{"paper_id":"0f18cb56-b679-4de8-8948-8ee7d5f72db2","body_text":"1 \n \nEffects of Agroforestry Trees on Microclimate and Enset ( Ensete ventricosum ) \nMorphophysiology in South Ethiopia \nAdmasu Yirgalema,c,*, Gezahegn Garo c, Rony Swennen e,f, Sabura Shara c, Bart Muys d, Oliver Honnay b, Karen \nVancampenhouta \na Division for Forest, Nature and Landscape, KU Leuven, Campus Geel, Belgium  \nb Dept. of Biology, Division of Ecology, Evolution and Biodiversity Conservation, KU Leuven, Kasteelpark Arenberg 31, \n2435, 3001, Leuven, Belgium  \nc Dept. of Horticulture, College of Agricultural Sciences, Arba Minch University, Arba Minch, Ethiopia  \nd Division Forest, Landscape and Nature, KU Leuven, Celestijnenlaan 200E-2411, 3001, Leuven, Belgium \ne Department of Biosystems, KU Leuven, Willem De Croylaan 42, 3001 Leuven, Belgium \n f International Institute of Tropical Agriculture, Sendusu, Uganda  \n \n* Corresponding author. Division for Forest, Nature and Landscape, KU Leuven, Campus Geel, Belgium. E -mail address: \nadmayir@gmail.com \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \n \nAbstract \nEnset (Ensete ventricosum), a multipurpose crop domesticated exclusively in Ethiopia, serves as a staple \nfood for millions of smallholder farmers. It is primarily cultivated as a monocrop in homegardens, \nleaving it vulnerable to climate change risks. One potential nature-based solution involves agroforestry \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n2 \n \nsystems; however, enset’s response to canopy cover remains unclear. This study examined how \nscattered trees in enset farms affected microclimate and enset morpho -physiology in South Ethiopia. \nTrees significantly modified the microclimate conditions in enset homegardens. The average daily \nreductions in air, soil surface, and soil temperatures ranged from -0.5 to -1.9 °C, -0.4 to -2.1 °C, and \n+0.4 to -1.0 °C, respectively. The minimum soil moisture offset ranged from +0.8% to +5.7%. Although \nthe tree identity effect on enset growth was negligible, planting position relative to the overstory trees \nsignificantly influenced enset responses. Most morphophysiological tra its were higher  under tree \ncanopies, with progressively lower values at the edge and outside the tree canopy. In contrast, leaf dry \nmatter content exhibited an inverse trend , aligning with the leaf economics spectrum . These results \ndemonstrate enset’s adaptability to canopy shade, suggesting potential for agroforestry expansion.  \nCultivar-specific shade tolerance and ideal shade levels to maintain enset productivity  should be \ninvestigated further. \nKEYWORDS: Climate change, Ensete ventricosum , Homegarden  agroforestry, Shade adaptation , \nSustainable production \n1. Introduction  \nIn 2023, an estimated 28.9% of the global population was moderately or severely food insecure (FAO \net al ., 2024) . This situation has worsened since 2015 , mainly  due to conflicts and  climate-related \nextremes, jeopardizing the progress toward achieving the UN Sustainable Development Goal of zero \nhunger by 2030 (IPCC, 2023) . Especially in sub -Saharan Africa, t he prevalence of severe food \ninsecurity is on the rise (FAO et al., 2024), and the population of 1.2 billion now is expected to surpass \n2 billion by 2050, increasing food demand by 3.9 percent per year (Cardell et al., 2024). It is key that \nincreasing agricultural production happens sustainably, without undermining the ecosystems’ capacity \nto sustain human well-being (Torres et al., 2021). In this context, neglected/orphan crops are gaining \nattention as an avenue to alleviate food insecurity challenges under a changing climate (Talabi et al., \n2022). Often cultivated by subsistence farmers, these crops have significant potential to contribute to \nsustainable food systems (Mabhaudhi et al., 2019; Tadele et al., 2024). They are often  more stress-\ntolerant and may have unique and beneficial nutritional profiles (Chapman et al., 2022). Enset (Ensete \nventricosum), a multipurpose perennial herbaceous plant  domesticated only in Ethiopia , is a prime \nexample of an underexploited crop and a potential candidate for famine insurance in climate-vulnerable \nregions (African Orphan Crops Consortium, 2013; FAO et al., 2024; National Research Council, 2006). \nDespite being a staple and food security crop for millions (Borrell et al., 2020; Kidane et al., 2021; \nTadele et al., 2024), enset has been relatively neglected by scientific research and is inarguably the least-\nstudied