Synergies and Trade-offs between Robusta Yield, Carbon Stocks and Biodiversity across Coffee Systems in the DR Congo

Synergies and Trade-offs between Robusta Yield, Carbon Stocks and Biodiversity across Coffee Systems in the DR Congo | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Synergies and Trade-offs between Robusta Yield, Carbon Stocks and Biodiversity across Coffee Systems in the DR Congo Ieben Broeckhoven, Jonas Depecker, Trésor Kasereka Muliwambene, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5165806/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 14 Feb, 2025 Read the published version in Agroforestry Systems → Version 1 posted 9 You are reading this latest preprint version Abstract The rapid decline of tropical rainforests, particularly in the Congo Basin, is predominantly driven by small-scale subsistence agricultural expansion. Tropical agroforestry, particularly coffee agroforestry, is seen as a potential way to balance agricultural productivity with biodiversity conservation and carbon sequestration, despite some possible trade-offs. However, substantial knowledge gaps persist regarding these trade-offs within and across coffee systems, especially in Africa. Here, we used a stratified random sampling design and general additive models to examine the relationship between yield, biodiversity, and carbon stocks in four coffee systems in the DR Congo (monocultures, cultivated agroforestry, wild agroforestry, and forest coffee) based on 79 inventoried plots. Our results demonstrate that coffee yields in cultivated agroforestry systems are not significantly different from monocultures, in contrast to lower yields in wild coffee agroforestry due to excessive shading (> 50%). Our study also shows the irreplaceable value of forest coffee systems in terms of biodiversity and carbon sequestration, suggesting that monoculture and agroforestry systems cannot serve as direct substitutes. Forest coffee systems contain three times more total organic carbon (TOC) than the agroforestry systems, which in turn contain almost double the amount of TOC as the coffee monocultures. Our findings revealed a steep decline in woody species diversity, including large changes in community composition, and carbon stocks from forest coffee to agroforestry, with comparatively smaller reductions from agroforestry to monocultures. On the one hand, our study identified convex relationships between woody species diversity and robusta coffee yield, as well as between carbon stocks and robusta yield. On the other hand, synergies are found between carbon stocks and woody plant diversity. One can thus say that coffee agroforestry systems allow the preservation of part of the biodiversity and carbon stocks while also supporting farmer’s livelihood. However, applying EUDR guidelines may hinder the adoption of these agroforestry systems due to the regulation’s inherent binary classification of forest versus non-forest. Agricultural intensification Agroforestry Coffea canephora Congo Basin Ecosystem Services EUDR Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction Africa stands out as being the home to the second-largest tropical rainforest globally, with approximately 89% situated within the Congo Basin (Harris et al., 2021 ; Malhi et al., 2013 ). Moreover, 60 % of these tropical rainforests of the Cngo Basin are found within the Democratic Republic of Congo (DRC) (Malhi et al., 2013 ). Furthermore, a significant portion (11 %) of the global tree cover loss between 2000 and 2021 occured in the Congo Basin (Global Forest Watch, 2023 ), with the highest deforestation rates observed in the DRC (FAO, 2020 ). The Tshopo province experienced the largest tree cover loss of all provinces within the DRC during this period (Global Forest Watch, 2023 ). Tropical rainforest loss is a major cause of climate change and biodiversity decline, with agriculture being the primary driver (Curtis et al., 2018; Feng et al., 2022; Jayathilake et al., 2021 ; Pendrill et al., 2022). Contrary to tropical forests in Asia and South America, the expansion of small-scale subsistence agriculture, instead of large-scale commodity-driven agriculture, is the major driver of deforestation in Africa, accounting for 97 % of all agriculture-driven forest loss (Branthomme et al., 2023 ; Curtis et al., 2018; Jayathilake et al., 2021 ; Pendrill et al., 2022). Notably, within the DRC, subsistence agriculture, mainly in the form of slash-and-burn agriculture, accounts for a staggering 93 % of deforestation, representing the secondhighest rate among the Congo Basin countries (Tegegne et al., 2016 ; Tyukavina et al., 2018 ). The proportion of the DRC’s deforestation caused by agriculture (93 %) significantly surpasses the regional average (68 %) for the Congo Basin countries (Tyukavina et al., 2018 ). In this context of deforestation driven by small-scale subsistence agriculture, tropical agroforestry is often advocated as it may reconcile the need for agricultural productivity, biodiversity conservation, carbon storage, and human well-being by leveraging synergies (Castle et al., 2021 ; Miller et al., 2020 ; Nair et al., 2022; Van Noordwijk et al., 2018 ; Wurz et al., 2022 ). A systematic review of productivity and ecosystem service provisioning of agroforestry in low- and middle-income countries worldwide found an overall positive impact of tropical agroforestry interventions on yield, although there was considerable heterogeneity depending on the type of intervention (Castle et al., 2021 ). Furthermore, the study revealed a small but overall positive impact on income. Surprisingly, there were only a limited number of studies dealing with the environmental benefits of agroforestry interventions. Nonetheless, the few existing studies found environmental benefits from agroforestry interventions (Castle et al., 2021 ). While trade-offs exist in agroforestry systems, it is vital to identify win-win opportunities. For example, Wurz et al. ( 2022 ) reported high yields in combination with high multi-taxa biodiversity in vanilla agroforests in Madagascar. Thus, the question whether similar synergies exist for coffee agroforestry systems arises. On the one hand, numerous studies have reported positive effects of coffee agroforestry systems on biodiversity conservation and carbon stocks compared to coffee monocultures. Several examples can be found mainly in Latin-America (Alvarez-Alvarez et al., 2021; Gordon et al., 2007; López-Gómez et al., 2008; Moguel & Toledo, 1999 ; Perfecto et al., 2003 , 2007; Philpott & Bichier, 2012), but also in India (Guillemot et al., 2018 ), Indonesia (Philpott et al., 2008 ), and Ethiopia (De Beenhouwer et al., 2016). On the other hand, competition for space, light, nutrients, and water, as well as changes in the incidence and severity of coffee pests and diseases, can positively and negatively influence coffee productivity within agroforestry systems (Durand-Bessart et al., 2020 ; Koutouleas et al., 2022 ; Mokondoko et al., 2022 ; Piato et al., 2020 ). Additionally, shade trees can be of significant economic value, generating additional income and income diversification, which is known to be particularly important during periods of low coffee prices (H. Davis et al., 2019 ; Leakey et al., 2005 ; Philpott et al., 2008 ; Rice, 2008 , 2011 ). Moreover, coffee agroforestry systems have been found to have equal or better economic importance than conventional systems (Jezeer et al., 2018). Studies have proven that coffee agroforestry systems can have higher profitability and cost-efficiency overall than conventional systems (Jezeer et al., 2017, 2019). Besides effects on coffee yield, variable impacts of shade on coffee quality have been reported (Muschler, 2001 ; Vaast et al., 2006 ). Currently, robusta coffee (Coffea canephora Pierre ex A. Froehner) makes up 43% of global coffee production, compared to 56% for arabica coffee (Coffea arabica L.) (ICO, 2023). The share of robusta production has been increasing over time, and this trend is expected to continue due to climate change (A. P. Davis et al., 2021 ; Piato et al., 2020 ). Historically, robusta coffee was one of the most valuable export commodities of the DRC, only second to copper (ICO et al., 2000). Moreover, in the 1990s, coffee contributed to three-quarters of the country’s agricultural export revenues and about 15 % of the GDP (ICO et al., 2000). However, the DRC’s robusta production and export levels arecurrently at a historical low (ICO, 2023). Thus, given that most of the coffee produced in the DRC is robusta (ICO, 2023), in combination with the increasing global demand for robusta, the cultivation of robusta coffee could represent a significant opportunity and source of livelihood for farmers in the DRC. Especially as agriculture in the DRC currently represents 55 % of employment and contributes to more than 30 % of GDP (World Bank, 2022 ). In addition to the potential in terms of agricultural production and livelihood generation, it should be noted that the DRC is an important region of origin for robusta coffee. The conservation of its standing coffee genetic diversity is of utmost importance for the future worldwide coffee production as wild populations carry valuable traits for coffee breeding, such as disease resistance (Lashermes et al., 2010 ; Silva et al., 2006 ), tolerance to climate change (A. P. Davis et al., 2012 ), and drought tolerance (Davis et al., 2021 ). Notable knowledge gaps are the limited research on robusta agroforestry systems compared to arabica, and the scarcity of African studies, particularly in the DRC (De Beenhouwer et al., 2013 ). Here, we make a first attempt to systematically quantify robusta coffee yields, carbon stocks and woody species diversity across different robusta coffee cultivation systems in the DRC. More specifically, we compared (i) coffee monocultures, (ii) cultivated coffee agroforestry systems characterised by commonly cultivated tree species intercropped with robusta, (ii) wild coffee agroforestry systems characterised by native undomesticated tree species in addition to commonly cultivated tree species intercropped with robusta, and (iv) forest coffee systems characterised by wild robusta, growing in a natural forest understorey. The recent EU Deforestation Regulation (EUDR) policy initiative, aims to reduce the EU’s impact on global deforestation and forest degradation, and thereby decreasing greenhouse gas emissions and biodiversity loss, has global ramifications (European Parliament and Council of the European Union, 2023), although its effects- and side effects – remain poorly understood. The associated EUDR map provides a reference and now allows to assess the impact for particular areas (Bourgoin et al., 2023, 2024). Our study aims to quantify potential trade-offs and synergies between robusta productivity, carbon stocks and woody species diversity. The specific objectives were fourfold: First, investigate the extent of variation in robusta yields between the different coffee systems. Second, quantify differences in carbon stocks, encompassing both above- and below-ground components. Third, investigate the variability in woody species diversity among the coffee typologies. Fourth, determine the existence of synergies and trade-offs between coffee yield, carbon stocks, and woody species diversity within these coffee systems, aiming to elucidate the precise nature of these interrelationships. Last, asses the classification of coffee systems as forest or non-forest in 2020 under EUDR. 2. Methodology 2.1. Study Area The research was carried out in the Tshopo Province, which is situated in the North-East of the Democratic Republic of Congo (DRC). The study area (0.5–1.4°N, 24–25°E; 380 m – 538 m) encompassed the territories of Isangi, Banalia, and Kisangani (Fig. 1 ). The region’s natural vegetation consists of moist semi-deciduous rainforest and monodominant Gilbertiodendron dewevrei (De Wild.) J.Léonard evergreen rainforest (Gilson et al., 1956). According to the Köppen-Geiger classification, the area has a tropical rainforest climate (Af) (Peel et al., 2007), with an average yearly precipitation of 1762 ± 295 mm in the Yangambi MAB reserve. The climate is characterised by two rainy seasons (August-November and April-May) and two dry seasons (December-February/March and June-July/August). Throughout the year, average temperatures remain high, with a minimum of 24.2 ± 0.4°C in July and a maximum of 25.5 ± 0.6°C in March. The dominant WRB soil type in the study area is a Ferralsol, formed from aeolian sediments, consisting mostly of quartz sand, kaolinite clay and hydrated iron oxides (Gilson et al., 1956; Van Ranst et al., 2010). However, due to seasonal inundations, Gleysols can also be found next to rivers and their tributaries (Gilson et al., 1956; Van Ranst et al., 2010). 2.2. The Four Coffee Systems Fieldwork was conducted between October and December 2021, the main coffee harvesting season. The coffee system typology was based partially on the share of dominant tree in total basal area and tree diversity following Sari et al. (2020). Additionally, canopy closure, origin of coffee planting material and input use were used to distinguish coffee system typologies (Fig. 2 ; Table 1 ). Coffee monocultures (MC) are characterised by robusta cultivars being the only woody species present and having an almost absent tree canopy (canopy closure < 5 %). Cultivated cofee agroforestry systems (CAF) are characterised by robusta cultivars being the most abundant woody species present, intercropped with commonly cultivated tree species and having a medium tree canopy closure (38.2 ± 4.75 %). Wild coffee aroforestry systems (WAF) are also characterised by robusta cultivars being the most abundant woody species present, intercropped with commonly cultivated tree species, but additionally contain native undomesticated fruit and forest tree species and having a higher tree canopy closure (52.9 ± 3.67 %). Forest coffeesystems (FC) are characterised by forest tree species, not wild robusta, being the most abundant woody species, and having a high tree canopy closure (> 85 %). The forest cofee plots were located in the Yangambi Man and Biosphere Reserve and the Ngazi Forest Reserve. Additionally, it should be noted that coffee farmers in our study area do not use any pesticides, insecticides or mineral fertilisers. Organic matter is sometimes applied in low quantities in the MC, CAF and WAF systems. Table 1 Coffee systems in Tshopo, DRC, and their defining characteristics Monoculture (n = 27) Cultivated Agroforestry (n = 15) Wild Agroforestry (n = 12) Forest Coffee (n = 25) ICRAF typology Monoculture Complex mixed agroforestry Complex mixed agroforestry Forest Tree species (/plot) 0 8 ± 1 8 ± 1 30 ± 1 Coffee as share of total BA 100% 31 ± 3% 33 ± 4% < 1% Dominant species Robusta coffee Commonly cultivated tree species Commonly cultivated & forest tree species Forest tree species Canopy closure 85% Coffee origin Local cultivars Local cultivars Local cultivars Wild Input use No mineral fertilisers, nor pesticides/insecticides No mineral fertilisers, nor pesticides/insecticides No mineral fertilisers, nor pesticides/insecticides No mineral fertilisers, nor pesticides/insecticides Both Cultivated and Wild Agroforestry fall under complex mixed agroforestry using the method of Sari et al. (2020), as coffee represents less than 50 % of the total baal area, and the number of tree species per plot is greater than 5. The four coffee typologies were then used as the basis for the stratified random sampling design. Specifically, 27 MC, 15 CAF, 12 WAF and 25 FC 25 m x 25 m plots were established (N = 79) with a minimum of 1 km distance between plots of the same typology. For forest coffee, five earlier established subplots (25 m x 25 m) within a larger plot of 125 m x 125 m were reused (Depecker et al., 2022 ). The data collected from the five subplots per main plot were then averaged to obtain one data point for each measured variable. 2.3. Measuring Robusta Coffee Yields Robusta yields were measured during the main harvesting season following Idol and Youkhana ( 2020 ) and calculated based on four randomly selected coffee plants per plot. To convert the number of ripe cherries to kilograms of green beans, a conversion factor of 3239 − 1 was used. This conversion factor was based on the average measurements using different robusta coffee varieties at the National Institute of Agronomic Research (INERA). A standardised moisture content for the green beans of 11 % was used. Tree canopyclosure (% shade) was measured using a model A convex spherical densitometer based on four data points per plot (Englund et al., 2011; Lemmon, 1956 ). 2.4. Assessing the Carbon Stocks of the Coffee Systems Carbon stocks were quantified based on the above- and below-ground biomass components per coffee system. For the above-ground biomass, three carbon pools were inventoried (Mg C/ha): above-ground living, deadwood, and litter biomass (IPCC, 2006 ). Biomass-related variables measured in the field for woody species with DBH ≥ 5cm included tree species, tree height (H), and diameter at breast height (DBH). For forest trees, an exponential regression model of the H:D relationship for the mixed forest in the Tshopo province (Kearsley et al., 2013 ), instead of a general H:D relation for Africa, was used (Eq. 1): \(\:H=(36.3576-31.6591)*{e}^{-0.0221*\:D}\) [Eq. 1] Dry wood densities (Mg/m 3 ) were obtained from the literature (Carsan et al., 2012 ; Depecker et al., 2022 b; Harja et al., 2019 ; NCSU, n.d.; Royal Museum for Central Africa, 2021). The dry weight of above-ground living biomass and deadwood biomass was calculated using allometric equations (Table 2 ). In the case of unidentified deadwood or if no species- or genus-specific value was available, the regional average dry wood density of 0.580 Mg/m 3 was used (Brown, 1997 ; IPCC, 2006 ). Litter was collected from two nested sub-plots (0.125m 2 each) within the sample plots, weighed, sun-dried, and then oven-dried for 72h at 50°C to calculate dry litter biomass. Different conversion factors were used to convert the dry biomass weight into carbon stocks (Mg C/ha): (i) for deadwood, the general conversion factor of 0.50 (IPCC, 2006 ), (ii) 0.37 for litter (IPCC, 2006 ), (iii) 0.379 for Musa spp. (Abdullah et al., 2013 ), and (iv) 0.45 for Coffea canephora were used (Van Noordwijk et al., 2002 ). Table 2 Allometric equations used for Above Ground Biomass (AGB) and fallen deadwood biomass estimations Variable Allometric equation Source AGB – General AGB = 0.0509 * ρ * DBH ² * H (Chave et al., 2005 ) AGB – Coffea Log( AGB ) = -1.181 + 1.991*Log( D 15 ) (Segura et al., 2006 ) AGB – Theobroma cacao Log( AGB ) = − 1.684 + 2.158 ∗ Log( D 30 ) + 0.892 ∗ Log( H ) (Somarriba et al., 2013 ) AGB – Elaeis guineensis AGB = -1.0007 * exp[-2.335 + 0.832 * ln( DBH 2 * H t ) (Migolet et al., 2020 ) AGB – Musa spp . AGB = -0.0927 + 0.0203 * DBH 2 (Alcudia-Aguilar et al., 2019 ) Fallen deadwood AGB = π * r 2 * ρ (Volume cylinder) AGB = estimated Above Ground Biomass [kg], ρ = wood density [Mg/m 3 ], DBH = Diameter at Breast Height [cm], D15 = Diameter at a height of 15cm [cm], D30 = Diameter at a height of 30cm [cm], r = radius, measured halfway the length of the deadwood [cm], H = Tree height [m], H t = Total tree height (height incl. trunk and crown) [m]. Two below-ground carbon pools (Mg C/ha) were considered: the root biomass and Soil Organic Carbon (SOC). The root biomass was derived from the above-ground biomass by multiplying it with the root:shoot ratio. For tropical moist forests, a root:shoot ratio of 0.235 ± 0.011 was used when living biomass was greater than 125 Mg/ha, while a root:shoot ratio of 0.205 ± 0.036 was used when living biomass was smaller than 125 Mg/ha (IPCC, 2006 ; Mokany et al., 2006). A root:shoot ratio of 0.096 + (0.022 * D15²) was used for coffee (Andrade et al., 2021 ). Composite soil samples based on eight different sampling points per plot were taken at three depth layers (0-5cm, 5-15cm and 15-30cm). These composite soil samples were mixed, sundried, and then oven-dried (50°C for 72 h), after which they were ground with a ball mill and sieved with a mesh size of 2 mm. Carbon content was determined using a Carlo Erba Elemental analyser (Ravindranath & Ostwald, 2008 ). Bulk density was measured from undisturbed soil samples using Kopecky rings at the same depths as the organic carbon samples. 2.5. Identifying Woody Plant Diversity In each plot, the abundance and DBH of the woody tree and shrub species with a DBH ≥ 5 cm were measured and recorded. This data was obtained from Depecker et al. ( 2022 ) for the forest coffee plots. For each plot, the observed (S obs ) and expected number of species (S exp ) and Hill numbers (N 1 , N 2 , E 1 and E 2 ) were calculated following Gotelli and Chao ( 2013 ). Moreover, the expected species richness (S exp ) was calculated using the Chao1-bc formula (Eq. 2) as several plots had no doubletons (Głowacki, 2011 ), with S exp being the expected number of species, S obs being the observed number of species, f 1 the number of singletons and f 2 the number of doubletons: \(\:{S}_{exp}={S}_{obs}\:+\:\frac{{f}_{1}\cdot\:\left({f}_{1}-1\right)}{2\cdot\:\left({f}_{2}+1\right)}\:\) [Eq. 2] The Hill numbers N 0 , N 1 , N 2 , E 1 (N 1 / N 0 ) and E 2 (N 2 / N 1 ) were calculated using the following formula (Hill, 1973 ), with N being the total number of organisms of all species and p i being the proportional abundance of species i out of S species: \(\:{N}_{q}={\sum\:}_{i=1}^{S}{\left({p}_{i}^{q}\right)}^{\frac{1}{1-q}\:}\) [Eq. 3] \(\:{E}_{q\:}=\:{N}_{q}/\:{N}_{q-1}\) [Eq. 4] 2.6. Forest Classification under EUDR The classification of coffee farms as forest or non-forest according to EUDR(Bourgoin et al., 2024; European Parliament and Council of the European Union, 2023) was evaluated using the Global Forest Cover 2020 map and ground-truthing data (Bourgoin et al., 2023, 2024). Analysis was based on the intersection of the coffee systems with forest pixels in the Global Forest Cover 2020 map (Bourgoin et al., 2023). Two different methods were applied using Google Earth Engine. One method used a GPS point in the coffee plot to represent the coffee system, while the other method used the minimum area of the inventoried plot as representing the coffee system. The inventoried plots had dimensions of 25m x 25m (625m 2 ). Most coffee farms were larger than the inventoried plot, therefore this can be seen as a conservative approach. 2.7. Statistical Analysis Statistical analyses were performed using R software (R Core Team, 2020 ). Prior to testing for significant differences between coffee systems, normality was tested. In the case of unpaired non-normally distributed data, the Kruskal-Wallis and pairwise Wilcoxon rank sum tests were used when comparing the coffee systems in terms of coffee yields, carbon stock elements, and biodiversity indices. In the case of unpaired normally distributed data, Tukey’s HSD was used when comparing the coffee systems. General Additive Models (GAMs) were used to model the relationships between coffee yield, carbon stocks and woody biodiversity, as they allow for the use of non-linear functions that can model more complex relationships while remaining interpretable. Modelling was done using the gam function from the mgcv package in R (R Core Team, 2020 ). Non-metric multidimensional scaling (NMDS) was used to visually represent the dissimilarity in tree community composition across coffee systems. A stress level of 0.07 was obtained for a two-axis solution, well below the threshold stress level of 0.20. The Bray-Curtis dissimilarity measure (Ricotta & Podani, 2017) and the metaMDS function in the vegan package were used for the ordination (Oksanen et al., 2018). Additionally, a permutational multivariate analysis of variance with 9999 permutations was performed using the adonis function in the vegan package to test for differences in community composition between the coffee systems. Pairwise comparisons with Benjamini-Hochberg correction among the coffee systems were made using the pairwise.adonis function in the vegan package based on 9999 permutations (Arbizu, 2020). 3. Results 3.1. Quantifying Robusta Coffee Yields and Management Variables The highest median robusta coffee yields per plant were found in CAF and MC, respectively, 0.96 ± 0.14 and 0.92 ± 0.07 kg green beans per plant (Table 3 ). The median coffee yield per plant tended to be higher in MC than in WAF (p = 0.098). Similarly, the yield per plant tended to be higher in CAF than in WAF (p = 0.111). No significant yield difference could be found between MC and CAF (p = 0.756). Similarly, the highest median robusta yields per ha were found in MC and CAF, respectively, about 800 ± 80 and 650 ± 135 kg green beans per ha (Table 3 ). While the median coffee yield per ha was found to be significantly higher in MC than in WAF (p = 0.018), no significant yield difference could be found between MC and CAF (p = 0.310), nor between CAF and WAF (p = 0.137). In addition, coffee yields per plant and per ha in FC were lower than all other coffee systems (p < 0.0001) and insignificant from a production perspective (< 1 kg/ha). No significant differences in coffee planting densities could be found between MC and CAF or WAF (p = 0.13). Regarding the canopy closure, the highest values were found for FC (91%), which was significantly greater than the 53% for WAF and 38% for CAF (p < 0.0001). Despite the large difference in median canopy closure between CAF and WAF, no significant differences could be found (p = 0.12). Table 3 Comparison of median robusta yields, coffee planting densities and canopy closure between different coffee systems in the Tshopo province, DR Congo. Monoculture (n = 27) Cultivated AF (n = 15) Wild AF (n = 12) Forest Coffee (n = 25) µ ± se µ ± se µ ± se µ ± se p-value Yield [kg green coffee/plant] 0.916 a (± 0.0746) 0.956 a,b (± 0.143) 0.586 b (± 0.101) 0.00205 c (± 0.00136) < 0.0001 Yield [kg green coffee/ha] 798 a (± 79.9) 648 a,b (± 134) 503 b (± 94.2) 0.131 c (± 0.0868) < 0.0001 Coffee density [plants/ha] 960 a (± 47.0) 896 a (± 61.6) 880 a (± 63.5) 64.0 b (± 3.16) < 0.0001 Canopy closure [%] 0.00 c (± 0.475) 38.2 b (± 4.75) 52.9 b (± 3.67) 91.4 a (± 0.785) < 0.0001 Different letters indicate significant differences between the different coffee systems based on the Kruskal-Wallis rank sum test and pairwise Wilcoxon rank sum tests with Benjamini-Hochberg correction for multiple testing for comparing non-normal distributions. 3.2. Quantifying Carbon Stocks Total organic carbon stocks were found to be largest in FC (289 ± 13.6 Mg C/ha), followed by WAF (99.6 ± 10.2 Mg C/ha) and CAF (91.9 ± 8.95 Mg C/ha), with MC (58.1 ± 2.63 Mg C/ha) having the lowest TOC (Fig. 3 , Table A1 ). FC contained roughly three times more TOC than CAF and WAF systems, which in turn contained almost double the amount of TOC than MC. Furthermore, significant differences were found when comparing the five carbon stock components across the coffee systems, except for litter biomass (Fig. 3 ; Table A1 ). Among these carbon stock elements, above-ground living and root biomass were identified as primary drivers for the differences in TOC between coffee systems and deadwood biomass to a lesser extent. In contrast, soil organic carbon (SOC) had a limited impact on the overall differences in TOC. Nonetheless, SOC accounted for 87% of the TOC found in MC while contributing 58% and 47% of TOC in CAF and WAF, respectively, and only 15% of TOC in FC systems. In addition, MC and CAF systems had significantly larger SOC stocks (0–30 cm) than FC, with p-values of 0.074 and 0.041, respectively (Fig. 3 ). Moreover, significant differences in SOC were only observed in the 0–5 cm layer (Table A2 ). There were no significant differences in soil organic carbon concentration. However, the bulk density was smaller in FC than in MC and CAF for all three depth layers (Table A2 ). 3.3. Quantifying Woody Biodiversity Our study recorded 11,001 individuals (DBH ≥ 5 cm) from 249 taxa across all plots. Of the 249 observed taxa, 85.27% were identified at the species level, 9.93% at the genus level, and 4.80% were unidentified taxa. In terms of observed woody species, 19 different woody species were found in CAF, 45 in WAF and 214 in FC. As the MC plots, by definition, lacked tree or shrub species apart from coffee, the MC plots were excluded from the calculation of the biodiversity indices. All biodiversity indices (S obs , S exp , N 1 , N 2 and E 1 ) were found to be significantly larger in FC than in CAF and WAF systems (Fig. 4 ), except for Hill’s E 2, which tended to be lower only in CAF compared to FC (p < 0.06). In comparison to FC, both agroforestry systems showed a significant reduction in observed species richness (73–74%) and expected species richness (78%) (Fig. 4 , Table A3 ). Furthermore, woody species diversity (N 1 and N 2 ) was reduced by 86–90% in CAF and WAF compared to FC. Additionally, the evenness of woody species (E 1 ) was reduced by 54% in CAF and WAF systems compared to FC, while Hill's evenness (E 2 ) was only 7% lower in CAF than in FC (Table A3 ). The woody plant communities in both agroforestry systems were found to differ greatly from FC communities along both NMDS axes (R 2 = 0.3242, P < 0.001) (Fig. 5 ). Additionally, the variability of community composition was smaller in the CAF than in the WAF system, with both having a smaller variability than the FC system. Of the 249 tree species observed, only three woody species were common among all three coffee systems, namely: Coffea canephora, Dacryodes edulis (G.Don) H.J.Lam. and Anonidium mannii (Oliv.) Engl. & Diels . Furthermore, FC and CAF only shared 3 woody species, the ones common to all three coffee systems. WAF was found to have 13 woody species in common with CAF and 14 in common with FC. Besides, 4 woody species were found to be unique to CAF, 21 unique to WAF and 200 unique to FC. 3.4. Identifying Synergies and Trade-offs between Robusta Yield, Woody Species Diversity and Carbon Stocks A significantly positive relationship was found between all biodiversity indices and TOC, except Hill's E2 index (Fig. 6 ), indicating co-benefits between woody species diversity and carbon stocks. The relationship between total organic carbon (TOC) and yield exhibited a convex pattern (Fig. 7 ). A sharp initial decline in TOC was observed when moving from FC (high TOC) to the CAF and WAF systems (medium TOC), which was accompanied by a large increase in coffee yields. Likewise, a trade-off could be identified when moving from WAF (medium TOC & medium yield) to MC (low TOC & high yield). However, when moving from CAF (medium TOC & high yield) to MC (low TOC & high yield), a loss of TOC was observed without an increase in coffee yield, in contrast to the other shifts where trade-offs between TOC and coffee yield occurred. Similar to the TOC-yield relation, the relationship between biodiversity and yield was examined using generalised additive models (GAMs), revealing a convex pattern (Fig. 7 ). When moving from FC to the coffee agroforestry systems, a rapid decline in woody species richness (S obs ) was observed, accompanied by a large increase in coffee yield. In addition, a trade-off could be identified when moving from WAF (medium biodiversity & medium yield) to MC (low biodiversity & high yield). However, when moving from CAF (medium biodiversity & high yield) to MC (low biodiversity & high yield), a loss of woody species diversity was observed without an increase in coffee yield, in contrast to the other shifts where trade-offs between woody species diversity and coffee yield occurred. 3.5. Classification of Coffee Systems as Forest under EUDR The comparison of coffee farms in our study with the Global Forest Cover 2020 map (Bourgoin et al., 2023, 2024) reveals that a significant proportion of coffee farms were classified as forest in 2020 (Table 4 ). The Forest Coffee plots were consistently and accurately categorised as forest. However, when analysing the plots inventoried on the coffee farms, 26% of monoculture and 58–60% of coffee agroforestry plots appeared in the so-called forest. Nevertheless, all coffee farms in the study had been planted for at least five years, with data collection occurring in 2021. This indicates that all farms were non-forest before 2020. Consequently, all plots classified as forest were incorrectly categorised except for the Forest Coffee plots. Even when plots were represented by a single point rather than the 25m x 25m plot, 20–42% of coffee farms were still erroneously classified as being forest in 2020. Additionally, as hypothesised, the analysis revealed that coffee agroforestry systems were more frequently misclassified as forest than monocultures, with agroforestry plots being twice as likely to be labelled as forest when using the plot method. Table 4 Forest and Non-Forest classification of coffee plots in Tshopo, DRC, based on EUDR criteria, comparing four different coffee systems and two classification methods. Method Classification Monoculture (n = 27) Cultivated Agroforestry (n = 15) Wild Agroforestry (n = 12) Forest Coffee (n = 25) Point Forest [%] 22.2 20.0 41.7 100 Non-Forest [%] 77.8 80.0 58.3 0.00 Plot (625m 2 ) Forest [%] 25.9 60.0 58.3 100 Non-Forest [%] 74.1 40.0 41.7 0.00 4. Discussion 4.1. Comparing Robusta Coffee Yields, Considering Farm Management Robusta coffee yields in forest coffee systems in our study (< 1 kg/ha) are several orders of magnitude lower than those of wild Coffea arabica in Ethiopian forests (15 kg/ha; Schmitt et al., 2010 ), which is likely the result of lower coffee densities and greater canopy closure in the Congolese tropical rainforests. The coffee yields in our study were found to have median yields ranging roughly from 0.50 to 0.80 ton green coffee/ha depending on the coffee system, excluding forest coffee (Table 3 ), and are significantly lower than those found for monocultures (1.25 ton/ha) and robusta-banana intercropping systems (1.09 ton/ha) in Uganda (van Asten et al., 2011 ). Furthermore, coffee yields for both monocultures and agroforestry systems were much lower in our study than in most studies in the meta-analysis (4.1 ± 2.88 ton/ha) on the drivers of yield gaps across different types of coffee systems (Mokondoko et al., 2022 ). This large yield gap is likely due to a lack of good agricultural practices, such as using improved coffee varieties, inorganic inputs, timely pruning, pest and disease management and also to a general low soil fertility in our study area (Pers. Obs.). On the one hand, coffee yields in our study were significantly higher in monocultures than in wild coffee agroforestry systems on a per hectare basis (Table 3 ). No significant difference was found in coffee planting density when comparing coffee monocultures with wild coffee agroforestry (p = 0.13), likely due to the limited sample size of the wild agroforestry plots (Table 3 ). Thus, the yield difference on a per ha basis mainly comes from the significant yield difference on a per plant basis. Furthermore, the yield differences per plant between monocultures and wild coffee agroforestry likely result from excessive shading (> 50%) in most wild agroforestry plots (Table 3 ). On the other hand, when comparing coffee monocultures with cultivated coffee agroforestry systems, no significant difference was found in coffee planting densities (p = 0.13). This might be due to the limited sample size of the cultivated agroforestry plots. However, the difference in median coffee planting density is not large (Table 3 ). It can be concluded that, despite a significantly higher median canopy closure and tree planting density in cultivated agroforestry systems compared to monocultures, coffee yields in our study are similar for monocultures and cultivated agroforestry systems, whether on a per plant basis (p = 0.757) or per ha (p = 0.310) (Table 3 ). The above is supported by a global meta-analysis on the drivers of yield gaps across coffee systems (Mokondoko et al., 2022 ), which found that shade cover and plant densities affected coffee yields more than climatic and biotic factors. Moreover, the meta-analysis found that coffee agroforestry systems with a shade cover of 35–50 % and tree density of 00–250 trees/ha, which corresponds to the cultivated coffee agroforestry systems in our study, have similar coffee yields as monocultural systems with a shade cover of less than 30 %. However, agroforesty systems with shade cover of 50–80 % and a tree density geater than 250 trees/ha, similar to the wild coffee agroforestry systems in our study, were found to result in significantly reduced coffee yields (Mokondoko et al., 2022 ). Thus, given the results from the meta-analysis and the finding that the median shade cover for wild coffee agroforestry systems in our study was 52.9 ± 3.67 %, this might indicate that the shade cover isabove the optimal range for coffee yield maximisation. Another meta-analysis on the effect of shade on robusta coffee yield and growth (Piato et al., 2020 ) was inconclusive, mainly due to the limited number of studies. Despite noteworthy higher median yield and smaller median canopy closure values for the cultivated coffee agroforestry system compared to wild coffee agroforestry, no significant differences were found in terms of yield per plant (p = 0.111), yield per ha (p = 0.137) and shade cover (p = 0.120) (Table 3 ). Most likely, this is due to the limited number of sampled plots for each of the two agroforestry systems, which resulted in limited statistical power. Additionally, it should be noted that coffee planting densities are likely below optimal in the monoculture system and possibly also in both agroforestry systems, as planting densities of roughly 960 and 890 coffee plants per hectare were found in our study for monocultures and agroforestry, respectively (Table 3 ). By comparison, densities of 1333 plants per ha are recommended by the local research institute (INERA) and a range of 1111–2222 plants per ha is recommended for robusta coffee in the literature (Niyibigira, 2019 ; Wintgens, 2004 ). Nonetheless, our findings align with van Asten et al. ( 2011 ) in smallholder farms in Uganda, who observed densities of 1014 and 880 robusta coffee plants per ha in monocultures and agroforestry systems. 4.2. Contrasting Carbon Stocks between Robusta Coffee Systems Forest coffee contained roughly three times more TOC than both agroforestry systems, which in turn was shown to contain roughly 60–70% more TOC than monocultures (Table A1 ). This trend is consistent with literature that compares monocultures, coffee agroforestry systems, and forest (coffee) systems (De Beenhouwer et al., 2016; Ehrenbergerová et al., 2016; Guillemot et al., 2018 ; Schmitt-Harsh et al., 2012 ; Solis et al., 2020 ; Soto-Pinto et al., 2010 ; Van Noordwijk et al., 2002 ; Vanderhaegen et al., 2015 ). Moreover, our study found that the TOC content in forest coffee systems was 289 ± 14 Mg C/ha, which falls roughly in the middle of the TOC range (198–413 Mg C/ha) reported in the literature (De Beenhouwer et al., 2016; Guillemot et al., 2018 ; Schmitt-Harsh et al., 2012 ; Van Noordwijk et al., 2002 ; Vanderhaegen et al., 2015 ). The average living above-ground carbon stock obtained for forest coffee systems in our study (179 ± 11 Mg C/ha) is consistent with results from other studies on forests in the Tshopo province, such as 162 ± 20 Mg C/ha in (Kearsley, 2015 ) and 185 ± 44 Mg C/ha in (P. Moonen, 2017 ). 8 In contrast, the TOC for both coffee agroforestry systems in our study (92–100 ± 10 Mg C/ha) was lower than the range of 120–219 Mg C/ha reported in previous studies (De Beenhouwer et al., 2016; Ehrenbergerová et al., 2016; Guillemot et al., 2018 ; Schmitt-Harsh et al., 2012 ; Solis et al., 2020 ; Soto-Pinto et al., 2010 ; Van Noordwijk et al., 2002 ; Vanderhaegen et al., 2015 ). This difference compared to other studies can be attributed to smaller living above-ground carbon stocks, and smaller SOC stocks in our study (Table A1 ) resulting from smaller soil organic carbon concentrations (Table A2 ). Moreover, the smaller living above-ground carbon stocks found for coffee agroforestry systems in our study might be attributed to tree densities being at the lower end, respectively 192 ± 25 and 240 ± 34 trees/ha for Cultivated and Wild agroforestry systems, compared to other studies on arabica and robusta coffee from India, Latin-America, and Ethiopia, which have reported a range of 124–800 trees/ha (De Beenhouwer et al., 2016b; Ehrenbergerová et al., 2016; Guillemot et al., 2018 ; Schmitt-Harsh et al., 2012 ; Solis et al., 2020 ; Soto-Pinto et al., 2010 ; Van Noordwijk et al., 2002 ; Vanderhaegen et al., 2015 ). For robusta coffee monocultures, the average TOC of 58 ± 3 Mg C/ha found in our study lies within the range of 52–113 Mg C/ha found for coffee monocultures in Peru and Indonesia (Ehrenbergerová et al., 2016; Solis et al., 2020 ; Van Noordwijk et al., 2002 ). Though within range, our TOC findings are at the lower end for monocultures due to relatively small SOC stocks (Table A1 ), mainly due to small soil organic carbon concentrations rather than smaller soil bulk densities (Table A2 ). When comparing the AGB of coffee systems (Table A1 ) with other agro- and ecosystems in the study area, it can be observed that the AGB of both coffee agroforestry systems (31–39 ± 6 Mg C/ha) is similar to the AGB in local cacao agroforests (44 ± 25 Mg C/ha) (Batsi et al., 2021 ). In addition, a study on the fallow systems (secondary forest regrowth after slash-and-burn) in the Tshopo province found that these fallow systems have an average AGB of 58 ± 46 Mg C/ha. However, this varies heavily depending on the number of passed rotation cycles and fallow age (Moonen et al., 2019 ). Thus, the AGB in secondary forest regrowth after slash-and-burn can be much smaller, similar or greater than in coffee agroforestry systems, while being almost always much greater than in coffee monocultures (6 ± 1 Mg C/ha). Furthermore, it should be noted that the AGB and total organic carbon in fallow systems are decreasing and will continue to decrease over time as the number of rotation cycles is increasing and the fallow period is shortening (Moonen et al., 2019 ). This indicates that, in the long term, the local coffee and cacao agroforestry systems hold greater potential in terms of carbon sequestration than the fallow systems. Moreover, (Moonen et al., 2019 ) noted that from the third and fourth fallow cycles onwards, a shift from fallows (secondary forest regrowth) to more permanent cropping systems, such as coffee and cacao agroforestry systems, frequently results in positive impacts on carbon stocks. 4.3. (Dis)similarities in Woody Plant Diversity and Community Composition Forest coffee systems exhibited higher woody biodiversity than both agroforestry systems, with both agroforestry systems exhibiting higher woody biodiversity than monocultures (Fig. 3 ). The large differences in species richness, diversity, evenness, and community composition indicate that forest coffee systems are characterised by a unique biodiversity that cannot be substituted by coffee agroforestry systems, and especially not by monocultures. Moreover, most woody species were only found in forest coffee, while some were only found in wild coffee agroforestry, though to a much lesser extent. Thus, coffee agroforestry systems contain significant woody species diversity, which complements the diversity found in forest coffee systems (Fig. 4 ). In addition, it should be noted that despite large differences in woody species community composition, several woody species are shared between wild coffee agroforestry and forest coffee systems, in contrast to cultivated coffee agroforestry and forest coffee. These findings align with the meta-analysis (De Beenhouwer et al., 2013 ), which concludes that converting natural forests to coffee agroforestry systems has an overall detrimental impact on total species richness, even more so on forest species. Likewise, as management intensifies from wild coffee agroforestry to cultivated coffee agroforestry, native forest species and native undomesticated fruit tree species tend to disappear, with only minimal to no replacement by other commonly cultivated tree species. As forest specialists often have specialised habitat requirements, they are more vulnerable to land-use change (Gibson et al., 2011 ; Hundera et al., 2013). Thus, woody species restricted to forests are greatly affected by management intensification, emphasising the indispensable role of natural forests in conserving forest species, even in an agroforest matrix, as shown by our results and supported by the meta-analysis of De Beenhouwer et al. ( 2013 ). Broader studies also support these results in the sense of the irreplaceable value of tropical forests for sustaining tropical biodiversity (Gardner et al., 2009 ; Gibson et al., 2011 ; Muñoz et al., 2013 ). Although agroforestry systems fail to conserve the strictest forest species, they can contribute to the conservation of many other species (Muñoz et al., 2013 ). Arthropods, mammals and birds were largely unaffected by the conversion of forest into agroforest (Daily et al., 2003 ; De Beenhouwer et al., 2013 ; Gibson et al., 2011 ; Pardini, 2004 ). These findings suggest that agroforestry systems can offer a partial solution to balance the need for agricultural production and biodiversity conservation. De Beenhouwer's meta-analysis (2013) found that intensifying high-diversity agroforestry systems towards sun plantations, containing only a sparse amount of shade trees that belong to one or very few species, led to a greater decrease in total species richness than converting natural forests into high-diversity agroforestry systems. This result can be attributed to the diverse and structurally complex canopy layer in certain diverse coffee agroforestry systems, which maintain a high level of biodiversity (Perfecto & Vandermeer, 2008 ; Schroth & Harvey, 2007 ). Our study observed the most significant decline in woody species richness (S obs ) occurring from forest coffee (29.57 ± 0.52) to cultivated coffee agroforestry (7.92 ± 0.54) and wild coffee agroforestry (8.17 ± 0.67), as opposed to, from coffee agroforestry to coffee monocultures. 4.4. Clarifying Synergies and Trade-offs between Robusta Yields, Woody Biodiversity and Carbon Stocks This study identified trade-offs between woody plant diversity and coffee yields, as well as between carbon stocks and coffee yield, alongside co-benefits between woody plant diversity and carbon stocks. Firstly, our findings regarding the relationship between woody biodiversity and coffee yield show a strong negative relationship (Fig. 6 ), which aligns with previous research (De Beenhouwer et al., 2013 ; Jha et al., 2014 ). However, contrary to other studies (De Beenhouwer et al., 2013 ; Perfecto et al., 2003 ), we observed a convex biodiversity-yield relationship, indicating a sharper decline in biodiversity during the shift from forest to coffee agroforestry and a less pronounced decline from agroforestry to monocultures. While in the case of a concave biodiversity-yield relationship, wildlife-friendly farming may be the best conservation option (Clough et al., 2011 ; De Beenhouwer et al., 2013 ; Perfecto et al., 2005 ), in our study context, the convex nature of the relationship indicates that wildlife-friendly farming alone does not suffice and stresses the importance of forest coffee conservation. Furthermore, when further intensifying management from cultivated agroforestry systems to monocultures, a loss of woody biodiversity was observed, without a gain in coffee yield, rather than a trade-off. Thus, as the much more biodiverse farming system of the two, cultivated coffee agroforestry may serve as the preferred management strategy for balancing biodiversity and coffee yield compared to coffee monocultures. Secondly, the convex TOC-yield relationship found in our study (Fig. 6 ), with a sharp decline in TOC from forest coffee to both agroforestry systems and a relatively smaller decline toward monocultures, aligns with the provisioning yield function suggested by De Beenhouwer et al. ( 2013 ). This again highlights the enormous value of forest coffee in terms of carbon sequestration compared to other coffee systems. However, given farmers’ need to balance production with carbon sequestration, other coffee systems will exist and will replace a part of the forest coffee system. Thus, it is important to note that, when intensifying management from cultivated agroforestry systems to monocultures, a loss of woody biodiversity was observed, without a gain in coffee yield, rather than a trade-off, which occurs when shifting between the other systems. Thus, as the farming system which contains much more carbon, cultivated coffee agroforestry is the preferred management strategy for balancing carbon sequestration and coffee yield compared to coffee monocultures. Thirdly, in contrast to the negative relationships found in our study between yield and biodiversity as well as between yield and TOC mentioned above, a significant positive relationship was found between the woody biodiversity indices and carbon stocks (Fig. 5 ). The latter finding supports previous studies (De Beenhouwer et al., 2016; Jha et al., 2014 ; Richards & Méndez, 2014 ; Sari et al., 2020; Zewdie et al., 2022 ). These results indicate co-benefits between woody biodiversity and carbon stocks (Fig. 5 ). Consequently, both objectives can be pursued simultaneously, highlighting the potential for synergies between biodiversity conservation and climate change mitigation. 4.5. Implications of EUDR Forest Classification for Coffee Systems The above-mentioned misclassification of coffee gardens as forest likely arises from the resemblance of agroforestry systems to forests in satellite imagery due to agroforestry's high tree canopy cover. Moreover, the cut-off value used by the EUDR is greater than 10% canopy cover (Bourgoin et al., 2024; European Parliament and Council of the European Union, 2023), while canopy closure found in our study for coffee agroforestry systems was 38 ± 4 % and 53 ± 4 % depending on the type of agroforestry (Table 3 ). Consequently, the EDR classifiction approach seems to disproportionately affect coffee systems with greater canopy cover. This will likely penalise sustainable practices like agroforestry, which have larger carbon stocks and woody diversity than monocultures (Figs. 3 and 4 ). This clearly contrasts with the environmental goals of the EUDR, which is “to bring down greenhouse gas emissions and biodiversity loss” (European Parliament and Council of the European Union, 2023). Additionally, since monocultures were often misclassified, while their canopy cover ranged from 0–5 %, it indicates that too much faith is given to these remotely sensed products. Another important conceptual deficiency of the EUDR is that it confuses land cover (as observed with remote sensing) with land use, which focuses more on land management and tenure. It has the perverse effect of leaving farmers who previously deforested their land unaffected, but it risks severely punishing farmers who apply more sustainable agroforestry practices because their farm resembles a forest too much. Thus, The EUDR is a very blunt instrument that risks disproportionally affecting smallholder farmers, many of whom have introduced sustainable agricultural practices. 5. Conclusion This study provides valuable insights into the synergies and trade-offs between robusta coffee yield, carbon stocks, and woody species biodiversity across coffee systems in the Tshopo province, DR Congo. The observed co-benefits between carbon stocks and biodiversity highlight the potential for synergies in coffee systems between biodiversity conservation and climate change mitigation. Trade-offs between robusta coffee yields on the one hand and woody biodiversity and carbon stocks on the other hand were found. On the one hand, the sharp decline in woody biodiversity and carbon stocks when intensifying forest coffee systems towards coffee agroforestry or coffee monocultures implies that forest coffee systems have an irreplaceable value. While significant trade-offs between woody biodiversity and carbon stocks, and coffee yield occur when intensifying wild coffee agroforestry to monocultures, this was not the case for cultivated agroforestry. Thus, from a perspective of balancing needs for production, biodiversity and carbon sequestration, cultivated coffee agroforestry systems can be recommended above coffee monocultures and wild coffee agroforestry. Even more so, coffee agroforestry provides additional advantages in terms of food diversification, income diversification and risk management. The profitability of the cultivated agroforestry system is likely greater than that of the low-input coffee monocultures in our study. However, additional research comparing the economic aspects of the different coffee systems would be welcome. Besides, further research on farmers’ knowledge of and preferences for (shade) tree species in agroforestry is recommended and will provide a better understanding of farmers’ choice between different coffee agroforestry systems. The comparison of coffee farms with the Global Forest Cover 2020 map used for EUDR revealed substantial misclassification, particularly for agroforestry systems, due to their high canopy cover resembling natural forests. This misclassification will likely penalise sustainable practices like agroforestry and reach the opposite of the EUDR’s environmental goals. Not only is a greater tree canopy cover threshold needed to differentiate between coffee systems and forest, but more attention should also be given to ownership and land tenure ship rights. In addition, our results indicate that excessive shading (> 50%) in many wild agroforestry plots severely affects coffee yields. Thus, identifying the optimal shade level for robusta cultivation would assist in minimising trade-offs and maximising synergies in coffee agroforestry systems. In line with this, optimal robusta and (shade) tree density combinations are poorly understood. However, given the too-low coffee planting densities currently used by most farmers in the Tshopo province, we recommend increasing their coffee planting density, especially in monocultures, but possibly also in coffee agroforestry. Furthermore, the coffee production systems in the Tshopo province should, in the first instance, be optimised by increasing coffee planting densities, pruning the principal stems to obtain an optimal number of stems per plant, seasonal pruning of the branches, and pruning the principal stems to ensure rejuvenation. Once these foundations have been laid, one could consider using fertilisers to address the low soil fertility in the region. Also, management of coffee pests and diseases, which is currently almost completely lacking, could greatly improve coffee yields, with a focus on the Coffee Berry Borer ( Hypothenemus hampei ) and wood-boring beetles, namely the Coffee Twig Borer ( Xylosandrus compactus Eichhoff 1875) and the West African Coffee Stem Borer ( Bixadus sierricola White 1858). Declarations Author Contribution Conceptualisation: IB, RM, OH and BV; Methodology: IB, BV, OH, RM and JD; Data collection: IB, TKM and JD; Lab work: IB; Data analysis: IB; Software; IB and JD; Writing – Original draft: IB; Visualisation: IB and JD; Writing - Review and editing: IB, TKM, JD, RM, OH and BV; Supervision: RM, OH and BV; Funding acquisition: RM and BV. Acknowledgement We would like to express our gratitude to Justin Angunizu Asimonyio (CSB) and Jean-Léon Kambale (CSB) for their assistance with the botanical identification of the woody species. Besides, we would like to thank Lore Fondu (KU Leuven) for her assistance with the analysis of soil samples. Data Availability Data used for figures and tables in this study have been deposited online via Zenodo: 10.5281/zenodo.13850422 References Abdullah, N., Sulaiman, F., & Taib, R. M. (2013). Characterisation of banana (Musa spp.) plantation wastes as a potential renewable energy source. In AIP Conference Proceedings (Vol. 1528, No. 1, pp. 325–330). American Institute of Physics. Aerts, R., Hundera, K., Berecha, G., Gijbels, P., Baeten, M., Van Mechelen, M., Hermy, M., Muys, B., & Honnay, O. (2011). Semi-Forest Coffee cultivation and the conservation of Ethiopian Afromontane rainforest fragments. Forest Ecology and Management, 261, 1034–1041. Alcudia-Aguilar, A., Martínez-Zurimendi, P., van der Wal, H., Castillo-Uzcanga, M. M., & Suárez-Sánchez, J. (2019). Allometric estimation of the biomass of Musa spp. in homegardens of Tabasco, Mexico. Tropical and Subtropical Agroecosystems, 22(1). Alvarez Alvarez, E. A., Almazán Núñez, R. C., González García, F., Brito Millán, M., Méndez Bahena, A., & García Ibáñez, S. (2020). Shade coffee plantations maintain woody plant diversity and structure in a cloud forest landscape of southern Mexico. J. For. Res.. Andrade, H. J., Segura, M. A., & Feria, M. (2021). Allometric models for estimating belowground biomass of individual coffee bushes growing in monoculture and agroforestry systems. Agroforestry Systems, 95, 215–226. Arbizu, P. M. (2017). pairwiseAdonis: Pairwise Multilevel comparison using Adonis . R package version Batsi, G., Sonwa, D. J., Mangaza, L., Ebuy, J., & Kahindo, J. M. (2021). Preliminary estimation of above-ground carbon storage in cocoa agroforests of Bengamisa-Yangambi forest landscape (Democratic Republic of Congo). AF Systems, 95 (8), 1505–1517. Beer, J. (1987). Advantages, disadvantages and desirable characteristics of shade trees for coffee, cacao and tea. AF Systems, 5 (3), 3–13. Branthomme, A., Merle, C., Kindgard, A., Lourenço, A., Ng, W.-T., D’Annunzio, R., & Shapiro, A. (2023). How much do large-scale and small-scale farming contribute to global deforestation? Results from a remote sensing pilot approach. Rome: FAO. https://doi.org/10.4060/cc5723en Brown, S. (1997). Estimating biomass and biomass change of tropical forests: A primer (Vol. 134). Food & Agriculture Org . Cardinale, B. J., Duffy, J. E., Gonzalez, A., Hooper, D. U., Perrings, C., Venail, P., … Naeem, S. (2012). Biodiversity loss and its impact on humanity. Nature, 486, 59–67. Carsan, S., Orwa, C., Harwood, C., Kindt, R., Stroebel, A., Neufeldt, H., & Jamnadass, R. (2012). African wood density database . World AF Centre, Nairobi. Castle, S. E., Miller, D. C., Ordonez, P. J., Baylis, K., & Hughes, K. (2021). The impacts of AF interventions on agricultural productivity, ecosystem services, and human well-being in low‐and middle‐income countries: A systematic review. Campbell Systematic Reviews, 17(2), e1167. Chao, A., Chazdon, R. L., Colwell, R. K., & Shen, T. J. (2005). A new statistical approach for assessing compositional similarity based on incidence and abundance data. Ecology Letters, 8, 148–159. Chave, J., Andalo, C., Brown, S., Cairns, M. A., Chambers, J. Q., Eamus, D., … Yamakura, T. (2005). Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia, 145, 87–99. Clough, Y., Barkmann, J., Juhrbandt, J., Kessler, M., Wanger, T. C., Anshary, A., … Tscharntke, T. (2011). Combining high biodiversity with high yields in tropical agroforests. Proceedings of the National Academy of Sciences , 108(20), 8311–8316. Cramer, P. J. S., & Wellman, F. L. (1957). Review of literature of coffee research in Indonesia. Daily, G. C., Ceballos, G., Pacheco, J., Suzán, G., & Sánchez-Azofeifa, A. R. (2003). Countryside biogeography of neotropical mammals: conservation opportunities in agricultural landscapes of Costa Rica. Conservation biology, 17(6), 1814–1826. Davis, A. P., D. Mieulet, J. Moat, D. Sarmu, and J. Haggar. (2021). Arabica-like Flavour in a Heat-Tolerant Wild Coffee Species. Nature Plants 7 (4): 413–18. https://doi.org/10.1038/s41477-021-00891-4 . Davis, A. P., Gole, T. W., Baena, S., & Moat, J. (2012). The impact of climate change on indigenous arabica coffee (Coffea arabica): predicting future trends and identifying priorities. PloS one, 7 (11), e47981. Davis, H., Rice, R., Rockwood, L., Wood, T., & Marra, P. (2019). The economic potential of fruit trees as shade in blue mountain coffee agroecosystems of the Yallahs River watershed, Jamaica WI. AF Systems, 93 , 581–589. De Beenhouwer, M., Aerts, R., & Honnay, O. (2013). A global meta-analysis of the biodiversity and ecosystem service benefits of coffee and cacao AF. Agriculture, Ecosystems & Environment, 175, 1–7. De Beenhouwer, M., Muleta, D., Peeters, B., Van Geel, M., Lievens, B., & Honnay, O. (2015a). DNA pyrosequencing evidence for large diversity differences between natural and managed coffee mycorrhizal fungal communities. Agronomy for Sustainable Development, 35, 241–249. De Beenhouwer, M., Van Geel, M., Ceulemans, T., Muleta, D., Lievens, B., & Honnay, O. (2015b). Agricultural intensification alters the arbuscular mycorrhizal fungi communities of Arabica coffee (Coffea arabica) in its Ethiopian centre of origin through changing soil characteristics. Soil Biology and Biochemistry, 91, 133–139. De Cáceres, M., and Legendre, P. (2009). Associations between species and groups of sites: indices and statistical inference. Ecology, 90, 3566–3574. Depecker, J., Asimonyio, J. A., Miteho, R., Hatangi, Y., Kambale, J. L., Verleysen, L., Stoffelen, P., Janssens, S. B., Dhed’a, B., Vandelook, F., & Honnay, O. (2022). The association between rainforest disturbance and recovery, tree community composition, and community traits in the Yangambi area in the Democratic Republic of the Congo. Journal of Tropical Ecology, pp.1–11. Dufrene, M., and Legendre, P. (1997). Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecological Monographs, 67, 345–366. Durand-Bessart, C., Tixier, P., Quinteros, A., Andreotti, F., Rapidel, B., Tauvel, C., & Allinne, C. (2020). Analysis of interactions amongst shade trees, coffee foliar diseases and coffee yield in multistrata AF systems. Crop Protection, 133 , 105137. Ehrenbergerova, L., Cienciala, E., Kučera, A., Guy, L., & Habrová, H. (2016). Carbon stock in agroforestry coffee plantations with different shade trees in Villa Rica, Peru. Agroforestry systems, 90 , 433–445. Englund, S. R., O'Brien, J. J., & Clark, D. B. (2000). Evaluation of digital and film hemispherical photography and spherical densiometry for measuring forest light environments. Canadian Journal of Forest Research, 30(12), 1999–2005. FAO and UNEP. (2020). The State of the World’s Forests 2020. Forests, biodiversity and people . Rome. FAO (2020). Global Forest Resources Assessment 2020 (FRA 2020). Food and Agriculture Organization of the United Nations, Rome. URL: https://doi.org/10.4060/ca9825en Gardner, T. A., Barlow, J., Chazdon, R., Ewers, R. M., Harvey, C. A., Peres, C. A., & Sodhi, N. S. (2009). Prospects for tropical forest biodiversity in a human-modified world. Ecology letters, 12(6), 561–582. Gibson, L., Lee, T. M., Koh, L. P., Brook, B. W., Gardner, T. A., Barlow, J., … Sodhi, N. S. (2011). Primary forests are irreplaceable for sustaining tropical biodiversity. Nature, 478(7369), 378–381. Global Forest Watch. (2023). Democratic Republic of the Congo (COD) Forest Data. In Global Forest Watch. Retrieved June 28, 2023, from https://www.globalforestwatch.org/dashboards/country/COD Głowacki, Ł. (2011). Accuracy of species richness estimators applied to fish in small and large temperate lowland rivers. Biodiversity and Conservation, 20, 1365–1384. Gordon, C., Manson, R., Sundberg, J., & Cruz-Angon, A. (2006). Biodiversity, profitability, and vegetation structure in a Mexican coffee agroecosystem. Agriculture, Ecosystems and Environment, 118 , 256–266. Gotelli, N. J., & Chao, A. (2013). Measuring and estimating species richness, species diversity, and biotic similarity from sampling data. Greig-Smith, P. (1983). Quantitative Plant Ecology . Blackwell Scientific Publications. Guillemot, J., Le Maire, G., Munishamappa, M., Charbonnier, F., & Vaast, P. (2018). Native coffee AF in the Western Ghats of India maintains higher carbon storage and tree diversity compared to exotic AF. Agriculture, Ecosystems & Environment, 265 , 461–469. Häger, A. (2012). The effects of management and plant diversity on carbon storage in coffee AF systems in Costa Rica. AF Systems, 86(2), 159–174. Hairiah, K., Dewi, S., Agus, F., Velarde, S., Ekadinata, A., Rahayu, S., Van Noordwijk, M. (2011 ). Measuring Carbon Stocks Across Land Use Systems: A Manual . World AF Centre (ICRAF), SEA Regional Office, Bogor Indonesia. Harja, D., Rahayu, S., and Pambudi, S. (2019). Tree functional attributes and ecological database . Harris, N. L., Gibbs, D. A., Baccini, A., Birdsey, R. A., de Bruin, S., Farina, M., Fatoyinbo, L., Hansen, M. C., Herold, M., Houghton, R. A., Potapov, P. V., Suarez, D. R., Roman-Cuesta, R. M., Saatchi, S. S., Slay, C. M., Turubanova, S. A., & Tyukavina, A. (2021). Global maps of twenty-first century forest carbon fluxes. Nature Climate Change, 11 , 234–240. Hill, M.O. (1973). Diversity and evenness: a unifying notation and its consequences. Ecology, 54, 427–432. Hundera, K., Aerts, R., Fontaine, A., Van Mechelen, M., Gijbels, P., Honnay, O., and Muys, B. (2012). Effects of wild coffee management intensity on composition, structure, and regeneration of Ethiopian montane rainforests. Environmental Management, 51, 801–809. ICRAF (2007). Wood Density Database . World AF Center: Wood Density Database. http://www.worldAF.org/sea/Products/AFDbases/WD/Index.htm (accessed on 2.15.22). ICO, CFC, and WB (2000) Study of Marketing and Trading Policies and Systems in Selected Coffee Producing Countries - Democratic Republic of the Congo . ICO (2023) Historical Data on the Global Coffee Trade - DRC Robusta Coffee Production. Statistical Publication of the International Coffee Organization . 2023. https://icocoffee.org/resources/historical-data-on-the-global-coffee-trade/ Idol, T.W., and Youkhana, A.H. (2020). A rapid visual estimation of fruits per lateral to predict coffee yield in Hawaii. AF Systems, 94(1), 81–93. InsideWood database, 2021. InsideWood NCSU. Published on the Internet. http://insidewood.lib.ncsu.edu/search (accessed on 2.15.21). IPCC (2006). IPCC Guidelines for National Greenhouse Gas Inventories, Forestry . The Institute for Global Environmental Strategies (IGES), Japan. Jayathilake, H. M., Prescott, G. W., Carrasco, L. R., Rao, M., & Symes, W. S. (2021). Drivers of deforestation and degradation for 28 tropical conservation landscapes. Ambio, 50 , 215–228. Jha, S., Bacon, C. M., Philpott, S. M., Ernesto Méndez, V., Läderach, P., & Rice, R. A. (2014). Shade coffee: update on a disappearing refuge for biodiversity. BioScience, 64(5), 416–428. Jones, A., Breuning-Madsen, H., Brossard, M., Dampha, A., Deckers, J., Dewitte, O., … Zougmoré, R. B. (2013). Soil atlas of Africa . European Commission. Kearsley, E., De Haulleville, T., Hufkens, K., Kidimbu, A., Toirambe, B., Baert, G., Huygens, D., Kebede, Y., Defourny, P., Bogaert, J., and Beeckman, H. (2013). Conventional tree height–diameter relationships significantly overestimate above-ground carbon stocks in the Central Congo Basin. Nature Communications, 4(1), 1–8. Kearsley, E. (2015). Carbon storage and functional diversity of tropical rainforest in the Congo Basin (Doctoral dissertation, Ghent University). Kindt, R., & Coe, R. (2005). Tree Diversity Analysis: A Manual and Software for Common Statistical Methods for Ecological and Biodiversity Studies . Nairobi, Kenya: World AF Centre (ICRAF). Koutouleas, A., Sarzynski, T., Bertrand, B., Bordeaux, M., Bosselmann, A. S., Campa, C., … Ræbild, A. (2022). Shade effects on yield across different Coffea arabica cultivars—how much is too much? A meta-analysis. Agronomy for Sustainable Development, 42(4), 55. Lashermes, P., Combes, M. C., Ribas, A., Cenci, A., Mahé, L., & Etienne, H. (2010). Genetic and physical mapping of the SH3 region that confers resistance to leaf rust in coffee tree (Coffea arabica L.). Tree Genetics & Genomes, 6 (6), 973–980. Leakey, R. R., Tchoundjeu, Z., Schreckenberg, K., Shackleton, S. E., & Shackleton, C. M. (2005). AF tree products (AFTPs): targeting poverty reduction and enhanced livelihoods. International Journal of Agricultural Sustainability, 3 (1), 1–23. Lemmon, P. E. (1956). A spherical densiometer for estimating forest overstory density. Forest science, 2(4), 314–320. López-Gómez, A. M., Williams-Linera, G., & Manson, R. H. (2007). Tree species diversity and vegetation structure in shade coffee farms in Veracruz, Mexico. Agriculture, Ecosystems and Environment, 124 , 160–172. Malhi, Y., Adu-Bredu, S., Asare, R. A., Lewis, S. L., & Mayaux, P. (2013). African rainforests: past, present and future. Philosophical Transactions of the Royal Society B: Biological Sciences, 368 , 20120312. Migolet, P., Goïta, K., Ngomanda, A., & Mekui Biyogo, A. P. (2020). Estimation of above-ground oil palm biomass in a mature plantation in the Congo Basin. Forests, 11(5), 544. Miller, D. C., Ordoñez, P. J., Brown, S. E., Forrest, S., Nava, N. J., Hughes, K., & Baylis, K. (2020). The impacts of AF on agricultural productivity, ecosystem services, and human well-being in low‐and middle‐income countries: An evidence and gap map. Campbell Systematic Reviews , 16(1). Moguel, P., & Toledo, V. (1999). Biodiversity conservation in traditional coffee systems of Mexico. Conservation Biology, 13 , 11–21. Mokondoko, P., Avila-Foucat, V. S., & Galeana-Pizaña, J. M. (2022). Biophysical drivers of yield gaps and ecosystem services across different coffee-based AF management types: A global meta-analysis. Agriculture, Ecosystems & Environment, 337 , 108024. Moonen, P. (2017). Integrated assessment of options to reduce emissions from deforestation in Tshopo, DR Congo. KU Leuven , Leuven, Belgium. Moonen, P. C., Verbist, B., Bosela, F. B., Norgrove, L., Dondeyne, S., Van Meerbeek, K., … & Muys, B. (2019). Disentangling how management affects biomass stock and productivity of tropical secondary forests fallows. Science of the Total Environment, 659, 101–114. Muñoz, J. C., Aerts, R., Thijs, K. W., Stevenson, P. R., Muys, B., & Sekercioglu, C. H. (2013). Contribution of woody habitat islands to the conservation of birds and their potential ecosystem services in an extensive Colombian rangeland. Agriculture, ecosystems and environment, 173, 13–19. Muschler, R. G. (2001). Shade improves coffee quality in a sub-optimal coffee-zone of Costa Rica. Agroforestry systems, 51 , 131–139. Nair, P. R., Kumar, B. M., & Nair, V. D. (2021). An Introduction to AF: Four Decades of Scientific Developments . Cham: Springer. Niyibigira, E. I. (2019). Robusta Coffee Handbook: A Sustainable Coffee Industry with High Stakeholder Value for Social Economic Transformation . The Ministry Of Agriculture, Animal Industry & Fisheries, Uganda. Oksanen, A. J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., … Wagner, H. (2020). Package 'vegan.' Pardini, R. (2004). Effects of forest fragmentation on small mammals in an Atlantic Forest landscape. Biodiversity & Conservation, 13, 2567–2586. Perfecto, I., Rice, R. A., Greenberg, R., & Van der Voort, M. E. (1996). Shade coffee: a disappearing refuge for biodiversity. BioScience, 46 , 598–608. Perfecto, I., Mas, A., Dietsch, T., & Vandermeer, J. (2003). Conservation of biodiversity in coffee agroecosystems: a tri-taxa comparison in southern Mexico. Biodiversity & Conservation, 12, 1239–1252. Perfecto, I., Vandermeer, J., Mas, A., & Pinto, L. S. (2005). Biodiversity, yield, and shade coffee certification. Ecological Economics, 54(4), 435–446. Perfecto, I., Armbrecht, I., Philpott, S. M., Soto-Pinto, L., & Dietsch, T. V. (2006). Shaded coffee and the stability of rainforest margins in northern Latin America. In T. Tscharntke, C. Leuschner, M. Zeller, E. Guhardja, & A. Bidin (Eds.), The stability of tropical rainforest margins, linking ecological, economic and social constraints of land use and conservation (pp. 227–263). Springer Verlag Berlin. Perfecto, I., & Vandermeer, J. (2008). Biodiversity conservation in tropical agroecosystems: a new conservation paradigm. Annals of the New York Academy of Sciences, 1134(1), 173–200. Perfecto, I., Vandermeer, J., & Philpott, S. M. (2014). Complex ecological interactions in the coffee agroecosystem. Annual Review of Ecology, Evolution, and Systematics , 45, 137–158.µ Philpott, S. M., Bichier, P., & Rice, R. A. (2008). Biodiversity conservation, yield, and alternative products in coffee agroecosystems in Sumatra, Indonesia. Biodiversity and Conservation, 17 , 1805–1820. Philpott, S. M., & Bichier, P. (2011). Effects of shade tree removal on birds in coffee agroecosystems in Chiapas, Mexico. Agriculture, Ecosystems and Environment, 149 , 171–180. Piato, K., Lefort, F., Subía, C., Caicedo, C., Calderón, D., Pico, J., & Norgrove, L. (2020). Effects of shade trees on robusta coffee growth, yield and quality: A meta-analysis. Agronomy for Sustainable Development, 40(6), 1–13. Pielou, E. C. (1969). An Introduction to Mathematical Ecology . New York: Wiley. PROTA. (2012). Protabase [WWW Document]. Prota4U Web Database . Retrieved from http://www.prota4u.org/ (Accessed on February 15, 2022). Ravindranath, N. H., & Ostwald, M. (2008). Carbon Inventory Methods: Handbook for Greenhouse Gas Inventory, Carbon Mitigation and Roundwood Production . Springer Science + Business Media B.V. R Core Team (2020). R: A language and environment for statistical computing . Vienna, Austria: R Foundation for Statistical Computing. Richards, M. B., & Méndez, V. E. (2014). Interactions between carbon sequestration and shade tree diversity in a smallholder coffee cooperative in El Salvador. Conservation Biology, 28(2), 489–497. Ricotta, C., Podani, J., & Uv, E. D. (2017). On some properties of the Bray-Curtis dissimilarity and their ecological meaning. Ecological Complexity, 31, 201–205. Rice, R. A. (2008). Agricultural intensification within AF: the case of coffee and wood products. Agriculture, ecosystems & environment, 128 (4), 212–218. Rice, R. A. (2011). Fruits from shade trees in coffee: how important are they? AF systems, 83 , 41–49. Royal Museum for Central Africa (RMCA) (2021). Data derived via DRYAD. Schmitt, C. B., Senbeta, F., Denich, M., Preisinger, H., & Boehmer, H. J. (2010). Wild coffee management and plant diversity in the montane rainforest of southwestern Ethiopia. African Journal of Ecology, 48(1), 78–86. Schmitt-Harsh, M., Evans, T. P., Castellanos, E., & Randolph, J. C. (2012). Carbon stocks in coffee agroforests and mixed dry tropical forests in the western highlands of Guatemala. AF Systems, 86, 141–157. Schroth, G., & Harvey, C. A. (2007). Biodiversity conservation in cocoa production landscapes: an overview. Biodiversity and Conservation, 16, 2237–2244. Segura, M., Kanninen, M., & Suárez, D. (2006). Allometric models for estimating above-ground biomass of shade trees and coffee bushes grown together. AF Systems, 68(2), 143–150. Shannon, C. E., & Weaver, W. (1949). The Mathematical Theory of Communication . Champaign, IL: University of Illinois Press. Silva, M. D. C., Várzea, V., Guerra-Guimarães, L., Azinheira, H. G., Fernandez, D., Petitot, A. S., … Nicole, M. (2006). Coffee resistance to the main diseases: leaf rust and coffee berry disease. Brazilian journal of plant physiology, 18 , 119–147. Solis, R., Vallejos-Torres, G., Arévalo, L., Marín-Díaz, J., Ñique-Alvarez, M., Engedal, T., & Bruun, T. B. (2020). Carbon stocks and the use of shade trees in different coffee growing systems in the Peruvian Amazon. The Journal of Agricultural Science, 158 (6), 450–460. Somarriba, E., Cerda, R., Orozco, L., Cifuentes, M., Dávila, H., Espin, T., Mavisoy, H., Ávila, G., Alvarado, E., Poveda, V., & Astorga, C. (2013). Carbon stocks and cocoa yields in AF systems of Central America. Agriculture, Ecosystems & Environment, 173, 46–57. Soto-Pinto, L., Anzueto, M., Mendoza, J., Ferrer, G. J., & de Jong, B. (2010). Carbon sequestration through agroforestry in indigenous communities of Chiapas, Mexico. Agroforestry Systems, 78 , 39–51. Tegegne, Y. T., Lindner, M., Fobissie, K., & Kanninen, M. (2016). Evolution of drivers of deforestation and forest degradation in the Congo Basin forests: exploring possible policy options to address forest loss. Land Use Policy, 51 , 312–324. Trefon, T. (2023). Bushmeat: Culture, Economy and Conservation in Central Africa . Tyukavina, A., Hansen, M. C., Potapov, P., Parker, D., Okpa, C., Stehman, S. V., Kommareddy, I., & Turubanova, S. (2018). Congo Basin forest loss dominated by increasing smallholder clearing. Science Advances, 4 , eaat2993. Vaast, P., Bertrand, B., Perriot, J. J., Guyot, B., & Genard, M. (2006). Fruit thinning and shade improve bean characteristics and beverage quality of coffee (Coffea arabica L.) under optimal conditions. Journal of the Science of Food and Agriculture, 86 (2), 197–204. van Asten, P. J., Wairegi, L. W. I., Mukasa, D., & Uringi, N. O. (2011). Agronomic and economic benefits of coffee–banana intercropping in Uganda’s smallholder farming systems. Agricultural Systems, 104(4), 326–334. Vanden Abeele, S., Janssens, S. B., Asimonyio Anio, J., Bawin, Y., Depecker, J., Kambale, B., … Vandelook, F. (2021). Genetic diversity of wild and cultivated Coffea canephora in north-eastern DR Congo and the implications for conservation. American Journal of Botany, 108(12), 2425–2434. Vanderhaegen, K., Verbist, B., Hundera, K., & Muys, B. (2015). REALU vs. REDD+: carbon and biodiversity in the Afromontane landscapes of SW Ethiopia. Forest Ecology and Management, 343, 22–33. Van Noordwijk, M., Rahayu, S., Hairiah, K., Wulan, Y. C., Farida, A., & Verbist, B. (2002). Carbon stock assessment for a forest-to-coffee conversion landscape in Sumber-Jaya (Lampung, Indonesia): from allometric equations to land use change analysis. Science in China Series C Life Siences, 45 , 75–86. Van Noordwijk, M., Duguma, L. A., Dewi, S., Leimona, B., Catacutan, D. C., Lusiana, B., … & Minang, P. A. (2018). SDG synergy between agriculture and forestry in the food, energy, water and income nexus: reinventing agroforestry?. Current opinion in environmental sustainability, 34, 33–42. Wintgens, J. N. (2004). Coffee: growing, processing, sustainable production. A guidebook for growers, processors, traders, and researchers . WILEY-VCH Verlag GmbH & Co. KGaA. World Bank Data Bank. (2021). World Development Indicators, DRC. URL: https://databank.worldbank.org/source/worlddevelopment-indicators World Bank (2022). Country profile: Democratic Republic of the Congo. Retrieved from https://data.worldbank.org/country/CD Wurz, A., Tscharntke, T., Martin, D. A., Osen, K., Rakotomalala, A. A., Raveloaritiana, E., … Grass, I. (2022). Win-win opportunities combining high yields with high multi-taxa biodiversity in tropical AF. Nature Communications, 13(1), 4127. Zewdie, B., Tack, A. J., Ayalew, B., Wondafrash, M., Nemomissa, S., & Hylander, K. (2022). Plant biodiversity declines with increasing coffee yield in Ethiopia's coffee agroforests. Journal of Applied Ecology, 59(5), 1198–1208. Additional Declarations No competing interests reported. Supplementary Files Appendices.docx Cite Share Download PDF Status: Published Journal Publication published 14 Feb, 2025 Read the published version in Agroforestry Systems → Version 1 posted Editorial decision: Revision requested 19 Nov, 2024 Reviews received at journal 15 Nov, 2024 Reviews received at journal 01 Oct, 2024 Reviewers agreed at journal 01 Oct, 2024 Reviewers agreed at journal 30 Sep, 2024 Reviewers invited by journal 30 Sep, 2024 Editor assigned by journal 30 Sep, 2024 Submission checks completed at journal 28 Sep, 2024 First submitted to journal 27 Sep, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5165806","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":380035834,"identity":"c12972c4-2fbf-44bd-9e84-725730c4f9db","order_by":0,"name":"Ieben Broeckhoven","email":"data:image/png;base64,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","orcid":"","institution":"KU Leuven","correspondingAuthor":true,"prefix":"","firstName":"Ieben","middleName":"","lastName":"Broeckhoven","suffix":""},{"id":380035836,"identity":"99c9e283-4ff7-443b-b8c5-2895b4a9fd17","order_by":1,"name":"Jonas Depecker","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Jonas","middleName":"","lastName":"Depecker","suffix":""},{"id":380035839,"identity":"f05ebff2-6f2b-45ac-801d-2bd8a35ceee1","order_by":2,"name":"Trésor Kasereka Muliwambene","email":"","orcid":"","institution":"UNIKIS","correspondingAuthor":false,"prefix":"","firstName":"Trésor","middleName":"Kasereka","lastName":"Muliwambene","suffix":""},{"id":380035841,"identity":"862c1f51-e0a5-4867-91d5-7c2aef419de9","order_by":3,"name":"Olivier Honnay","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Olivier","middleName":"","lastName":"Honnay","suffix":""},{"id":380035842,"identity":"34bd040f-17a5-4bd1-8515-6fc70c7243d4","order_by":4,"name":"Roel Merckx","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Roel","middleName":"","lastName":"Merckx","suffix":""},{"id":380035843,"identity":"e44c164d-fed1-4b9f-89da-a970c854320e","order_by":5,"name":"Bruno Verbist","email":"","orcid":"","institution":"KU Leuven","correspondingAuthor":false,"prefix":"","firstName":"Bruno","middleName":"","lastName":"Verbist","suffix":""}],"badges":[],"createdAt":"2024-09-27 14:23:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5165806/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5165806/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10457-025-01140-9","type":"published","date":"2025-02-14T15:56:54+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":72317762,"identity":"bc3c74ce-c720-42e4-bec4-3e5a6b961d46","added_by":"auto","created_at":"2024-12-25 08:32:37","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":121886,"visible":true,"origin":"","legend":"\u003cp\u003eMap of the study area in the Tshopo province, DR Congo, with the location of coffee plots, encompassing the four coffee typologies: Monoculture (MC), Cultivated coffee Agroforestry (CAF), Wild coffee Agroforestry (WAF) and Forest Coffee (FC).\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5165806/v1/04e8d924ca39f49f8a4b4654.jpeg"},{"id":72317672,"identity":"90483f00-e03f-41fe-bd51-0e8753caa8e8","added_by":"auto","created_at":"2024-12-25 08:24:37","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1251953,"visible":true,"origin":"","legend":"\u003cp\u003eRepresentative examples of the four different robusta coffee systems found in the Tshopo province, DR Congo: Monoculture (MC), Cultivated coffee Agroforestry (CAF), Wild coffee Agroforestry (WAF) and Forest Coffee (FC).\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5165806/v1/0e49ad646a118a722931a13e.jpeg"},{"id":72317667,"identity":"d501e4cf-b70b-4985-94fd-c87ada105857","added_by":"auto","created_at":"2024-12-25 08:24:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":24256,"visible":true,"origin":"","legend":"\u003cp\u003eThe median carbon stocks (± se) are compared per component between three coffee systems (Mg C ha-1) based on 27 monoculture (MC), 15 Cultivated Agroforestry (CAF), 12 Wild Agroforestry (WAF) and 25 Forest Coffee (FC) plots (N=79). Different letters and colours indicate significant differences between the different coffee systems based on the Kruskal-Wallis rank sum test and pairwise Wilcoxon rank sum tests with Benjamini-Hochberg correction for multiple testing for comparing non-normal distributions.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-5165806/v1/a1ac3b858dbdec8ad4ea7df3.png"},{"id":72317669,"identity":"789195ec-181d-41b9-a484-85c4adcf2cc6","added_by":"auto","created_at":"2024-12-25 08:24:37","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":25572,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of the biodiversity indices for the agroforestry and forest coffee systems based 12 Cultivated Agroforestry (CAF), 12 Wild Agroforestry (WAF) and 25 forest coffee (FC) plots (N=49). Different letters and colours indicate significant differences between the different coffee systems based on the Kruskal-Wallis rank sum test and pairwise Wilcoxon rank sum tests with Benjamini-Hochberg correction for multiple testing for comparing non-normal distributions.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-5165806/v1/319851100ac1aa756259eae6.png"},{"id":72317761,"identity":"a4406fda-6af7-4a45-b86d-028d988cb094","added_by":"auto","created_at":"2024-12-25 08:32:37","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":8888,"visible":true,"origin":"","legend":"\u003cp\u003eNon-metric multidimensional scaling (NMDS) ordination of 144 coffee plots in the Tshopo province, DRC, comparing 15 Cultivated Agroforestry (CAF), 12 Wild Agroforestry (WAF) and 125 forest coffee (FC) plots. The ordination is based on Bray-Curtis calculated using tree species abundances (DBH ≥ 5cm). The ellipses presented are the dispersion ellipses using the standard deviation of the mean. Ellipses: green (FC), blue (WAF), red (CAF). The tree species community compositions significantly differed between the forest coffee and agroforestry systems. Only 3 out of 249 woody species were common among the three coffee systems.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-5165806/v1/a2d168e87fc780fd2e7c3b4b.png"},{"id":72317673,"identity":"1cbaf6d6-3f2f-4413-a4d4-7a6614a15133","added_by":"auto","created_at":"2024-12-25 08:24:37","extension":"jpeg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":170781,"visible":true,"origin":"","legend":"\u003cp\u003eOverview relations between the biodiversity indices and the Total Organic Carbon (TOC) stock in megagram carbon per ha (Mg C ha\u003csup\u003e-1\u003c/sup\u003e) based on 12 Cultivated Agroforestry (CAF), 12 Wild Agroforestry (WAF) and 25 forest coffee plots (FC) (N = 49). S\u003csub\u003eobs\u003c/sub\u003e is the observed species richness, S\u003csub\u003eexp\u003c/sub\u003e is the expected species richness, N\u003csub\u003e1\u003c/sub\u003e and N\u003csub\u003e2\u003c/sub\u003e represent Hill’s woody species diversity, while E\u003csub\u003e1\u003c/sub\u003e and E\u003csub\u003e2\u003c/sub\u003e represent Hill’s evenness of woody species. Shaded area’s represent the 95% confidence intervals.\u003c/p\u003e","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5165806/v1/bf8b6fb29fbe55513597e3e3.jpeg"},{"id":72317671,"identity":"61f40b25-7fa6-4141-8257-76e2edb4dbd0","added_by":"auto","created_at":"2024-12-25 08:24:37","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":69429,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between coffee yield (kg green coffee ha-1) and (A) Total Organic Carbon (Mg C\u0026nbsp; ha\u003csup\u003e-1\u003c/sup\u003e) and (B) observed number of species Sobs as biodiversity index based on 27 monoculture (MC), 12 Cultivated Agroforestry (Cultivated), 12 Wild Agroforestry (Wild) and 25 forest coffee (FC) plots (N = 76). Shaded area’s represent the 95% confidence intervals.\u003c/p\u003e","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5165806/v1/fca1273b1bbf053c2ae7d1b8.jpeg"},{"id":76487414,"identity":"cde2a3f2-1fb1-4b61-aa9f-148551938712","added_by":"auto","created_at":"2025-02-17 16:05:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3040096,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5165806/v1/3ec8eddf-b60d-45a0-a362-eb9a6ec79796.pdf"},{"id":72317666,"identity":"9d5b9a10-506b-4a4c-8492-d8e388ad36eb","added_by":"auto","created_at":"2024-12-25 08:24:37","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":26280,"visible":true,"origin":"","legend":"","description":"","filename":"Appendices.docx","url":"https://assets-eu.researchsquare.com/files/rs-5165806/v1/b8d52bb724a7344db3d800bc.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Synergies and Trade-offs between Robusta Yield, Carbon Stocks and Biodiversity across Coffee Systems in the DR Congo","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAfrica stands out as being the home to the second-largest tropical rainforest globally, with approximately 89% situated within the Congo Basin (Harris et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Malhi et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Moreover, 60 % of these tropical rainforests of the Cngo Basin are found within the Democratic Republic of Congo (DRC) (Malhi et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Furthermore, a significant portion (11 %) of the global tree cover loss between 2000 and 2021 occured in the Congo Basin (Global Forest Watch, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), with the highest deforestation rates observed in the DRC (FAO, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The Tshopo province experienced the largest tree cover loss of all provinces within the DRC during this period (Global Forest Watch, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Tropical rainforest loss is a major cause of climate change and biodiversity decline, with agriculture being the primary driver (Curtis et al., 2018; Feng et al., 2022; Jayathilake et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Pendrill et al., 2022). Contrary to tropical forests in Asia and South America, the expansion of small-scale subsistence agriculture, instead of large-scale commodity-driven agriculture, is the major driver of deforestation in Africa, accounting for 97 % of all agriculture-driven forest loss (Branthomme et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Curtis et al., 2018; Jayathilake et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Pendrill et al., 2022). Notably, within the DRC, subsistence agriculture, mainly in the form of slash-and-burn agriculture, accounts for a staggering 93 % of deforestation, representing the secondhighest rate among the Congo Basin countries (Tegegne et al., \u003cspan citationid=\"CR104\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Tyukavina et al., \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The proportion of the DRC\u0026rsquo;s deforestation caused by agriculture (93 %) significantly surpasses the regional average (68 %) for the Congo Basin countries (Tyukavina et al., \u003cspan citationid=\"CR106\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn this context of deforestation driven by small-scale subsistence agriculture, tropical agroforestry is often advocated as it may reconcile the need for agricultural productivity, biodiversity conservation, carbon storage, and human well-being by leveraging synergies (Castle et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Miller et al., \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Nair et al., 2022; Van Noordwijk et al., \u003cspan citationid=\"CR112\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Wurz et al., \u003cspan citationid=\"CR116\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). A systematic review of productivity and ecosystem service provisioning of agroforestry in low- and middle-income countries worldwide found an overall positive impact of tropical agroforestry interventions on yield, although there was considerable heterogeneity depending on the type of intervention (Castle et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Furthermore, the study revealed a small but overall positive impact on income. Surprisingly, there were only a limited number of studies dealing with the environmental benefits of agroforestry interventions. Nonetheless, the few existing studies found environmental benefits from agroforestry interventions (Castle et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). While trade-offs exist in agroforestry systems, it is vital to identify win-win opportunities. For example, Wurz et al. (\u003cspan citationid=\"CR116\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) reported high yields in combination with high multi-taxa biodiversity in vanilla agroforests in Madagascar. Thus, the question whether similar synergies exist for coffee agroforestry systems arises.\u003c/p\u003e \u003cp\u003eOn the one hand, numerous studies have reported positive effects of coffee agroforestry systems on biodiversity conservation and carbon stocks compared to coffee monocultures. Several examples can be found mainly in Latin-America (Alvarez-Alvarez et al., 2021; Gordon et al., 2007; L\u0026oacute;pez-G\u0026oacute;mez et al., 2008; Moguel \u0026amp; Toledo, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Perfecto et al., \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2003\u003c/span\u003e, 2007; Philpott \u0026amp; Bichier, 2012), but also in India (Guillemot et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), Indonesia (Philpott et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2008\u003c/span\u003e), and Ethiopia (De Beenhouwer et al., 2016). On the other hand, competition for space, light, nutrients, and water, as well as changes in the incidence and severity of coffee pests and diseases, can positively and negatively influence coffee productivity within agroforestry systems (Durand-Bessart et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Koutouleas et al., \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Mokondoko et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Piato et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Additionally, shade trees can be of significant economic value, generating additional income and income diversification, which is known to be particularly important during periods of low coffee prices (H. Davis et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Leakey et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Philpott et al., \u003cspan citationid=\"CR83\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Rice, \u003cspan citationid=\"CR92\" class=\"CitationRef\"\u003e2008\u003c/span\u003e, \u003cspan citationid=\"CR93\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Moreover, coffee agroforestry systems have been found to have equal or better economic importance than conventional systems (Jezeer et al., 2018). Studies have proven that coffee agroforestry systems can have higher profitability and cost-efficiency overall than conventional systems (Jezeer et al., 2017, 2019). Besides effects on coffee yield, variable impacts of shade on coffee quality have been reported (Muschler, \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Vaast et al., \u003cspan citationid=\"CR107\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCurrently, robusta coffee \u003cem\u003e(Coffea canephora Pierre ex A. Froehner)\u003c/em\u003e makes up 43% of global coffee production, compared to 56% for arabica coffee \u003cem\u003e(Coffea arabica L.)\u003c/em\u003e (ICO, 2023). The share of robusta production has been increasing over time, and this trend is expected to continue due to climate change (A. P. Davis et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Piato et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Historically, robusta coffee was one of the most valuable export commodities of the DRC, only second to copper (ICO et al., 2000). Moreover, in the 1990s, coffee contributed to three-quarters of the country\u0026rsquo;s agricultural export revenues and about 15 % of the GDP (ICO et al., 2000). However, the DRC\u0026rsquo;s robusta production and export levels arecurrently at a historical low (ICO, 2023). Thus, given that most of the coffee produced in the DRC is robusta (ICO, 2023), in combination with the increasing global demand for robusta, the cultivation of robusta coffee could represent a significant opportunity and source of livelihood for farmers in the DRC. Especially as agriculture in the DRC currently represents 55 % of employment and contributes to more than 30 % of GDP (World Bank, \u003cspan citationid=\"CR115\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn addition to the potential in terms of agricultural production and livelihood generation, it should be noted that the DRC is an important region of origin for robusta coffee. The conservation of its standing coffee genetic diversity is of utmost importance for the future worldwide coffee production as wild populations carry valuable traits for coffee breeding, such as disease resistance (Lashermes et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Silva et al., \u003cspan citationid=\"CR100\" class=\"CitationRef\"\u003e2006\u003c/span\u003e), tolerance to climate change (A. P. Davis et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2012\u003c/span\u003e), and drought tolerance (Davis et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNotable knowledge gaps are the limited research on robusta agroforestry systems compared to arabica, and the scarcity of African studies, particularly in the DRC (De Beenhouwer et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Here, we make a first attempt to systematically quantify robusta coffee yields, carbon stocks and woody species diversity across different robusta coffee cultivation systems in the DRC. More specifically, we compared (i) coffee monocultures, (ii) cultivated coffee agroforestry systems characterised by commonly cultivated tree species intercropped with robusta, (ii) wild coffee agroforestry systems characterised by native undomesticated tree species in addition to commonly cultivated tree species intercropped with robusta, and (iv) forest coffee systems characterised by wild robusta, growing in a natural forest understorey.\u003c/p\u003e \u003cp\u003eThe recent EU Deforestation Regulation (EUDR) policy initiative, aims to reduce the EU\u0026rsquo;s impact on global deforestation and forest degradation, and thereby decreasing greenhouse gas emissions and biodiversity loss, has global ramifications (European Parliament and Council of the European Union, 2023), although its effects- and side effects \u0026ndash; remain poorly understood. The associated EUDR map provides a reference and now allows to assess the impact for particular areas (Bourgoin et al., 2023, 2024).\u003c/p\u003e \u003cp\u003eOur study aims to quantify potential trade-offs and synergies between robusta productivity, carbon stocks and woody species diversity. The specific objectives were fourfold: First, investigate the extent of variation in robusta yields between the different coffee systems. Second, quantify differences in carbon stocks, encompassing both above- and below-ground components. Third, investigate the variability in woody species diversity among the coffee typologies. Fourth, determine the existence of synergies and trade-offs between coffee yield, carbon stocks, and woody species diversity within these coffee systems, aiming to elucidate the precise nature of these interrelationships. Last, asses the classification of coffee systems as forest or non-forest in 2020 under EUDR.\u003c/p\u003e"},{"header":"2. Methodology","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Study Area\u003c/h2\u003e \u003cp\u003eThe research was carried out in the Tshopo Province, which is situated in the North-East of the Democratic Republic of Congo (DRC). The study area (0.5\u0026ndash;1.4\u0026deg;N, 24\u0026ndash;25\u0026deg;E; 380 m \u0026ndash; 538 m) encompassed the territories of Isangi, Banalia, and Kisangani (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The region\u0026rsquo;s natural vegetation consists of moist semi-deciduous rainforest and monodominant \u003cem\u003eGilbertiodendron dewevrei (De Wild.) J.L\u0026eacute;onard\u003c/em\u003e evergreen rainforest (Gilson et al., 1956). According to the K\u0026ouml;ppen-Geiger classification, the area has a tropical rainforest climate (Af) (Peel et al., 2007), with an average yearly precipitation of 1762\u0026thinsp;\u0026plusmn;\u0026thinsp;295 mm in the Yangambi MAB reserve. The climate is characterised by two rainy seasons (August-November and April-May) and two dry seasons (December-February/March and June-July/August). Throughout the year, average temperatures remain high, with a minimum of 24.2\u0026thinsp;\u0026plusmn;\u0026thinsp;0.4\u0026deg;C in July and a maximum of 25.5\u0026thinsp;\u0026plusmn;\u0026thinsp;0.6\u0026deg;C in March. The dominant WRB soil type in the study area is a Ferralsol, formed from aeolian sediments, consisting mostly of quartz sand, kaolinite clay and hydrated iron oxides (Gilson et al., 1956; Van Ranst et al., 2010). However, due to seasonal inundations, Gleysols can also be found next to rivers and their tributaries (Gilson et al., 1956; Van Ranst et al., 2010).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. The Four Coffee Systems\u003c/h2\u003e \u003cp\u003eFieldwork was conducted between October and December 2021, the main coffee harvesting season. The coffee system typology was based partially on the share of dominant tree in total basal area and tree diversity following Sari et al. (2020). Additionally, canopy closure, origin of coffee planting material and input use were used to distinguish coffee system typologies (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Coffee monocultures (MC) are characterised by robusta cultivars being the only woody species present and having an almost absent tree canopy (canopy closure\u0026thinsp;\u0026lt;\u0026thinsp;5 %). Cultivated cofee agroforestry systems (CAF) are characterised by robusta cultivars being the most abundant woody species present, intercropped with commonly cultivated tree species and having a medium tree canopy closure (38.2\u0026thinsp;\u0026plusmn;\u0026thinsp;4.75 %). Wild coffee aroforestry systems (WAF) are also characterised by robusta cultivars being the most abundant woody species present, intercropped with commonly cultivated tree species, but additionally contain native undomesticated fruit and forest tree species and having a higher tree canopy closure (52.9\u0026thinsp;\u0026plusmn;\u0026thinsp;3.67 %). Forest coffeesystems (FC) are characterised by forest tree species, not wild robusta, being the most abundant woody species, and having a high tree canopy closure (\u0026gt;\u0026thinsp;85 %). The forest cofee plots were located in the Yangambi Man and Biosphere Reserve and the Ngazi Forest Reserve. Additionally, it should be noted that coffee farmers in our study area do not use any pesticides, insecticides or mineral fertilisers. Organic matter is sometimes applied in low quantities in the MC, CAF and WAF systems.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCoffee systems in Tshopo, DRC, and their defining characteristics\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c3\" namest=\"c2\"\u003e \u003cp\u003eMonoculture\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;27)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCultivated Agroforestry\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;15)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eWild Agroforestry\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;12)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eForest Coffee\u003c/p\u003e \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;25)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eICRAF typology\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eMonoculture\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e\u003cem\u003eComplex mixed agroforestry\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003eComplex mixed agroforestry\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003eForest\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTree species (/plot)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e8\u0026thinsp;\u0026plusmn;\u0026thinsp;1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8\u0026thinsp;\u0026plusmn;\u0026thinsp;1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e30\u0026thinsp;\u0026plusmn;\u0026thinsp;1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCoffee as share of total BA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e100%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e31\u0026thinsp;\u0026plusmn;\u0026thinsp;3%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e33\u0026thinsp;\u0026plusmn;\u0026thinsp;4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;1%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDominant species\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRobusta coffee\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eCommonly cultivated tree species\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eCommonly cultivated\u003c/p\u003e \u003cp\u003e\u0026amp; forest tree species\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eForest tree species\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCanopy closure\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003e38\u0026thinsp;\u0026plusmn;\u0026thinsp;4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e53\u0026thinsp;\u0026plusmn;\u0026thinsp;4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u0026gt;\u0026thinsp;85%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCoffee origin\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLocal cultivars\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eLocal cultivars\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLocal cultivars\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eWild\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInput use\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eNo mineral fertilisers,\u003c/p\u003e \u003cp\u003enor pesticides/insecticides\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e \u003cp\u003eNo mineral fertilisers,\u003c/p\u003e \u003cp\u003enor pesticides/insecticides\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eNo mineral fertilisers,\u003c/p\u003e \u003cp\u003enor pesticides/insecticides\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eNo mineral fertilisers,\u003c/p\u003e \u003cp\u003enor pesticides/insecticides\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eBoth Cultivated and Wild Agroforestry fall under complex mixed agroforestry using the method of Sari et al. (2020), as coffee represents less than 50 % of the total baal area, and the number of tree species per plot is greater than 5.\u003c/em\u003e \u003c/p\u003e \u003cp\u003eThe four coffee typologies were then used as the basis for the stratified random sampling design. Specifically, 27 MC, 15 CAF, 12 WAF and 25 FC 25 m x 25 m plots were established (N\u0026thinsp;=\u0026thinsp;79) with a minimum of 1 km distance between plots of the same typology. For forest coffee, five earlier established subplots (25 m x 25 m) within a larger plot of 125 m x 125 m were reused (Depecker et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The data collected from the five subplots per main plot were then averaged to obtain one data point for each measured variable.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Measuring Robusta Coffee Yields\u003c/h2\u003e \u003cp\u003eRobusta yields were measured during the main harvesting season following Idol and Youkhana (\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) and calculated based on four randomly selected coffee plants per plot. To convert the number of ripe cherries to kilograms of green beans, a conversion factor of 3239\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e was used. This conversion factor was based on the average measurements using different robusta coffee varieties at the National Institute of Agronomic Research (INERA). A standardised moisture content for the green beans of 11 % was used. Tree canopyclosure (% shade) was measured using a model A convex spherical densitometer based on four data points per plot (Englund et al., 2011; Lemmon, \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e1956\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4. Assessing the Carbon Stocks of the Coffee Systems\u003c/h2\u003e \u003cp\u003eCarbon stocks were quantified based on the above- and below-ground biomass components per coffee system. For the above-ground biomass, three carbon pools were inventoried (Mg C/ha): above-ground living, deadwood, and litter biomass (IPCC, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). Biomass-related variables measured in the field for woody species with DBH\u0026thinsp;\u0026ge;\u0026thinsp;5cm included tree species, tree height (H), and diameter at breast height (DBH). For forest trees, an exponential regression model of the H:D relationship for the mixed forest in the Tshopo province (Kearsley et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), instead of a general H:D relation for Africa, was used (Eq.\u0026nbsp;1):\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:H=(36.3576-31.6591)*{e}^{-0.0221*\\:D}\\)\u003c/span\u003e \u003c/span\u003e [Eq.\u0026nbsp;1]\u003c/p\u003e \u003cp\u003eDry wood densities (Mg/m\u003csup\u003e3\u003c/sup\u003e) were obtained from the literature (Carsan et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Depecker et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2022\u003c/span\u003eb; Harja et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; NCSU, n.d.; Royal Museum for Central Africa, 2021). The dry weight of above-ground living biomass and deadwood biomass was calculated using allometric equations (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). In the case of unidentified deadwood or if no species- or genus-specific value was available, the regional average dry wood density of 0.580 Mg/m\u003csup\u003e3\u003c/sup\u003e was used (Brown, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; IPCC, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). Litter was collected from two nested sub-plots (0.125m\u003csup\u003e2\u003c/sup\u003e each) within the sample plots, weighed, sun-dried, and then oven-dried for 72h at 50\u0026deg;C to calculate dry litter biomass. Different conversion factors were used to convert the dry biomass weight into carbon stocks (Mg C/ha): (i) for deadwood, the general conversion factor of 0.50 (IPCC, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2006\u003c/span\u003e), (ii) 0.37 for litter (IPCC, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2006\u003c/span\u003e), (iii) 0.379 for \u003cem\u003eMusa spp.\u003c/em\u003e (Abdullah et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), and (iv) 0.45 for \u003cem\u003eCoffea canephora\u003c/em\u003e were used (Van Noordwijk et al., \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2002\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAllometric equations used for Above Ground Biomass (AGB) and fallen deadwood biomass estimations\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAllometric equation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eSource\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAGB \u0026ndash; General\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eAGB\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.0509 * \u003cem\u003eρ\u003c/em\u003e * \u003cem\u003eDBH\u003c/em\u003e\u0026sup2; * \u003cem\u003eH\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(Chave et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2005\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAGB \u0026ndash; \u003cem\u003eCoffea\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLog(\u003cem\u003eAGB\u003c/em\u003e) = -1.181\u0026thinsp;+\u0026thinsp;1.991*Log(\u003cem\u003eD\u003c/em\u003e\u003csub\u003e\u003cem\u003e15\u003c/em\u003e\u003c/sub\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(Segura et al., \u003cspan citationid=\"CR98\" class=\"CitationRef\"\u003e2006\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAGB \u0026ndash; \u003cem\u003eTheobroma cacao\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLog(\u003cem\u003eAGB\u003c/em\u003e)\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;1.684\u0026thinsp;+\u0026thinsp;2.158 \u0026lowast; Log(\u003cem\u003eD\u003c/em\u003e\u003csub\u003e\u003cem\u003e30\u003c/em\u003e\u003c/sub\u003e)\u0026thinsp;+\u0026thinsp;0.892 \u0026lowast; Log(\u003cem\u003eH\u003c/em\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(Somarriba et al., \u003cspan citationid=\"CR102\" class=\"CitationRef\"\u003e2013\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAGB \u0026ndash; \u003cem\u003eElaeis guineensis\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eAGB\u003c/em\u003e = -1.0007 * exp[-2.335\u0026thinsp;+\u0026thinsp;0.832 * ln(\u003cem\u003eDBH\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e * \u003cem\u003eH\u003c/em\u003e\u003csub\u003e\u003cem\u003et\u003c/em\u003e\u003c/sub\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(Migolet et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2020\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAGB \u0026ndash; \u003cem\u003eMusa spp\u003c/em\u003e.\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eAGB\u003c/em\u003e = -0.0927\u0026thinsp;+\u0026thinsp;0.0203 * \u003cem\u003eDBH\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(Alcudia-Aguilar et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2019\u003c/span\u003e)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFallen deadwood\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eAGB\u003c/em\u003e\u0026thinsp;=\u0026thinsp;π * \u003cem\u003er\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e * \u003cem\u003eρ\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e(Volume cylinder)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cem\u003eAGB\u0026thinsp;=\u0026thinsp;estimated Above Ground Biomass [kg], ρ\u0026thinsp;=\u0026thinsp;wood density [Mg/m\u003c/em\u003e \u003csup\u003e \u003cem\u003e3\u003c/em\u003e \u003c/sup\u003e \u003cem\u003e], DBH\u0026thinsp;=\u0026thinsp;Diameter at Breast Height [cm], D15\u0026thinsp;=\u0026thinsp;Diameter at a height of 15cm [cm], D30\u0026thinsp;=\u0026thinsp;Diameter at a height of 30cm [cm], r\u0026thinsp;=\u0026thinsp;radius, measured halfway the length of the deadwood [cm], H\u0026thinsp;=\u0026thinsp;Tree height [m], H\u003c/em\u003e \u003csub\u003e \u003cem\u003et\u003c/em\u003e \u003c/sub\u003e \u003cem\u003e= Total tree height (height incl. trunk and crown) [m].\u003c/em\u003e\u003c/p\u003e \u003cp\u003eTwo below-ground carbon pools (Mg C/ha) were considered: the root biomass and Soil Organic Carbon (SOC). The root biomass was derived from the above-ground biomass by multiplying it with the root:shoot ratio. For tropical moist forests, a root:shoot ratio of 0.235\u0026thinsp;\u0026plusmn;\u0026thinsp;0.011 was used when living biomass was greater than 125 Mg/ha, while a root:shoot ratio of 0.205\u0026thinsp;\u0026plusmn;\u0026thinsp;0.036 was used when living biomass was smaller than 125 Mg/ha (IPCC, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Mokany et al., 2006). A root:shoot ratio of 0.096 + (0.022 * D15\u0026sup2;) was used for coffee (Andrade et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Composite soil samples based on eight different sampling points per plot were taken at three depth layers (0-5cm, 5-15cm and 15-30cm). These composite soil samples were mixed, sundried, and then oven-dried (50\u0026deg;C for 72 h), after which they were ground with a ball mill and sieved with a mesh size of 2 mm. Carbon content was determined using a Carlo Erba Elemental analyser (Ravindranath \u0026amp; Ostwald, \u003cspan citationid=\"CR88\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Bulk density was measured from undisturbed soil samples using Kopecky rings at the same depths as the organic carbon samples.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5. Identifying Woody Plant Diversity\u003c/h2\u003e \u003cp\u003eIn each plot, the abundance and DBH of the woody tree and shrub species with a DBH\u0026thinsp;\u0026ge;\u0026thinsp;5 cm were measured and recorded. This data was obtained from Depecker et al. (\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) for the forest coffee plots. For each plot, the observed (S\u003csub\u003eobs\u003c/sub\u003e) and expected number of species (S\u003csub\u003eexp\u003c/sub\u003e) and Hill numbers (N\u003csub\u003e1\u003c/sub\u003e, N\u003csub\u003e2\u003c/sub\u003e, E\u003csub\u003e1\u003c/sub\u003e and E\u003csub\u003e2\u003c/sub\u003e) were calculated following Gotelli and Chao (\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Moreover, the expected species richness (S\u003csub\u003eexp\u003c/sub\u003e) was calculated using the Chao1-bc formula (Eq.\u0026nbsp;2) as several plots had no doubletons (Głowacki, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2011\u003c/span\u003e), with S\u003csub\u003eexp\u003c/sub\u003e being the expected number of species, S\u003csub\u003eobs\u003c/sub\u003e being the observed number of species, f\u003csub\u003e1\u003c/sub\u003e the number of singletons and f\u003csub\u003e2\u003c/sub\u003e the number of doubletons:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{S}_{exp}={S}_{obs}\\:+\\:\\frac{{f}_{1}\\cdot\\:\\left({f}_{1}-1\\right)}{2\\cdot\\:\\left({f}_{2}+1\\right)}\\:\\)\u003c/span\u003e \u003c/span\u003e [Eq.\u0026nbsp;2]\u003c/p\u003e \u003cp\u003eThe Hill numbers N\u003csub\u003e0\u003c/sub\u003e, N\u003csub\u003e1\u003c/sub\u003e, N\u003csub\u003e2\u003c/sub\u003e, E\u003csub\u003e1\u003c/sub\u003e (N\u003csub\u003e1\u003c/sub\u003e/ N\u003csub\u003e0\u003c/sub\u003e) and E\u003csub\u003e2\u003c/sub\u003e (N\u003csub\u003e2\u003c/sub\u003e/ N\u003csub\u003e1\u003c/sub\u003e) were calculated using the following formula (Hill, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e1973\u003c/span\u003e), with \u003cem\u003eN\u003c/em\u003e being the total number of organisms of all species and \u003cem\u003ep\u003c/em\u003e\u003csub\u003e\u003cem\u003ei\u003c/em\u003e\u003c/sub\u003e being the proportional abundance of species \u003cem\u003ei\u003c/em\u003e out of \u003cem\u003eS\u003c/em\u003e species:\u003c/p\u003e \u003cp\u003e \u003cspan class=\"InlineEquation\"\u003e \u003cspan class=\"mathinline\"\u003e\\(\\:{N}_{q}={\\sum\\:}_{i=1}^{S}{\\left({p}_{i}^{q}\\right)}^{\\frac{1}{1-q}\\:}\\)\u003c/span\u003e \u003c/span\u003e [Eq.\u0026nbsp;3] \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{E}_{q\\:}=\\:{N}_{q}/\\:{N}_{q-1}\\)\u003c/span\u003e\u003c/span\u003e [Eq.\u0026nbsp;4]\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6. Forest Classification under EUDR\u003c/h2\u003e \u003cp\u003eThe classification of coffee farms as forest or non-forest according to EUDR(Bourgoin et al., 2024; European Parliament and Council of the European Union, 2023) was evaluated using the Global Forest Cover 2020 map and ground-truthing data (Bourgoin et al., 2023, 2024). Analysis was based on the intersection of the coffee systems with forest pixels in the Global Forest Cover 2020 map (Bourgoin et al., 2023). Two different methods were applied using Google Earth Engine. One method used a GPS point in the coffee plot to represent the coffee system, while the other method used the minimum area of the inventoried plot as representing the coffee system. The inventoried plots had dimensions of 25m x 25m (625m\u003csup\u003e2\u003c/sup\u003e). Most coffee farms were larger than the inventoried plot, therefore this can be seen as a conservative approach.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7. Statistical Analysis\u003c/h2\u003e \u003cp\u003eStatistical analyses were performed using R software (R Core Team, \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Prior to testing for significant differences between coffee systems, normality was tested. In the case of unpaired non-normally distributed data, the Kruskal-Wallis and pairwise Wilcoxon rank sum tests were used when comparing the coffee systems in terms of coffee yields, carbon stock elements, and biodiversity indices. In the case of unpaired normally distributed data, Tukey\u0026rsquo;s HSD was used when comparing the coffee systems.\u003c/p\u003e \u003cp\u003eGeneral Additive Models (GAMs) were used to model the relationships between coffee yield, carbon stocks and woody biodiversity, as they allow for the use of non-linear functions that can model more complex relationships while remaining interpretable. Modelling was done using the \u003cem\u003egam\u003c/em\u003e function from the \u003cem\u003emgcv\u003c/em\u003e package in R (R Core Team, \u003cspan citationid=\"CR89\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eNon-metric multidimensional scaling (NMDS) was used to visually represent the dissimilarity in tree community composition across coffee systems. A stress level of 0.07 was obtained for a two-axis solution, well below the threshold stress level of 0.20. The Bray-Curtis dissimilarity measure (Ricotta \u0026amp; Podani, 2017) and the \u003cem\u003emetaMDS\u003c/em\u003e function in the \u003cem\u003evegan\u003c/em\u003e package were used for the ordination (Oksanen et al., 2018). Additionally, a permutational multivariate analysis of variance with 9999 permutations was performed using the \u003cem\u003eadonis\u003c/em\u003e function in the \u003cem\u003evegan\u003c/em\u003e package to test for differences in community composition between the coffee systems. Pairwise comparisons with Benjamini-Hochberg correction among the coffee systems were made using the \u003cem\u003epairwise.adonis\u003c/em\u003e function in the \u003cem\u003evegan\u003c/em\u003e package based on 9999 permutations (Arbizu, 2020).\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1. Quantifying Robusta Coffee Yields and Management Variables\u003c/h2\u003e\n \u003cp\u003eThe highest median robusta coffee yields per plant were found in CAF and MC, respectively, 0.96\u0026thinsp;\u0026plusmn;\u0026thinsp;0.14 and 0.92\u0026thinsp;\u0026plusmn;\u0026thinsp;0.07 kg green beans per plant (Table \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). The median coffee yield per plant tended to be higher in MC than in WAF (p\u0026thinsp;=\u0026thinsp;0.098). Similarly, the yield per plant tended to be higher in CAF than in WAF (p\u0026thinsp;=\u0026thinsp;0.111). No significant yield difference could be found between MC and CAF (p\u0026thinsp;=\u0026thinsp;0.756).\u003c/p\u003e\n \u003cp\u003eSimilarly, the highest median robusta yields per ha were found in MC and CAF, respectively, about 800\u0026thinsp;\u0026plusmn;\u0026thinsp;80 and 650\u0026thinsp;\u0026plusmn;\u0026thinsp;135 kg green beans per ha (Table \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). While the median coffee yield per ha was found to be significantly higher in MC than in WAF (p\u0026thinsp;=\u0026thinsp;0.018), no significant yield difference could be found between MC and CAF (p\u0026thinsp;=\u0026thinsp;0.310), nor between CAF and WAF (p\u0026thinsp;=\u0026thinsp;0.137). In addition, coffee yields per plant and per ha in FC were lower than all other coffee systems (p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) and insignificant from a production perspective (\u0026lt;\u0026thinsp;1 kg/ha).\u003c/p\u003e\n \u003cp\u003eNo significant differences in coffee planting densities could be found between MC and CAF or WAF (p\u0026thinsp;=\u0026thinsp;0.13). Regarding the canopy closure, the highest values were found for FC (91%), which was significantly greater than the 53% for WAF and 38% for CAF (p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001). Despite the large difference in median canopy closure between CAF and WAF, no significant differences could be found (p\u0026thinsp;=\u0026thinsp;0.12).\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable id=\"Tab3\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eComparison of median robusta yields, coffee planting densities and canopy closure between different coffee systems in the Tshopo province, DR Congo.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"12\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colspan=\"3\"\u003e\n \u003cp\u003eMonoculture\u003c/p\u003e\n \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;27)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eCultivated AF\u003c/p\u003e\n \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;15)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eWild AF\u003c/p\u003e\n \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;12)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"3\"\u003e\n \u003cp\u003eForest Coffee\u003c/p\u003e\n \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;25)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026micro;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026plusmn; se\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026micro;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026plusmn; se\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026micro;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026plusmn; se\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026micro;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u0026plusmn; se\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003ep-value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eYield [kg green coffee/plant]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.916\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;0.0746)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.956\u003csup\u003ea,b\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;0.143)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.586\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;0.101)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.00205\u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;0.00136)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eYield [kg green coffee/ha]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e798\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;79.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e648\u003csup\u003ea,b\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;134)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e503\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;94.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.131\u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;0.0868)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eCoffee density [plants/ha]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e960\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;47.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e896\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;61.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e880\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;63.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e64.0\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;3.16)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eCanopy closure [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.00\u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;0.475)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e38.2\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;4.75)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e52.9\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;3.67)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e91.4\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e(\u0026plusmn;\u0026thinsp;0.785)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003e\u0026lt;\u0026thinsp;0.0001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cem\u003eDifferent letters indicate significant differences between the different coffee systems based on the Kruskal-Wallis rank sum test and pairwise Wilcoxon rank sum tests with Benjamini-Hochberg correction for multiple testing for comparing non-normal distributions.\u003c/em\u003e\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2. Quantifying Carbon Stocks\u003c/h2\u003e\n \u003cp\u003eTotal organic carbon stocks were found to be largest in FC (289\u0026thinsp;\u0026plusmn;\u0026thinsp;13.6 Mg C/ha), followed by WAF (99.6\u0026thinsp;\u0026plusmn;\u0026thinsp;10.2 Mg C/ha) and CAF (91.9\u0026thinsp;\u0026plusmn;\u0026thinsp;8.95 Mg C/ha), with MC (58.1\u0026thinsp;\u0026plusmn;\u0026thinsp;2.63 Mg C/ha) having the lowest TOC (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e, Table \u003cspan class=\"InternalRef\"\u003eA1\u003c/span\u003e). FC contained roughly three times more TOC than CAF and WAF systems, which in turn contained almost double the amount of TOC than MC.\u003c/p\u003e\n \u003cp\u003eFurthermore, significant differences were found when comparing the five carbon stock components across the coffee systems, except for litter biomass (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e; Table \u003cspan class=\"InternalRef\"\u003eA1\u003c/span\u003e). Among these carbon stock elements, above-ground living and root biomass were identified as primary drivers for the differences in TOC between coffee systems and deadwood biomass to a lesser extent. In contrast, soil organic carbon (SOC) had a limited impact on the overall differences in TOC. Nonetheless, SOC accounted for 87% of the TOC found in MC while contributing 58% and 47% of TOC in CAF and WAF, respectively, and only 15% of TOC in FC systems.\u003c/p\u003e\n \u003cp\u003eIn addition, MC and CAF systems had significantly larger SOC stocks (0\u0026ndash;30 cm) than FC, with p-values of 0.074 and 0.041, respectively (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). Moreover, significant differences in SOC were only observed in the 0\u0026ndash;5 cm layer (Table \u003cspan class=\"InternalRef\"\u003eA2\u003c/span\u003e). There were no significant differences in soil organic carbon concentration. However, the bulk density was smaller in FC than in MC and CAF for all three depth layers (Table \u003cspan class=\"InternalRef\"\u003eA2\u003c/span\u003e).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003e3.3. Quantifying Woody Biodiversity\u003c/h2\u003e\n \u003cp\u003eOur study recorded 11,001 individuals (DBH\u0026thinsp;\u0026ge;\u0026thinsp;5 cm) from 249 taxa across all plots. Of the 249 observed taxa, 85.27% were identified at the species level, 9.93% at the genus level, and 4.80% were unidentified taxa. In terms of observed woody species, 19 different woody species were found in CAF, 45 in WAF and 214 in FC. As the MC plots, by definition, lacked tree or shrub species apart from coffee, the MC plots were excluded from the calculation of the biodiversity indices.\u003c/p\u003e\n \u003cp\u003eAll biodiversity indices (S\u003csub\u003eobs\u003c/sub\u003e, S\u003csub\u003eexp\u003c/sub\u003e, N\u003csub\u003e1\u003c/sub\u003e, N\u003csub\u003e2\u003c/sub\u003e and E\u003csub\u003e1\u003c/sub\u003e) were found to be significantly larger in FC than in CAF and WAF systems (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e), except for Hill\u0026rsquo;s E\u003csub\u003e2,\u003c/sub\u003e which tended to be lower only in CAF compared to FC (p\u0026thinsp;\u0026lt;\u0026thinsp;0.06). In comparison to FC, both agroforestry systems showed a significant reduction in observed species richness (73\u0026ndash;74%) and expected species richness (78%) (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e, Table \u003cspan class=\"InternalRef\"\u003eA3\u003c/span\u003e). Furthermore, woody species diversity (N\u003csub\u003e1\u003c/sub\u003e and N\u003csub\u003e2\u003c/sub\u003e) was reduced by 86\u0026ndash;90% in CAF and WAF compared to FC. Additionally, the evenness of woody species (E\u003csub\u003e1\u003c/sub\u003e) was reduced by 54% in CAF and WAF systems compared to FC, while Hill\u0026apos;s evenness (E\u003csub\u003e2\u003c/sub\u003e) was only 7% lower in CAF than in FC (Table \u003cspan class=\"InternalRef\"\u003eA3\u003c/span\u003e).\u003c/p\u003e\n \u003cp\u003eThe woody plant communities in both agroforestry systems were found to differ greatly from FC communities along both NMDS axes (R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.3242, P\u0026thinsp;\u0026lt;\u0026thinsp;0.001) (Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e). Additionally, the variability of community composition was smaller in the CAF than in the WAF system, with both having a smaller variability than the FC system. Of the 249 tree species observed, only three woody species were common among all three coffee systems, namely: \u003cem\u003eCoffea canephora, Dacryodes edulis (G.Don) H.J.Lam.\u003c/em\u003e and \u003cem\u003eAnonidium mannii (Oliv.) Engl. \u0026amp; Diels\u003c/em\u003e. Furthermore, FC and CAF only shared 3 woody species, the ones common to all three coffee systems. WAF was found to have 13 woody species in common with CAF and 14 in common with FC. Besides, 4 woody species were found to be unique to CAF, 21 unique to WAF and 200 unique to FC.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\n \u003ch2\u003e3.4. Identifying Synergies and Trade-offs between Robusta Yield, Woody Species Diversity and Carbon Stocks\u003c/h2\u003e\n \u003cp\u003eA significantly positive relationship was found between all biodiversity indices and TOC, except Hill\u0026apos;s E2 index (Fig. \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e), indicating co-benefits between woody species diversity and carbon stocks.\u003c/p\u003e\n \u003cp\u003eThe relationship between total organic carbon (TOC) and yield exhibited a convex pattern (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e). A sharp initial decline in TOC was observed when moving from FC (high TOC) to the CAF and WAF systems (medium TOC), which was accompanied by a large increase in coffee yields. Likewise, a trade-off could be identified when moving from WAF (medium TOC \u0026amp; medium yield) to MC (low TOC \u0026amp; high yield). However, when moving from CAF (medium TOC \u0026amp; high yield) to MC (low TOC \u0026amp; high yield), a loss of TOC was observed without an increase in coffee yield, in contrast to the other shifts where trade-offs between TOC and coffee yield occurred.\u003c/p\u003e\n \u003cp\u003eSimilar to the TOC-yield relation, the relationship between biodiversity and yield was examined using generalised additive models (GAMs), revealing a convex pattern (Fig. \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003e). When moving from FC to the coffee agroforestry systems, a rapid decline in woody species richness (S\u003csub\u003eobs\u003c/sub\u003e) was observed, accompanied by a large increase in coffee yield. In addition, a trade-off could be identified when moving from WAF (medium biodiversity \u0026amp; medium yield) to MC (low biodiversity \u0026amp; high yield). However, when moving from CAF (medium biodiversity \u0026amp; high yield) to MC (low biodiversity \u0026amp; high yield), a loss of woody species diversity was observed without an increase in coffee yield, in contrast to the other shifts where trade-offs between woody species diversity and coffee yield occurred.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec15\" class=\"Section2\"\u003e\n \u003ch2\u003e3.5. Classification of Coffee Systems as Forest under EUDR\u003c/h2\u003e\n \u003cp\u003eThe comparison of coffee farms in our study with the Global Forest Cover 2020 map (Bourgoin et al., 2023, 2024) reveals that a significant proportion of coffee farms were classified as forest in 2020 (Table \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e). The Forest Coffee plots were consistently and accurately categorised as forest. However, when analysing the plots inventoried on the coffee farms, 26% of monoculture and 58\u0026ndash;60% of coffee agroforestry plots appeared in the so-called forest. Nevertheless, all coffee farms in the study had been planted for at least five years, with data collection occurring in 2021. This indicates that all farms were non-forest before 2020. Consequently, all plots classified as forest were incorrectly categorised except for the Forest Coffee plots. Even when plots were represented by a single point rather than the 25m x 25m plot, 20\u0026ndash;42% of coffee farms were still erroneously classified as being forest in 2020. Additionally, as hypothesised, the analysis revealed that coffee agroforestry systems were more frequently misclassified as forest than monocultures, with agroforestry plots being twice as likely to be labelled as forest when using the plot method.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\u0026nbsp;\u003ctable id=\"Tab7\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eForest and Non-Forest classification of coffee plots in Tshopo, DRC, based on EUDR criteria, comparing four different coffee systems and two classification methods.\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003ccolgroup cols=\"6\"\u003e\u003c/colgroup\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eMethod\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eClassification\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eMonoculture (n\u0026thinsp;=\u0026thinsp;27)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCultivated Agroforestry (n\u0026thinsp;=\u0026thinsp;15)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eWild Agroforestry (n\u0026thinsp;=\u0026thinsp;12)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eForest Coffee\u003c/p\u003e\n \u003cp\u003e(n\u0026thinsp;=\u0026thinsp;25)\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003ePoint\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eForest [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e22.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e41.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-Forest [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e77.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e80.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e58.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cem\u003ePlot (625m\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eForest [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e25.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e60.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e58.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e100\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-Forest [%]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e74.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e40.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e41.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.00\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n\u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Comparing Robusta Coffee Yields, Considering Farm Management\u003c/h2\u003e \u003cp\u003eRobusta coffee yields in forest coffee systems in our study (\u0026lt;\u0026thinsp;1 kg/ha) are several orders of magnitude lower than those of wild \u003cem\u003eCoffea arabica\u003c/em\u003e in Ethiopian forests (15 kg/ha; Schmitt et al., \u003cspan citationid=\"CR95\" class=\"CitationRef\"\u003e2010\u003c/span\u003e), which is likely the result of lower coffee densities and greater canopy closure in the Congolese tropical rainforests.\u003c/p\u003e \u003cp\u003eThe coffee yields in our study were found to have median yields ranging roughly from 0.50 to 0.80 ton green coffee/ha depending on the coffee system, excluding forest coffee (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), and are significantly lower than those found for monocultures (1.25 ton/ha) and robusta-banana intercropping systems (1.09 ton/ha) in Uganda (van Asten et al., \u003cspan citationid=\"CR108\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Furthermore, coffee yields for both monocultures and agroforestry systems were much lower in our study than in most studies in the meta-analysis (4.1\u0026thinsp;\u0026plusmn;\u0026thinsp;2.88 ton/ha) on the drivers of yield gaps across different types of coffee systems (Mokondoko et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). This large yield gap is likely due to a lack of good agricultural practices, such as using improved coffee varieties, inorganic inputs, timely pruning, pest and disease management and also to a general low soil fertility in our study area (Pers. Obs.).\u003c/p\u003e \u003cp\u003eOn the one hand, coffee yields in our study were significantly higher in monocultures than in wild coffee agroforestry systems on a per hectare basis (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). No significant difference was found in coffee planting density when comparing coffee monocultures with wild coffee agroforestry (p\u0026thinsp;=\u0026thinsp;0.13), likely due to the limited sample size of the wild agroforestry plots (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Thus, the yield difference on a per ha basis mainly comes from the significant yield difference on a per plant basis. Furthermore, the yield differences per plant between monocultures and wild coffee agroforestry likely result from excessive shading (\u0026gt;\u0026thinsp;50%) in most wild agroforestry plots (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eOn the other hand, when comparing coffee monocultures with cultivated coffee agroforestry systems, no significant difference was found in coffee planting densities (p\u0026thinsp;=\u0026thinsp;0.13). This might be due to the limited sample size of the cultivated agroforestry plots. However, the difference in median coffee planting density is not large (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). It can be concluded that, despite a significantly higher median canopy closure and tree planting density in cultivated agroforestry systems compared to monocultures, coffee yields in our study are similar for monocultures and cultivated agroforestry systems, whether on a per plant basis (p\u0026thinsp;=\u0026thinsp;0.757) or per ha (p\u0026thinsp;=\u0026thinsp;0.310) (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe above is supported by a global meta-analysis on the drivers of yield gaps across coffee systems (Mokondoko et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2022\u003c/span\u003e), which found that shade cover and plant densities affected coffee yields more than climatic and biotic factors. Moreover, the meta-analysis found that coffee agroforestry systems with a shade cover of 35\u0026ndash;50 % and tree density of 00\u0026ndash;250 trees/ha, which corresponds to the cultivated coffee agroforestry systems in our study, have similar coffee yields as monocultural systems with a shade cover of less than 30 %. However, agroforesty systems with shade cover of 50\u0026ndash;80 % and a tree density geater than 250 trees/ha, similar to the wild coffee agroforestry systems in our study, were found to result in significantly reduced coffee yields (Mokondoko et al., \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Thus, given the results from the meta-analysis and the finding that the median shade cover for wild coffee agroforestry systems in our study was 52.9\u0026thinsp;\u0026plusmn;\u0026thinsp;3.67 %, this might indicate that the shade cover isabove the optimal range for coffee yield maximisation. Another meta-analysis on the effect of shade on robusta coffee yield and growth (Piato et al., \u003cspan citationid=\"CR85\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) was inconclusive, mainly due to the limited number of studies.\u003c/p\u003e \u003cp\u003eDespite noteworthy higher median yield and smaller median canopy closure values for the cultivated coffee agroforestry system compared to wild coffee agroforestry, no significant differences were found in terms of yield per plant (p\u0026thinsp;=\u0026thinsp;0.111), yield per ha (p\u0026thinsp;=\u0026thinsp;0.137) and shade cover (p\u0026thinsp;=\u0026thinsp;0.120) (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Most likely, this is due to the limited number of sampled plots for each of the two agroforestry systems, which resulted in limited statistical power.\u003c/p\u003e \u003cp\u003eAdditionally, it should be noted that coffee planting densities are likely below optimal in the monoculture system and possibly also in both agroforestry systems, as planting densities of roughly 960 and 890 coffee plants per hectare were found in our study for monocultures and agroforestry, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). By comparison, densities of 1333 plants per ha are recommended by the local research institute (INERA) and a range of 1111\u0026ndash;2222 plants per ha is recommended for robusta coffee in the literature (Niyibigira, \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Wintgens, \u003cspan citationid=\"CR113\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). Nonetheless, our findings align with van Asten et al. (\u003cspan citationid=\"CR108\" class=\"CitationRef\"\u003e2011\u003c/span\u003e) in smallholder farms in Uganda, who observed densities of 1014 and 880 robusta coffee plants per ha in monocultures and agroforestry systems.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e4.2. Contrasting Carbon Stocks between Robusta Coffee Systems\u003c/h2\u003e \u003cp\u003eForest coffee contained roughly three times more TOC than both agroforestry systems, which in turn was shown to contain roughly 60\u0026ndash;70% more TOC than monocultures (Table \u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003eA1\u003c/span\u003e). This trend is consistent with literature that compares monocultures, coffee agroforestry systems, and forest (coffee) systems (De Beenhouwer et al., 2016; Ehrenbergerov\u0026aacute; et al., 2016; Guillemot et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Schmitt-Harsh et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Solis et al., \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Soto-Pinto et al., \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Van Noordwijk et al., \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Vanderhaegen et al., \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eMoreover, our study found that the TOC content in forest coffee systems was 289\u0026thinsp;\u0026plusmn;\u0026thinsp;14 Mg C/ha, which falls roughly in the middle of the TOC range (198\u0026ndash;413 Mg C/ha) reported in the literature (De Beenhouwer et al., 2016; Guillemot et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Schmitt-Harsh et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Van Noordwijk et al., \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Vanderhaegen et al., \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). The average living above-ground carbon stock obtained for forest coffee systems in our study (179\u0026thinsp;\u0026plusmn;\u0026thinsp;11 Mg C/ha) is consistent with results from other studies on forests in the Tshopo province, such as 162\u0026thinsp;\u0026plusmn;\u0026thinsp;20 Mg C/ha in (Kearsley, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2015\u003c/span\u003e) and 185\u0026thinsp;\u0026plusmn;\u0026thinsp;44 Mg C/ha in (P. Moonen, \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). 8\u003c/p\u003e \u003cp\u003eIn contrast, the TOC for both coffee agroforestry systems in our study (92\u0026ndash;100\u0026thinsp;\u0026plusmn;\u0026thinsp;10 Mg C/ha) was lower than the range of 120\u0026ndash;219 Mg C/ha reported in previous studies (De Beenhouwer et al., 2016; Ehrenbergerov\u0026aacute; et al., 2016; Guillemot et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Schmitt-Harsh et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Solis et al., \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Soto-Pinto et al., \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Van Noordwijk et al., \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Vanderhaegen et al., \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). This difference compared to other studies can be attributed to smaller living above-ground carbon stocks, and smaller SOC stocks in our study (Table \u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003eA1\u003c/span\u003e) resulting from smaller soil organic carbon concentrations (Table \u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003eA2\u003c/span\u003e). Moreover, the smaller living above-ground carbon stocks found for coffee agroforestry systems in our study might be attributed to tree densities being at the lower end, respectively 192\u0026thinsp;\u0026plusmn;\u0026thinsp;25 and 240\u0026thinsp;\u0026plusmn;\u0026thinsp;34 trees/ha for Cultivated and Wild agroforestry systems, compared to other studies on arabica and robusta coffee from India, Latin-America, and Ethiopia, which have reported a range of 124\u0026ndash;800 trees/ha (De Beenhouwer et al., 2016b; Ehrenbergerov\u0026aacute; et al., 2016; Guillemot et al., \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Schmitt-Harsh et al., \u003cspan citationid=\"CR96\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; Solis et al., \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Soto-Pinto et al., \u003cspan citationid=\"CR103\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Van Noordwijk et al., \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Vanderhaegen et al., \u003cspan citationid=\"CR110\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor robusta coffee monocultures, the average TOC of 58\u0026thinsp;\u0026plusmn;\u0026thinsp;3 Mg C/ha found in our study lies within the range of 52\u0026ndash;113 Mg C/ha found for coffee monocultures in Peru and Indonesia (Ehrenbergerov\u0026aacute; et al., 2016; Solis et al., \u003cspan citationid=\"CR101\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Van Noordwijk et al., \u003cspan citationid=\"CR111\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). Though within range, our TOC findings are at the lower end for monocultures due to relatively small SOC stocks (Table \u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003eA1\u003c/span\u003e), mainly due to small soil organic carbon concentrations rather than smaller soil bulk densities (Table \u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003eA2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eWhen comparing the AGB of coffee systems (Table \u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003eA1\u003c/span\u003e) with other agro- and ecosystems in the study area, it can be observed that the AGB of both coffee agroforestry systems (31\u0026ndash;39\u0026thinsp;\u0026plusmn;\u0026thinsp;6 Mg C/ha) is similar to the AGB in local cacao agroforests (44\u0026thinsp;\u0026plusmn;\u0026thinsp;25 Mg C/ha) (Batsi et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In addition, a study on the fallow systems (secondary forest regrowth after slash-and-burn) in the Tshopo province found that these fallow systems have an average AGB of 58\u0026thinsp;\u0026plusmn;\u0026thinsp;46 Mg C/ha. However, this varies heavily depending on the number of passed rotation cycles and fallow age (Moonen et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Thus, the AGB in secondary forest regrowth after slash-and-burn can be much smaller, similar or greater than in coffee agroforestry systems, while being almost always much greater than in coffee monocultures (6\u0026thinsp;\u0026plusmn;\u0026thinsp;1 Mg C/ha). Furthermore, it should be noted that the AGB and total organic carbon in fallow systems are decreasing and will continue to decrease over time as the number of rotation cycles is increasing and the fallow period is shortening (Moonen et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). This indicates that, in the long term, the local coffee and cacao agroforestry systems hold greater potential in terms of carbon sequestration than the fallow systems. Moreover, (Moonen et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) noted that from the third and fourth fallow cycles onwards, a shift from fallows (secondary forest regrowth) to more permanent cropping systems, such as coffee and cacao agroforestry systems, frequently results in positive impacts on carbon stocks.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e4.3. (Dis)similarities in Woody Plant Diversity and Community Composition\u003c/h2\u003e \u003cp\u003eForest coffee systems exhibited higher woody biodiversity than both agroforestry systems, with both agroforestry systems exhibiting higher woody biodiversity than monocultures (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The large differences in species richness, diversity, evenness, and community composition indicate that forest coffee systems are characterised by a unique biodiversity that cannot be substituted by coffee agroforestry systems, and especially not by monocultures.\u003c/p\u003e \u003cp\u003eMoreover, most woody species were only found in forest coffee, while some were only found in wild coffee agroforestry, though to a much lesser extent. Thus, coffee agroforestry systems contain significant woody species diversity, which complements the diversity found in forest coffee systems (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). In addition, it should be noted that despite large differences in woody species community composition, several woody species are shared between wild coffee agroforestry and forest coffee systems, in contrast to cultivated coffee agroforestry and forest coffee.\u003c/p\u003e \u003cp\u003eThese findings align with the meta-analysis (De Beenhouwer et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), which concludes that converting natural forests to coffee agroforestry systems has an overall detrimental impact on total species richness, even more so on forest species. Likewise, as management intensifies from wild coffee agroforestry to cultivated coffee agroforestry, native forest species and native undomesticated fruit tree species tend to disappear, with only minimal to no replacement by other commonly cultivated tree species.\u003c/p\u003e \u003cp\u003eAs forest specialists often have specialised habitat requirements, they are more vulnerable to land-use change (Gibson et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Hundera et al., 2013). Thus, woody species restricted to forests are greatly affected by management intensification, emphasising the indispensable role of natural forests in conserving forest species, even in an agroforest matrix, as shown by our results and supported by the meta-analysis of De Beenhouwer et al. (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Broader studies also support these results in the sense of the irreplaceable value of tropical forests for sustaining tropical biodiversity (Gardner et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Gibson et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Mu\u0026ntilde;oz et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Although agroforestry systems fail to conserve the strictest forest species, they can contribute to the conservation of many other species (Mu\u0026ntilde;oz et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Arthropods, mammals and birds were largely unaffected by the conversion of forest into agroforest (Daily et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; De Beenhouwer et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Gibson et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Pardini, \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). These findings suggest that agroforestry systems can offer a partial solution to balance the need for agricultural production and biodiversity conservation.\u003c/p\u003e \u003cp\u003eDe Beenhouwer's meta-analysis (2013) found that intensifying high-diversity agroforestry systems towards sun plantations, containing only a sparse amount of shade trees that belong to one or very few species, led to a greater decrease in total species richness than converting natural forests into high-diversity agroforestry systems. This result can be attributed to the diverse and structurally complex canopy layer in certain diverse coffee agroforestry systems, which maintain a high level of biodiversity (Perfecto \u0026amp; Vandermeer, \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Schroth \u0026amp; Harvey, \u003cspan citationid=\"CR97\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Our study observed the most significant decline in woody species richness (S\u003csub\u003eobs\u003c/sub\u003e) occurring from forest coffee (29.57\u0026thinsp;\u0026plusmn;\u0026thinsp;0.52) to cultivated coffee agroforestry (7.92\u0026thinsp;\u0026plusmn;\u0026thinsp;0.54) and wild coffee agroforestry (8.17\u0026thinsp;\u0026plusmn;\u0026thinsp;0.67), as opposed to, from coffee agroforestry to coffee monocultures.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e4.4. Clarifying Synergies and Trade-offs between Robusta Yields, Woody Biodiversity and Carbon Stocks\u003c/h2\u003e \u003cp\u003eThis study identified trade-offs between woody plant diversity and coffee yields, as well as between carbon stocks and coffee yield, alongside co-benefits between woody plant diversity and carbon stocks.\u003c/p\u003e \u003cp\u003eFirstly, our findings regarding the relationship between woody biodiversity and coffee yield show a strong negative relationship (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e6\u003c/span\u003e), which aligns with previous research (De Beenhouwer et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Jha et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). However, contrary to other studies (De Beenhouwer et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Perfecto et al., \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e2003\u003c/span\u003e), we observed a convex biodiversity-yield relationship, indicating a sharper decline in biodiversity during the shift from forest to coffee agroforestry and a less pronounced decline from agroforestry to monocultures. While in the case of a concave biodiversity-yield relationship, wildlife-friendly farming may be the best conservation option (Clough et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; De Beenhouwer et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Perfecto et al., \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2005\u003c/span\u003e), in our study context, the convex nature of the relationship indicates that wildlife-friendly farming alone does not suffice and stresses the importance of forest coffee conservation. Furthermore, when further intensifying management from cultivated agroforestry systems to monocultures, a loss of woody biodiversity was observed, without a gain in coffee yield, rather than a trade-off. Thus, as the much more biodiverse farming system of the two, cultivated coffee agroforestry may serve as the preferred management strategy for balancing biodiversity and coffee yield compared to coffee monocultures.\u003c/p\u003e \u003cp\u003eSecondly, the convex TOC-yield relationship found in our study (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e6\u003c/span\u003e), with a sharp decline in TOC from forest coffee to both agroforestry systems and a relatively smaller decline toward monocultures, aligns with the provisioning yield function suggested by De Beenhouwer et al. (\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). This again highlights the enormous value of forest coffee in terms of carbon sequestration compared to other coffee systems. However, given farmers\u0026rsquo; need to balance production with carbon sequestration, other coffee systems will exist and will replace a part of the forest coffee system. Thus, it is important to note that, when intensifying management from cultivated agroforestry systems to monocultures, a loss of woody biodiversity was observed, without a gain in coffee yield, rather than a trade-off, which occurs when shifting between the other systems. Thus, as the farming system which contains much more carbon, cultivated coffee agroforestry is the preferred management strategy for balancing carbon sequestration and coffee yield compared to coffee monocultures.\u003c/p\u003e \u003cp\u003eThirdly, in contrast to the negative relationships found in our study between yield and biodiversity as well as between yield and TOC mentioned above, a significant positive relationship was found between the woody biodiversity indices and carbon stocks (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). The latter finding supports previous studies (De Beenhouwer et al., 2016; Jha et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Richards \u0026amp; M\u0026eacute;ndez, \u003cspan citationid=\"CR90\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Sari et al., 2020; Zewdie et al., \u003cspan citationid=\"CR117\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). These results indicate co-benefits between woody biodiversity and carbon stocks (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Consequently, both objectives can be pursued simultaneously, highlighting the potential for synergies between biodiversity conservation and climate change mitigation.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e4.5. Implications of EUDR Forest Classification for Coffee Systems\u003c/h2\u003e \u003cp\u003eThe above-mentioned misclassification of coffee gardens as forest likely arises from the resemblance of agroforestry systems to forests in satellite imagery due to agroforestry's high tree canopy cover. Moreover, the cut-off value used by the EUDR is greater than 10% canopy cover (Bourgoin et al., 2024; European Parliament and Council of the European Union, 2023), while canopy closure found in our study for coffee agroforestry systems was 38\u0026thinsp;\u0026plusmn;\u0026thinsp;4 % and 53\u0026thinsp;\u0026plusmn;\u0026thinsp;4 % depending on the type of agroforestry (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Consequently, the EDR classifiction approach seems to disproportionately affect coffee systems with greater canopy cover. This will likely penalise sustainable practices like agroforestry, which have larger carbon stocks and woody diversity than monocultures (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This clearly contrasts with the environmental goals of the EUDR, which is \u0026ldquo;to bring down greenhouse gas emissions and biodiversity loss\u0026rdquo; (European Parliament and Council of the European Union, 2023). Additionally, since monocultures were often misclassified, while their canopy cover ranged from 0\u0026ndash;5 %, it indicates that too much faith is given to these remotely sensed products.\u003c/p\u003e \u003cp\u003eAnother important conceptual deficiency of the EUDR is that it confuses land cover (as observed with remote sensing) with land use, which focuses more on land management and tenure. It has the perverse effect of leaving farmers who previously deforested their land unaffected, but it risks severely punishing farmers who apply more sustainable agroforestry practices because their farm resembles a forest too much. Thus, The EUDR is a very blunt instrument that risks disproportionally affecting smallholder farmers, many of whom have introduced sustainable agricultural practices.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusion","content":"\u003cp\u003eThis study provides valuable insights into the synergies and trade-offs between robusta coffee yield, carbon stocks, and woody species biodiversity across coffee systems in the Tshopo province, DR Congo. The observed co-benefits between carbon stocks and biodiversity highlight the potential for synergies in coffee systems between biodiversity conservation and climate change mitigation.\u003c/p\u003e \u003cp\u003eTrade-offs between robusta coffee yields on the one hand and woody biodiversity and carbon stocks on the other hand were found. On the one hand, the sharp decline in woody biodiversity and carbon stocks when intensifying forest coffee systems towards coffee agroforestry or coffee monocultures implies that forest coffee systems have an irreplaceable value. While significant trade-offs between woody biodiversity and carbon stocks, and coffee yield occur when intensifying wild coffee agroforestry to monocultures, this was not the case for cultivated agroforestry. Thus, from a perspective of balancing needs for production, biodiversity and carbon sequestration, cultivated coffee agroforestry systems can be recommended above coffee monocultures and wild coffee agroforestry. Even more so, coffee agroforestry provides additional advantages in terms of food diversification, income diversification and risk management. The profitability of the cultivated agroforestry system is likely greater than that of the low-input coffee monocultures in our study. However, additional research comparing the economic aspects of the different coffee systems would be welcome. Besides, further research on farmers\u0026rsquo; knowledge of and preferences for (shade) tree species in agroforestry is recommended and will provide a better understanding of farmers\u0026rsquo; choice between different coffee agroforestry systems.\u003c/p\u003e \u003cp\u003eThe comparison of coffee farms with the Global Forest Cover 2020 map used for EUDR revealed substantial misclassification, particularly for agroforestry systems, due to their high canopy cover resembling natural forests. This misclassification will likely penalise sustainable practices like agroforestry and reach the opposite of the EUDR\u0026rsquo;s environmental goals. Not only is a greater tree canopy cover threshold needed to differentiate between coffee systems and forest, but more attention should also be given to ownership and land tenure ship rights.\u003c/p\u003e \u003cp\u003eIn addition, our results indicate that excessive shading (\u0026gt;\u0026thinsp;50%) in many wild agroforestry plots severely affects coffee yields. Thus, identifying the optimal shade level for robusta cultivation would assist in minimising trade-offs and maximising synergies in coffee agroforestry systems. In line with this, optimal robusta and (shade) tree density combinations are poorly understood. However, given the too-low coffee planting densities currently used by most farmers in the Tshopo province, we recommend increasing their coffee planting density, especially in monocultures, but possibly also in coffee agroforestry.\u003c/p\u003e \u003cp\u003eFurthermore, the coffee production systems in the Tshopo province should, in the first instance, be optimised by increasing coffee planting densities, pruning the principal stems to obtain an optimal number of stems per plant, seasonal pruning of the branches, and pruning the principal stems to ensure rejuvenation. Once these foundations have been laid, one could consider using fertilisers to address the low soil fertility in the region. Also, management of coffee pests and diseases, which is currently almost completely lacking, could greatly improve coffee yields, with a focus on the Coffee Berry Borer (\u003cem\u003eHypothenemus hampei\u003c/em\u003e) and wood-boring beetles, namely the Coffee Twig Borer (\u003cem\u003eXylosandrus compactus\u003c/em\u003e Eichhoff 1875) and the West African Coffee Stem Borer (\u003cem\u003eBixadus sierricola\u003c/em\u003e White 1858).\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eConceptualisation: IB, RM, OH and BV; Methodology: IB, BV, OH, RM and JD; Data collection: IB, TKM and JD; Lab work: IB; Data analysis: IB; Software; IB and JD; Writing \u0026ndash; Original draft: IB; Visualisation: IB and JD; Writing - Review and editing: IB, TKM, JD, RM, OH and BV; Supervision: RM, OH and BV; Funding acquisition: RM and BV.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe would like to express our gratitude to Justin Angunizu Asimonyio (CSB) and Jean-L\u0026eacute;on Kambale (CSB) for their assistance with the botanical identification of the woody species. Besides, we would like to thank Lore Fondu (KU Leuven) for her assistance with the analysis of soil samples.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eData used for figures and tables in this study have been deposited online via Zenodo: 10.5281/zenodo.13850422\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbdullah, N., Sulaiman, F., \u0026amp; Taib, R. M. (2013). Characterisation of banana (Musa spp.) plantation wastes as a potential renewable energy source. In \u003cem\u003eAIP Conference Proceedings\u003c/em\u003e (Vol. 1528, No. 1, pp. 325\u0026ndash;330). American Institute of Physics.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAerts, R., Hundera, K., Berecha, G., Gijbels, P., Baeten, M., Van Mechelen, M., Hermy, M., Muys, B., \u0026amp; Honnay, O. (2011). Semi-Forest Coffee cultivation and the conservation of Ethiopian Afromontane rainforest fragments. Forest Ecology and Management, 261, 1034\u0026ndash;1041.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlcudia-Aguilar, A., Mart\u0026iacute;nez-Zurimendi, P., van der Wal, H., Castillo-Uzcanga, M. M., \u0026amp; Su\u0026aacute;rez-S\u0026aacute;nchez, J. (2019). Allometric estimation of the biomass of Musa spp. in homegardens of Tabasco, Mexico. Tropical and Subtropical Agroecosystems, 22(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAlvarez Alvarez, E. A., Almaz\u0026aacute;n N\u0026uacute;\u0026ntilde;ez, R. C., Gonz\u0026aacute;lez Garc\u0026iacute;a, F., Brito Mill\u0026aacute;n, M., M\u0026eacute;ndez Bahena, A., \u0026amp; Garc\u0026iacute;a Ib\u0026aacute;\u0026ntilde;ez, S. (2020). Shade coffee plantations maintain woody plant diversity and structure in a cloud forest landscape of southern Mexico. J. For. Res..\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAndrade, H. J., Segura, M. A., \u0026amp; Feria, M. (2021). Allometric models for estimating belowground biomass of individual coffee bushes growing in monoculture and agroforestry systems. Agroforestry Systems, 95, 215\u0026ndash;226.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArbizu, P. M. (2017). \u003cem\u003epairwiseAdonis: Pairwise Multilevel comparison using Adonis\u003c/em\u003e. R package version\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBatsi, G., Sonwa, D. J., Mangaza, L., Ebuy, J., \u0026amp; Kahindo, J. M. (2021). Preliminary estimation of above-ground carbon storage in cocoa agroforests of Bengamisa-Yangambi forest landscape (Democratic Republic of Congo). AF Systems, \u003cem\u003e95\u003c/em\u003e(8), 1505\u0026ndash;1517.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeer, J. (1987). Advantages, disadvantages and desirable characteristics of shade trees for coffee, cacao and tea. AF Systems, \u003cem\u003e5\u003c/em\u003e(3), 3\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBranthomme, A., Merle, C., Kindgard, A., Louren\u0026ccedil;o, A., Ng, W.-T., D\u0026rsquo;Annunzio, R., \u0026amp; Shapiro, A. (2023). How much do large-scale and small-scale farming contribute to global deforestation? Results from a remote sensing pilot approach. Rome: FAO. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4060/cc5723en\u003c/span\u003e\u003cspan address=\"10.4060/cc5723en\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrown, S. (1997). Estimating biomass and biomass change of tropical forests: A primer (Vol. 134). \u003cem\u003eFood \u0026amp; Agriculture Org\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCardinale, B. J., Duffy, J. E., Gonzalez, A., Hooper, D. U., Perrings, C., Venail, P., \u0026hellip; Naeem, S. (2012). Biodiversity loss and its impact on humanity. Nature, 486, 59\u0026ndash;67.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarsan, S., Orwa, C., Harwood, C., Kindt, R., Stroebel, A., Neufeldt, H., \u0026amp; Jamnadass, R. (2012). \u003cem\u003eAfrican wood density database\u003c/em\u003e. World AF Centre, Nairobi.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCastle, S. E., Miller, D. C., Ordonez, P. J., Baylis, K., \u0026amp; Hughes, K. (2021). The impacts of AF interventions on agricultural productivity, ecosystem services, and human well-being in low‐and middle‐income countries: A systematic review. Campbell Systematic Reviews, 17(2), e1167.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChao, A., Chazdon, R. L., Colwell, R. K., \u0026amp; Shen, T. J. (2005). A new statistical approach for assessing compositional similarity based on incidence and abundance data. Ecology Letters, 8, 148\u0026ndash;159.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChave, J., Andalo, C., Brown, S., Cairns, M. A., Chambers, J. Q., Eamus, D., \u0026hellip; Yamakura, T. (2005). Tree allometry and improved estimation of carbon stocks and balance in tropical forests. Oecologia, 145, 87\u0026ndash;99.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClough, Y., Barkmann, J., Juhrbandt, J., Kessler, M., Wanger, T. C., Anshary, A., \u0026hellip; Tscharntke, T. (2011). Combining high biodiversity with high yields in tropical agroforests. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, 108(20), 8311\u0026ndash;8316.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCramer, P. J. S., \u0026amp; Wellman, F. L. (1957). Review of literature of coffee research in Indonesia.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDaily, G. C., Ceballos, G., Pacheco, J., Suz\u0026aacute;n, G., \u0026amp; S\u0026aacute;nchez-Azofeifa, A. R. (2003). Countryside biogeography of neotropical mammals: conservation opportunities in agricultural landscapes of Costa Rica. Conservation biology, 17(6), 1814\u0026ndash;1826.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavis, A. P., D. Mieulet, J. Moat, D. Sarmu, and J. Haggar. (2021). Arabica-like Flavour in a Heat-Tolerant Wild Coffee Species. Nature Plants 7 (4): 413\u0026ndash;18. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41477-021-00891-4\u003c/span\u003e\u003cspan address=\"10.1038/s41477-021-00891-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavis, A. P., Gole, T. W., Baena, S., \u0026amp; Moat, J. (2012). The impact of climate change on indigenous arabica coffee (Coffea arabica): predicting future trends and identifying priorities. PloS one, \u003cem\u003e7\u003c/em\u003e(11), e47981.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDavis, H., Rice, R., Rockwood, L., Wood, T., \u0026amp; Marra, P. (2019). The economic potential of fruit trees as shade in blue mountain coffee agroecosystems of the Yallahs River watershed, Jamaica WI. AF Systems, \u003cem\u003e93\u003c/em\u003e, 581\u0026ndash;589.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Beenhouwer, M., Aerts, R., \u0026amp; Honnay, O. (2013). A global meta-analysis of the biodiversity and ecosystem service benefits of coffee and cacao AF. Agriculture, Ecosystems \u0026amp; Environment, 175, 1\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Beenhouwer, M., Muleta, D., Peeters, B., Van Geel, M., Lievens, B., \u0026amp; Honnay, O. (2015a). DNA pyrosequencing evidence for large diversity differences between natural and managed coffee mycorrhizal fungal communities. Agronomy for Sustainable Development, 35, 241\u0026ndash;249.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Beenhouwer, M., Van Geel, M., Ceulemans, T., Muleta, D., Lievens, B., \u0026amp; Honnay, O. (2015b). Agricultural intensification alters the arbuscular mycorrhizal fungi communities of Arabica coffee (Coffea arabica) in its Ethiopian centre of origin through changing soil characteristics. Soil Biology and Biochemistry, 91, 133\u0026ndash;139.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe C\u0026aacute;ceres, M., and Legendre, P. (2009). Associations between species and groups of sites: indices and statistical inference. Ecology, 90, 3566\u0026ndash;3574.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDepecker, J., Asimonyio, J. A., Miteho, R., Hatangi, Y., Kambale, J. L., Verleysen, L., Stoffelen, P., Janssens, S. B., Dhed\u0026rsquo;a, B., Vandelook, F., \u0026amp; Honnay, O. (2022). The association between rainforest disturbance and recovery, tree community composition, and community traits in the Yangambi area in the Democratic Republic of the Congo. Journal of Tropical Ecology, pp.1\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDufrene, M., and Legendre, P. (1997). Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecological Monographs, 67, 345\u0026ndash;366.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDurand-Bessart, C., Tixier, P., Quinteros, A., Andreotti, F., Rapidel, B., Tauvel, C., \u0026amp; Allinne, C. (2020). Analysis of interactions amongst shade trees, coffee foliar diseases and coffee yield in multistrata AF systems. Crop Protection, \u003cem\u003e133\u003c/em\u003e, 105137.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEhrenbergerova, L., Cienciala, E., Kučera, A., Guy, L., \u0026amp; Habrov\u0026aacute;, H. (2016). Carbon stock in agroforestry coffee plantations with different shade trees in Villa Rica, Peru. Agroforestry systems, \u003cem\u003e90\u003c/em\u003e, 433\u0026ndash;445.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEnglund, S. R., O'Brien, J. J., \u0026amp; Clark, D. B. (2000). Evaluation of digital and film hemispherical photography and spherical densiometry for measuring forest light environments. Canadian Journal of Forest Research, 30(12), 1999\u0026ndash;2005.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFAO and UNEP. (2020). \u003cem\u003eThe State of the World\u0026rsquo;s Forests 2020. Forests, biodiversity and people\u003c/em\u003e. Rome.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFAO (2020). Global Forest Resources Assessment 2020 (FRA 2020). Food and Agriculture Organization of the United Nations, Rome. URL: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4060/ca9825en\u003c/span\u003e\u003cspan address=\"10.4060/ca9825en\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGardner, T. A., Barlow, J., Chazdon, R., Ewers, R. M., Harvey, C. A., Peres, C. A., \u0026amp; Sodhi, N. S. (2009). Prospects for tropical forest biodiversity in a human-modified world. Ecology letters, 12(6), 561\u0026ndash;582.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGibson, L., Lee, T. M., Koh, L. P., Brook, B. W., Gardner, T. A., Barlow, J., \u0026hellip; Sodhi, N. S. (2011). Primary forests are irreplaceable for sustaining tropical biodiversity. Nature, 478(7369), 378\u0026ndash;381.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlobal Forest Watch. (2023). Democratic Republic of the Congo (COD) Forest Data. In Global Forest Watch. Retrieved June 28, 2023, from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.globalforestwatch.org/dashboards/country/COD\u003c/span\u003e\u003cspan address=\"https://www.globalforestwatch.org/dashboards/country/COD\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGłowacki, Ł. (2011). Accuracy of species richness estimators applied to fish in small and large temperate lowland rivers. Biodiversity and Conservation, 20, 1365\u0026ndash;1384.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGordon, C., Manson, R., Sundberg, J., \u0026amp; Cruz-Angon, A. (2006). Biodiversity, profitability, and vegetation structure in a Mexican coffee agroecosystem. Agriculture, Ecosystems and Environment, \u003cem\u003e118\u003c/em\u003e, 256\u0026ndash;266.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGotelli, N. J., \u0026amp; Chao, A. (2013). \u003cem\u003eMeasuring and estimating species richness, species diversity, and biotic similarity from sampling data.\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGreig-Smith, P. (1983). \u003cem\u003eQuantitative Plant Ecology\u003c/em\u003e. Blackwell Scientific Publications.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGuillemot, J., Le Maire, G., Munishamappa, M., Charbonnier, F., \u0026amp; Vaast, P. (2018). Native coffee AF in the Western Ghats of India maintains higher carbon storage and tree diversity compared to exotic AF. Agriculture, Ecosystems \u0026amp; Environment, \u003cem\u003e265\u003c/em\u003e, 461\u0026ndash;469.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eH\u0026auml;ger, A. (2012). The effects of management and plant diversity on carbon storage in coffee AF systems in Costa Rica. AF Systems, 86(2), 159\u0026ndash;174.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHairiah, K., Dewi, S., Agus, F., Velarde, S., Ekadinata, A., Rahayu, S., Van Noordwijk, M. (2011\u003cem\u003e). Measuring Carbon Stocks Across Land Use Systems: A Manual\u003c/em\u003e. World AF Centre (ICRAF), SEA Regional Office, Bogor Indonesia.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarja, D., Rahayu, S., and Pambudi, S. (2019). \u003cem\u003eTree functional attributes and ecological database\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarris, N. L., Gibbs, D. A., Baccini, A., Birdsey, R. A., de Bruin, S., Farina, M., Fatoyinbo, L., Hansen, M. C., Herold, M., Houghton, R. A., Potapov, P. V., Suarez, D. R., Roman-Cuesta, R. M., Saatchi, S. S., Slay, C. M., Turubanova, S. A., \u0026amp; Tyukavina, A. (2021). Global maps of twenty-first century forest carbon fluxes. Nature Climate Change, \u003cem\u003e11\u003c/em\u003e, 234\u0026ndash;240.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHill, M.O. (1973). Diversity and evenness: a unifying notation and its consequences. Ecology, 54, 427\u0026ndash;432.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHundera, K., Aerts, R., Fontaine, A., Van Mechelen, M., Gijbels, P., Honnay, O., and Muys, B. (2012). Effects of wild coffee management intensity on composition, structure, and regeneration of Ethiopian montane rainforests. Environmental Management, 51, 801\u0026ndash;809.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eICRAF (2007). \u003cem\u003eWood Density Database\u003c/em\u003e. World AF Center: Wood Density Database. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.worldAF.org/sea/Products/AFDbases/WD/Index.htm\u003c/span\u003e\u003cspan address=\"http://www.worldAF.org/sea/Products/AFDbases/WD/Index.htm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (accessed on 2.15.22).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eICO, CFC, and WB (2000) \u003cem\u003eStudy of Marketing and Trading Policies and Systems in Selected Coffee Producing Countries - Democratic Republic of the Congo\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eICO (2023) Historical Data on the Global Coffee Trade - DRC Robusta Coffee Production. \u003cem\u003eStatistical Publication of the International Coffee Organization\u003c/em\u003e. 2023. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://icocoffee.org/resources/historical-data-on-the-global-coffee-trade/\u003c/span\u003e\u003cspan address=\"https://icocoffee.org/resources/historical-data-on-the-global-coffee-trade/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIdol, T.W., and Youkhana, A.H. (2020). A rapid visual estimation of fruits per lateral to predict coffee yield in Hawaii. AF Systems, 94(1), 81\u0026ndash;93.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eInsideWood database, 2021. \u003cem\u003eInsideWood\u003c/em\u003e NCSU. Published on the Internet. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://insidewood.lib.ncsu.edu/search\u003c/span\u003e\u003cspan address=\"http://insidewood.lib.ncsu.edu/search\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (accessed on 2.15.21).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIPCC (2006). \u003cem\u003eIPCC Guidelines for National Greenhouse Gas Inventories, Forestry\u003c/em\u003e. The Institute for Global Environmental Strategies (IGES), Japan.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJayathilake, H. M., Prescott, G. W., Carrasco, L. R., Rao, M., \u0026amp; Symes, W. S. (2021). Drivers of deforestation and degradation for 28 tropical conservation landscapes. Ambio, \u003cem\u003e50\u003c/em\u003e, 215\u0026ndash;228.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJha, S., Bacon, C. M., Philpott, S. M., Ernesto M\u0026eacute;ndez, V., L\u0026auml;derach, P., \u0026amp; Rice, R. A. (2014). Shade coffee: update on a disappearing refuge for biodiversity. BioScience, 64(5), 416\u0026ndash;428.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJones, A., Breuning-Madsen, H., Brossard, M., Dampha, A., Deckers, J., Dewitte, O., \u0026hellip; Zougmor\u0026eacute;, R. B. (2013). \u003cem\u003eSoil atlas of Africa\u003c/em\u003e. European Commission.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKearsley, E., De Haulleville, T., Hufkens, K., Kidimbu, A., Toirambe, B., Baert, G., Huygens, D., Kebede, Y., Defourny, P., Bogaert, J., and Beeckman, H. (2013). Conventional tree height\u0026ndash;diameter relationships significantly overestimate above-ground carbon stocks in the Central Congo Basin. Nature Communications, 4(1), 1\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKearsley, E. (2015). \u003cem\u003eCarbon storage and functional diversity of tropical rainforest in the Congo Basin\u003c/em\u003e (Doctoral dissertation, Ghent University).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKindt, R., \u0026amp; Coe, R. (2005). \u003cem\u003eTree Diversity Analysis: A Manual and Software for Common Statistical Methods for Ecological and Biodiversity Studies\u003c/em\u003e. Nairobi, Kenya: World AF Centre (ICRAF).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKoutouleas, A., Sarzynski, T., Bertrand, B., Bordeaux, M., Bosselmann, A. S., Campa, C., \u0026hellip; R\u0026aelig;bild, A. (2022). Shade effects on yield across different Coffea arabica cultivars\u0026mdash;how much is too much? A meta-analysis. Agronomy for Sustainable Development, 42(4), 55.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLashermes, P., Combes, M. C., Ribas, A., Cenci, A., Mah\u0026eacute;, L., \u0026amp; Etienne, H. (2010). Genetic and physical mapping of the SH3 region that confers resistance to leaf rust in coffee tree (Coffea arabica L.). Tree Genetics \u0026amp; Genomes, \u003cem\u003e6\u003c/em\u003e(6), 973\u0026ndash;980.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeakey, R. R., Tchoundjeu, Z., Schreckenberg, K., Shackleton, S. E., \u0026amp; Shackleton, C. M. (2005). AF tree products (AFTPs): targeting poverty reduction and enhanced livelihoods. International Journal of Agricultural Sustainability, \u003cem\u003e3\u003c/em\u003e(1), 1\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLemmon, P. E. (1956). A spherical densiometer for estimating forest overstory density. Forest science, 2(4), 314\u0026ndash;320.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eL\u0026oacute;pez-G\u0026oacute;mez, A. M., Williams-Linera, G., \u0026amp; Manson, R. H. (2007). Tree species diversity and vegetation structure in shade coffee farms in Veracruz, Mexico. Agriculture, Ecosystems and Environment, \u003cem\u003e124\u003c/em\u003e, 160\u0026ndash;172.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalhi, Y., Adu-Bredu, S., Asare, R. A., Lewis, S. L., \u0026amp; Mayaux, P. (2013). African rainforests: past, present and future. Philosophical Transactions of the Royal Society B: Biological Sciences, \u003cem\u003e368\u003c/em\u003e, 20120312.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMigolet, P., Go\u0026iuml;ta, K., Ngomanda, A., \u0026amp; Mekui Biyogo, A. P. (2020). Estimation of above-ground oil palm biomass in a mature plantation in the Congo Basin. Forests, 11(5), 544.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMiller, D. C., Ordo\u0026ntilde;ez, P. J., Brown, S. E., Forrest, S., Nava, N. J., Hughes, K., \u0026amp; Baylis, K. (2020). The impacts of AF on agricultural productivity, ecosystem services, and human well-being in low‐and middle‐income countries: An evidence and gap map. \u003cem\u003eCampbell Systematic Reviews\u003c/em\u003e, 16(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoguel, P., \u0026amp; Toledo, V. (1999). Biodiversity conservation in traditional coffee systems of Mexico. Conservation Biology, \u003cem\u003e13\u003c/em\u003e, 11\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMokondoko, P., Avila-Foucat, V. S., \u0026amp; Galeana-Piza\u0026ntilde;a, J. M. (2022). Biophysical drivers of yield gaps and ecosystem services across different coffee-based AF management types: A global meta-analysis. Agriculture, Ecosystems \u0026amp; Environment, \u003cem\u003e337\u003c/em\u003e, 108024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoonen, P. (2017). \u003cem\u003eIntegrated assessment of options to reduce emissions from deforestation in Tshopo, DR Congo. KU Leuven\u003c/em\u003e, Leuven, Belgium.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMoonen, P. C., Verbist, B., Bosela, F. B., Norgrove, L., Dondeyne, S., Van Meerbeek, K., \u0026hellip; \u0026amp; Muys, B. (2019). Disentangling how management affects biomass stock and productivity of tropical secondary forests fallows. Science of the Total Environment, 659, 101\u0026ndash;114.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMu\u0026ntilde;oz, J. C., Aerts, R., Thijs, K. W., Stevenson, P. R., Muys, B., \u0026amp; Sekercioglu, C. H. (2013). Contribution of woody habitat islands to the conservation of birds and their potential ecosystem services in an extensive Colombian rangeland. Agriculture, ecosystems and environment, 173, 13\u0026ndash;19.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuschler, R. G. (2001). Shade improves coffee quality in a sub-optimal coffee-zone of Costa Rica. Agroforestry systems, \u003cem\u003e51\u003c/em\u003e, 131\u0026ndash;139.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNair, P. R., Kumar, B. M., \u0026amp; Nair, V. D. (2021). \u003cem\u003eAn Introduction to AF: Four Decades of Scientific Developments\u003c/em\u003e. Cham: Springer.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNiyibigira, E. I. (2019). \u003cem\u003eRobusta Coffee Handbook: A Sustainable Coffee Industry with High Stakeholder Value for Social Economic Transformation\u003c/em\u003e. The Ministry Of Agriculture, Animal Industry \u0026amp; Fisheries, Uganda.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOksanen, A. J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., \u0026hellip; Wagner, H. (2020). \u003cem\u003ePackage 'vegan.'\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePardini, R. (2004). Effects of forest fragmentation on small mammals in an Atlantic Forest landscape. Biodiversity \u0026amp; Conservation, 13, 2567\u0026ndash;2586.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerfecto, I., Rice, R. A., Greenberg, R., \u0026amp; Van der Voort, M. E. (1996). Shade coffee: a disappearing refuge for biodiversity. BioScience, \u003cem\u003e46\u003c/em\u003e, 598\u0026ndash;608.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerfecto, I., Mas, A., Dietsch, T., \u0026amp; Vandermeer, J. (2003). Conservation of biodiversity in coffee agroecosystems: a tri-taxa comparison in southern Mexico. Biodiversity \u0026amp; Conservation, 12, 1239\u0026ndash;1252.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerfecto, I., Vandermeer, J., Mas, A., \u0026amp; Pinto, L. S. (2005). Biodiversity, yield, and shade coffee certification. Ecological Economics, 54(4), 435\u0026ndash;446.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerfecto, I., Armbrecht, I., Philpott, S. M., Soto-Pinto, L., \u0026amp; Dietsch, T. V. (2006). Shaded coffee and the stability of rainforest margins in northern Latin America. In T. Tscharntke, C. Leuschner, M. Zeller, E. Guhardja, \u0026amp; A. Bidin (Eds.), \u003cem\u003eThe stability of tropical rainforest margins, linking ecological, economic and social constraints of land use and conservation\u003c/em\u003e (pp. 227\u0026ndash;263). Springer Verlag Berlin.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerfecto, I., \u0026amp; Vandermeer, J. (2008). Biodiversity conservation in tropical agroecosystems: a new conservation paradigm. Annals of the New York Academy of Sciences, 1134(1), 173\u0026ndash;200.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerfecto, I., Vandermeer, J., \u0026amp; Philpott, S. M. (2014). Complex ecological interactions in the coffee agroecosystem. \u003cem\u003eAnnual Review of Ecology, Evolution, and Systematics\u003c/em\u003e, 45, 137\u0026ndash;158.\u0026micro;\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePhilpott, S. M., Bichier, P., \u0026amp; Rice, R. A. (2008). Biodiversity conservation, yield, and alternative products in coffee agroecosystems in Sumatra, Indonesia. Biodiversity and Conservation, \u003cem\u003e17\u003c/em\u003e, 1805\u0026ndash;1820.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePhilpott, S. M., \u0026amp; Bichier, P. (2011). Effects of shade tree removal on birds in coffee agroecosystems in Chiapas, Mexico. Agriculture, Ecosystems and Environment, \u003cem\u003e149\u003c/em\u003e, 171\u0026ndash;180.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePiato, K., Lefort, F., Sub\u0026iacute;a, C., Caicedo, C., Calder\u0026oacute;n, D., Pico, J., \u0026amp; Norgrove, L. (2020). Effects of shade trees on robusta coffee growth, yield and quality: A meta-analysis. Agronomy for Sustainable Development, 40(6), 1\u0026ndash;13.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePielou, E. C. (1969). \u003cem\u003eAn Introduction to Mathematical Ecology\u003c/em\u003e. New York: Wiley.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePROTA. (2012). Protabase [WWW Document]. \u003cem\u003eProta4U Web Database\u003c/em\u003e. Retrieved from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.prota4u.org/\u003c/span\u003e\u003cspan address=\"http://www.prota4u.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (Accessed on February 15, 2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRavindranath, N. H., \u0026amp; Ostwald, M. (2008). \u003cem\u003eCarbon Inventory Methods: Handbook for Greenhouse Gas Inventory, Carbon Mitigation and Roundwood Production\u003c/em\u003e. Springer Science\u0026thinsp;+\u0026thinsp;Business Media B.V.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR Core Team (2020). \u003cem\u003eR: A language and environment for statistical computing\u003c/em\u003e. Vienna, Austria: R Foundation for Statistical Computing.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRichards, M. B., \u0026amp; M\u0026eacute;ndez, V. E. (2014). Interactions between carbon sequestration and shade tree diversity in a smallholder coffee cooperative in El Salvador. Conservation Biology, 28(2), 489\u0026ndash;497.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRicotta, C., Podani, J., \u0026amp; Uv, E. D. (2017). On some properties of the Bray-Curtis dissimilarity and their ecological meaning. Ecological Complexity, 31, 201\u0026ndash;205.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRice, R. A. (2008). Agricultural intensification within AF: the case of coffee and wood products. Agriculture, ecosystems \u0026amp; environment, \u003cem\u003e128\u003c/em\u003e(4), 212\u0026ndash;218.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRice, R. A. (2011). Fruits from shade trees in coffee: how important are they? AF systems, \u003cem\u003e83\u003c/em\u003e, 41\u0026ndash;49.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRoyal Museum for Central Africa (RMCA) (2021). Data derived via DRYAD.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmitt, C. B., Senbeta, F., Denich, M., Preisinger, H., \u0026amp; Boehmer, H. J. (2010). Wild coffee management and plant diversity in the montane rainforest of southwestern Ethiopia. African Journal of Ecology, 48(1), 78\u0026ndash;86.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchmitt-Harsh, M., Evans, T. P., Castellanos, E., \u0026amp; Randolph, J. C. (2012). Carbon stocks in coffee agroforests and mixed dry tropical forests in the western highlands of Guatemala. AF Systems, 86, 141\u0026ndash;157.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchroth, G., \u0026amp; Harvey, C. A. (2007). Biodiversity conservation in cocoa production landscapes: an overview. Biodiversity and Conservation, 16, 2237\u0026ndash;2244.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSegura, M., Kanninen, M., \u0026amp; Su\u0026aacute;rez, D. (2006). Allometric models for estimating above-ground biomass of shade trees and coffee bushes grown together. AF Systems, 68(2), 143\u0026ndash;150.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShannon, C. E., \u0026amp; Weaver, W. (1949). \u003cem\u003eThe Mathematical Theory of Communication\u003c/em\u003e. Champaign, IL: University of Illinois Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSilva, M. D. C., V\u0026aacute;rzea, V., Guerra-Guimar\u0026atilde;es, L., Azinheira, H. G., Fernandez, D., Petitot, A. S., \u0026hellip; Nicole, M. (2006). Coffee resistance to the main diseases: leaf rust and coffee berry disease. Brazilian journal of plant physiology, \u003cem\u003e18\u003c/em\u003e, 119\u0026ndash;147.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSolis, R., Vallejos-Torres, G., Ar\u0026eacute;valo, L., Mar\u0026iacute;n-D\u0026iacute;az, J., \u0026Ntilde;ique-Alvarez, M., Engedal, T., \u0026amp; Bruun, T. B. (2020). Carbon stocks and the use of shade trees in different coffee growing systems in the Peruvian Amazon. The Journal of Agricultural Science, \u003cem\u003e158\u003c/em\u003e(6), 450\u0026ndash;460.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSomarriba, E., Cerda, R., Orozco, L., Cifuentes, M., D\u0026aacute;vila, H., Espin, T., Mavisoy, H., \u0026Aacute;vila, G., Alvarado, E., Poveda, V., \u0026amp; Astorga, C. (2013). Carbon stocks and cocoa yields in AF systems of Central America. Agriculture, Ecosystems \u0026amp; Environment, 173, 46\u0026ndash;57.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoto-Pinto, L., Anzueto, M., Mendoza, J., Ferrer, G. J., \u0026amp; de Jong, B. (2010). Carbon sequestration through agroforestry in indigenous communities of Chiapas, Mexico. Agroforestry Systems, \u003cem\u003e78\u003c/em\u003e, 39\u0026ndash;51.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTegegne, Y. T., Lindner, M., Fobissie, K., \u0026amp; Kanninen, M. (2016). Evolution of drivers of deforestation and forest degradation in the Congo Basin forests: exploring possible policy options to address forest loss. Land Use Policy, \u003cem\u003e51\u003c/em\u003e, 312\u0026ndash;324.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrefon, T. (2023). \u003cem\u003eBushmeat: Culture, Economy and Conservation in Central Africa\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTyukavina, A., Hansen, M. C., Potapov, P., Parker, D., Okpa, C., Stehman, S. V., Kommareddy, I., \u0026amp; Turubanova, S. (2018). Congo Basin forest loss dominated by increasing smallholder clearing. Science Advances, \u003cem\u003e4\u003c/em\u003e, eaat2993.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVaast, P., Bertrand, B., Perriot, J. J., Guyot, B., \u0026amp; Genard, M. (2006). Fruit thinning and shade improve bean characteristics and beverage quality of coffee (Coffea arabica L.) under optimal conditions. Journal of the Science of Food and Agriculture, \u003cem\u003e86\u003c/em\u003e(2), 197\u0026ndash;204.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Asten, P. J., Wairegi, L. W. I., Mukasa, D., \u0026amp; Uringi, N. O. (2011). Agronomic and economic benefits of coffee\u0026ndash;banana intercropping in Uganda\u0026rsquo;s smallholder farming systems. Agricultural Systems, 104(4), 326\u0026ndash;334.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVanden Abeele, S., Janssens, S. B., Asimonyio Anio, J., Bawin, Y., Depecker, J., Kambale, B., \u0026hellip; Vandelook, F. (2021). Genetic diversity of wild and cultivated Coffea canephora in north-eastern DR Congo and the implications for conservation. American Journal of Botany, 108(12), 2425\u0026ndash;2434.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVanderhaegen, K., Verbist, B., Hundera, K., \u0026amp; Muys, B. (2015). REALU vs. REDD+: carbon and biodiversity in the Afromontane landscapes of SW Ethiopia. Forest Ecology and Management, 343, 22\u0026ndash;33.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Noordwijk, M., Rahayu, S., Hairiah, K., Wulan, Y. C., Farida, A., \u0026amp; Verbist, B. (2002). Carbon stock assessment for a forest-to-coffee conversion landscape in Sumber-Jaya (Lampung, Indonesia): from allometric equations to land use change analysis. Science in China Series C Life Siences, \u003cem\u003e45\u003c/em\u003e, 75\u0026ndash;86.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVan Noordwijk, M., Duguma, L. A., Dewi, S., Leimona, B., Catacutan, D. C., Lusiana, B., \u0026hellip; \u0026amp; Minang, P. A. (2018). SDG synergy between agriculture and forestry in the food, energy, water and income nexus: reinventing agroforestry?. Current opinion in environmental sustainability, 34, 33\u0026ndash;42.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWintgens, J. N. (2004). \u003cem\u003eCoffee: growing, processing, sustainable production. A guidebook for growers, processors, traders, and researchers\u003c/em\u003e. WILEY-VCH Verlag GmbH \u0026amp; Co. KGaA.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Bank Data Bank. (2021). World Development Indicators, DRC. URL: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://databank.worldbank.org/source/worlddevelopment-indicators\u003c/span\u003e\u003cspan address=\"https://databank.worldbank.org/source/worlddevelopment-indicators\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWorld Bank (2022). Country profile: Democratic Republic of the Congo. Retrieved from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://data.worldbank.org/country/CD\u003c/span\u003e\u003cspan address=\"https://data.worldbank.org/country/CD\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWurz, A., Tscharntke, T., Martin, D. A., Osen, K., Rakotomalala, A. A., Raveloaritiana, E., \u0026hellip; Grass, I. (2022). Win-win opportunities combining high yields with high multi-taxa biodiversity in tropical AF. Nature Communications, 13(1), 4127.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZewdie, B., Tack, A. J., Ayalew, B., Wondafrash, M., Nemomissa, S., \u0026amp; Hylander, K. (2022). Plant biodiversity declines with increasing coffee yield in Ethiopia's coffee agroforests. Journal of Applied Ecology, 59(5), 1198\u0026ndash;1208.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"agroforestry-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"agfo","sideBox":"Learn more about [Agroforestry Systems](http://link.springer.com/journal/10457)","snPcode":"10457","submissionUrl":"https://submission.nature.com/new-submission/10457/3","title":"Agroforestry Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Agricultural intensification, Agroforestry, Coffea canephora, Congo Basin, Ecosystem Services, EUDR","lastPublishedDoi":"10.21203/rs.3.rs-5165806/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5165806/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe rapid decline of tropical rainforests, particularly in the Congo Basin, is predominantly driven by small-scale subsistence agricultural expansion. Tropical agroforestry, particularly coffee agroforestry, is seen as a potential way to balance agricultural productivity with biodiversity conservation and carbon sequestration, despite some possible trade-offs. However, substantial knowledge gaps persist regarding these trade-offs within and across coffee systems, especially in Africa. Here, we used a stratified random sampling design and general additive models to examine the relationship between yield, biodiversity, and carbon stocks in four coffee systems in the DR Congo (monocultures, cultivated agroforestry, wild agroforestry, and forest coffee) based on 79 inventoried plots. Our results demonstrate that coffee yields in cultivated agroforestry systems are not significantly different from monocultures, in contrast to lower yields in wild coffee agroforestry due to excessive shading (\u0026gt;\u0026thinsp;50%). Our study also shows the irreplaceable value of forest coffee systems in terms of biodiversity and carbon sequestration, suggesting that monoculture and agroforestry systems cannot serve as direct substitutes. Forest coffee systems contain three times more total organic carbon (TOC) than the agroforestry systems, which in turn contain almost double the amount of TOC as the coffee monocultures. Our findings revealed a steep decline in woody species diversity, including large changes in community composition, and carbon stocks from forest coffee to agroforestry, with comparatively smaller reductions from agroforestry to monocultures. On the one hand, our study identified convex relationships between woody species diversity and robusta coffee yield, as well as between carbon stocks and robusta yield. On the other hand, synergies are found between carbon stocks and woody plant diversity. One can thus say that coffee agroforestry systems allow the preservation of part of the biodiversity and carbon stocks while also supporting farmer\u0026rsquo;s livelihood. However, applying EUDR guidelines may hinder the adoption of these agroforestry systems due to the regulation\u0026rsquo;s inherent binary classification of forest versus non-forest.\u003c/p\u003e","manuscriptTitle":"Synergies and Trade-offs between Robusta Yield, Carbon Stocks and Biodiversity across Coffee Systems in the DR Congo","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-25 08:24:32","doi":"10.21203/rs.3.rs-5165806/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-11-19T14:02:39+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-11-16T04:58:34+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-10-01T18:07:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"9542174608918700107881287244817048798","date":"2024-10-01T14:50:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"81653481237393871713646466904825520477","date":"2024-09-30T15:19:54+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-09-30T15:02:57+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-09-30T10:30:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-09-28T12:30:48+00:00","index":"","fulltext":""},{"type":"submitted","content":"Agroforestry Systems","date":"2024-09-27T14:15:09+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"agroforestry-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"agfo","sideBox":"Learn more about [Agroforestry Systems](http://link.springer.com/journal/10457)","snPcode":"10457","submissionUrl":"https://submission.nature.com/new-submission/10457/3","title":"Agroforestry Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"4fd1e907-a0ff-4982-b559-0b62a6ce1b93","owner":[],"postedDate":"December 25th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-02-17T15:58:37+00:00","versionOfRecord":{"articleIdentity":"rs-5165806","link":"https://doi.org/10.1007/s10457-025-01140-9","journal":{"identity":"agroforestry-systems","isVorOnly":false,"title":"Agroforestry Systems"},"publishedOn":"2025-02-14 15:56:54","publishedOnDateReadable":"February 14th, 2025"},"versionCreatedAt":"2024-12-25 08:24:32","video":"","vorDoi":"10.1007/s10457-025-01140-9","vorDoiUrl":"https://doi.org/10.1007/s10457-025-01140-9","workflowStages":[]},"version":"v1","identity":"rs-5165806","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5165806","identity":"rs-5165806","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}