African crop (Venkatesan et al., 2022). Known as the ‘tree against hunger’ (Brandt et al., 1997), \nenset offers flexible harvest timing, high yield, long storage, and putative drought tolerance, enabling \nfarmers to bridge adverse periods of harsh conditions (Birmeta et al., 2004; Borrell et al., 2020; Chase \net al., 2023; Yemata, 2020). It is considered a first-rated climate-smart crop due to its ability to withstand \nprolonged drought lasting more than five years (Kudama et al., 2022). Due to this capability, areas with \na more frequent severe drought history in southwestern Ethiopia have a much larger proportion of enset \nproduction (Chase et al., 2023). According to Tsegaye & Struik (2001), its potential yields per hectare \nmay be higher than any other crop cultivated in Ethiopia. Enset-producing households in Ethiopia are \ntherefore less vulnerable to shocks and perceive less risk (Feyisa et al., 2022). Enset is now estimated \nto be a staple crop feeding 20 million people in  its current cultivation range in Ethiopia. But its wild \ndistribution suggests that it could serve a much wider area in Ethiopia, Kenya, Uganda, and Rwanda, \nand enset cultivation might prove feasible for an additional 87.2 –111.5 million people (Koch et al., \n2021).    \nThe enset production system in most parts of southern Ethiopia is a monoculture  in homegardens  \n(Degefa & Dawit, 2018) , surrounding houses (Shara et al., 2021). Although enset requires relatively \nfew off-farm inputs (Senbeta et al., 2022), those further from the house  often remain undernourished \ndue to limited manure application, resulting in reduced growth because of low soil nutrient availability \nand reduced organic matter, higher temperature and moisture stress, or due to perceptions that preferred \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n3 \n \nfood types need less input (Amede & Taboge, 2007; Shara et al., 2021). Ongoing climate change in the \nCentral Rift of Ethiopia is likely to exacerbate this, which makes adaptation strategies increasingly \nurgent (Belay et al ., 2017) . The introduction of trees into farming systems (agroforestry) has the \npotential to address these climate change-related challenges (Torres et al., 2021). Indeed, the potential \nbenefits of agroforestry are well -known. For example, compared to monocropping, agroforestry \nsystems are better buffered against extreme climate events like temperature fluctuations (Niether et al., \n2018). Additionally, agroforestry supports long -term productivity (IPCC, 2023) and diversifies farm \nincome (Kassie, 2017). Agroforestry systems in East Africa contribute to livelihoods by providing food, \nfodder, firewood, and income (Muthuri et al., 2023). Moreover, these practices help to reduce runoff \nand soil loss and improve slope stability, infiltration rates, and soil moisture content (Hemp, 2006; \nKuyah et al., 2019). Enset is a monocotyledonous crop with a fibrous rooting system that grows from \nthe corm and primarily exploits nutrients from the topsoil  (Brandt et al., 1997; Zewdie et al., 2008). \nWoody trees , on the other hand, enhance soil fertility through nutrient cycling and enrich the soil \nthrough litter deposition (Negash & Starr, 2013), which therefore enhances nutrient availability. Enset \nwild relatives  are reported to grow in forest gaps (Birmeta et al ., 2004; Borrell et al ., 2019) , and \ninterestingly, in some parts of Southern Ethiopia, it is also cultivated  in homegarden agroforestry \nsystems (Abebe & Bongers, 2012; Mellisse et al ., 2018) . Those systems, which were originally \ndominated by natural forests, integrated Ensete ventricosum and Coffea arabica following selective \ncanopy thinning  (Negash & Starr, 2013) . Thus, fields are enriched with diverse, multipurpose tree \nspecies that can provide food and agroecological services, such as improving soil fertility, controlling \nerosion, mitigating climate change, and conserving biological diversity  (Abebe et al., 2006; Lelamo, \n2021; Wolka et al., 2021). However, such traditional enset-based homegardens are in decline due to \nsocioeconomic changes (Abebe, 2018; Sahle et al ., 2022) . Nevertheless, these traditional systems \nsuggest that it is possible to grow enset under a (light) canopy of trees and benefit from the added \nbenefits of agroforestry.  \nWhile the general benefits of introducing trees in cropping systems are well -known, tree-crop \ninteractions can also lead to trade -offs (Gonçalves et al ., 2021) . Such trade -offs often arise from \ncompetition between trees and crops for space, water, nutrients, and light (Tschora & Cherubini, 2020), \nresulting in a reduction in the biomass harvest of the main crop (Bertsch-Hoermann et al., 2021; Niether \net al., 2020). Among these factors, competition for light is recognized as a critical factor (Scordia et al., \n2023) and it is strongly influenced by the canopy structure of the tree species , which affects  air \ntemperature, relative humidity, wind speed, and solar radiation intensity (Feng et al., 2023; Speak et al., \n2020). Despite the enset's potential importance, relatively little is known about its biology and ecology \n(Borrell et al., 2019). The wide range of suitable agroecological conditions for enset cultivation suggests \na large phenotypic plasticity to changes in the environment, such as elevation (Negash, 2001; Yemataw \net al ., 2018). Such plasticity enables species to thrive in heterogeneous environments (Nürnberger, \n2013; Sommer, 2020)  and can facilitate subsequent adaptive evolution (Lalejini et al., 2021; West -\nEberhard, 2008). However, despite the enset’s traditional cultivation under tree canopies in agroforestry \nsystems, there is hardly any data on its interaction with trees, particularly regarding responses to canopy \ncover remains poorly documented.  \nHere, we capitalized on the presence of  woody tree species in enset-producing homegardens in the \nGamo highlands of South Ethiopia to evaluate the potential benefits of these trees in enset production \nsystems. The specific objectives were twofold: first, to examine the effects of canopy trees of different \nspecies in enset homegardens on microclimate regulation; and second, to evaluate the effects of the tree \nspecies canopy cover on the morphophysiological traits of enset. We hypothesized that the different tree \nspecies vary in canopy structure, governing the understory enset plants' exposure to light and the overall \nmicroclimate. Additionally, we hypothesized that enset exhibits phenotypic plasticity in response to \nshading imposed by the canopy cover of the tree species. Our findings supports the promotion of enset-\nbased homegarden agroforestry systems by addressing knowledge gaps related to enset's phenotypic \ninteraction with woody tree species. \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n4 \n \n2. Material and methods \nDescription of the study area \nThe study was conducted in Chencha  Zuria, Gerese Zuria, and Kamba Zuria woredas of the Gamo \nhighlands of south Ethiopia at an elevation ranging from 2100 to 3000 m a.s.l. The Gamo zone is \ngeographically located between 5°55′ N and 6°20′ N latitude and between 37°10′ E and 37°40′ E \nlongitude in the South Ethiopian Rift (Shalishe et al., 2023).  \n \nFigure 1. Sampling plots (individual enset farms) in three woredas (Chencha Zuria, Gerese Zuria, and Kemba \nZuria) of the Gamo highlands of South Ethiopia. \nThe mean temperature ranges between 23°C and 14°C in the lowlands and highlands, respectively; the \nmean annual rainfall is between 750mm and 1700mm (Berhanu et al., 2013; Coltorti et al., 2019). The \nlandscape is characterized by steep slopes affected by landslides and dissected by concave valleys and \ngullies, with the uppermost part characterized by gently undulating surfaces across a range of altitudes \n(Coltorti et al., 2019). The region’s natural vegetation is Afromontane forest; however, the landscape \nhas been severely deforested and is now predominantly covered by annual crops (Assefa & Bork, 2014, \n2016). The Gamo people are predominantly farmers, with mixed crops and livestock production \n(Gemeda et al., 2023). Enset is among the main food crops and is grown in traditional homegardens \nsurrounding the house (Shara et al., 2021).   Unlike the typical enset-agroforestry systems in the Gedeo \nand Sidama Zones, in Gamo, trees are rarely incorporated into enset plots and are instead mainly found \nalong farm boundaries, near roads, or in small patches of woodlands (own observation).   \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n5 \n \nStudy species selection \nA reconnaissance field visit was conducted in the Gamo highlands' enset-producing homegardens in \nthree woredas to identify the commonly present tree species and the availability of trees integrated with \nthe enset farming system. During the survey, over 150 enset homegardens were randomly selected for \nobservation, and seven tree species were identified as commonly present. These species, listed in order \nof importance/dominance, are Croton macrostachyus, Ficus sur, Erythrina brucei, Hagenia abyssinica, \nCordia africana, Persea americana, and Prunus africana.  \nTable 1. Selected tree species planted in Enset homegardens of Gamo highlands, their family, ecological services, \nand characteristics (Kuria et al., 2017) \nTree species  Family Ecological Services Growth rate \nCroton macrostachyus Euphorbiaceae \nLive fence, shade, erosion control, \nriverbank stabilization , windbreak, \nmulching \nFairly fast-\ngrowing  \nFicus sur Moraceae Live fence, shade, erosion control Unknown \nErythrina brucei Fabaceae Ornamental, live fence, erosion control, \nsoil fertility improvement through \nNitrogen-fixing, and mulch/leaves \nSlow growing \nHagenia abyssinica Rosaceae \nOrnamental, live fence, erosion control, \nsoil fertility improvement through \nmulch/leaves \nSlow growing \nCordia africana Boraginaceae \nOrnamental, live fence, dead fence, shade, \nerosion control, soil fertility improvement \nthrough mulch/leaves \nReasonably \nfast-growing \nExperimental setup \nIn each homegarden, a solitary tree was selected, and three circular plots with different radii were laid \nout under the tree canopy. Enset leaf sampling and data recordings were done at the middle of the tree \ncanopy, at the edge of the tree canopy, and outside the tree canopy. The selected trees were at least 10 \nmeters away from the nearest woody plants , and the direction towards the houses  was omitted from \nsampling, as it is reported to be often overfertilized with manure (Shara et al., 2021).  \n \nFigure 2. Enset farm with a solitary tree (A) and sampling scheme along three distances from a tree trunk (B). \nEnset data recording and sampling were done in the middle of the canopy (a), at the edge (b) , and outside the \ncanopy (c). Microclimat e data recording loggers were installed in the middle and outside of the tree canopy  for \nsix months (February 2024 to July 2024). Arrows indicate direction towards the houses omitted from sampling.  \nA \n B \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n6 \n \nData collection \nTree age was estimated through interviews with household heads or farmers, while other biometric \nattributes were measured directly. Diameter at breast height (DBH) was measured using a diameter tape \nor a caliper, depending on the size of the tree trunk. The crown diameter (CD) was measured using a \ntape, and two measurements were made perpendicularly. Then, the crown area (CA) was estimated from \nthe CD, assuming a circular crown shape (Snowdon et al., 2001). The variables of interest in this study \nwere the tree canopy gap light transmission (canopy openness/closure) and CA. Canopy openness (CO) \nrefers to the proportion of sky unobscured by vegetation/canopy (Jennings et al., 1999). It was assessed \nby taking hemispherical photographs with an Android smartphone equipped with a fisheye lens  and \nanalysed with Gap Light Analysis Mobile App, following the methods described by Cameron et al. \n(2021) and Díaz (2022) . The photographs were taken un der the tree canopy, and the images  were \nvisually inspected to ensure that they were fully under the canopy. For microclimate data recording, 29 \nfarms were sub-selected from the 38 farms for feasibility reasons. Temperature-moisture-sensor (TMS) \ndata logger s (model: s tandard TMS4) were used to measure microclimate parameters  at 15-minute \nintervals starting from February 2024 to the end of July 2024. They recorded temperature at +15, 0, and \n−8 cm relative to the soil surface (further referred to as air, surface, and soil temperature) and volumetric \nsoil moisture to a depth of approximately 14 cm (Wild et al., 2019). They were installed both under the \ncanopy and outside the canopy of the selected tree species. From the 38 homegarden, three enset plants \nof two to three years old were selected at each of the three distances (a total of 342 individual enset \nplants). Fluorometric parameters namely maximum quantum efficiency of Photosystem II (Fv/Fm) and \nperformance index (PIabs), and leaf chlorophyll content were measured from the third active leaf at the \nlamina’s central position using a portable Handy PEA (Hansatech Instruments Ltd., King’s Lynn, \nNorfolk, England) and a chlorophyll content meter (model: CCM -200 plus), respectively. The Fv/Fm \nis a sensitive indicator of plant photosynthetic performance, with healthy samples achieving a maximum \nvalue of approximately 0.85. While PIabs is an indicator of sample vitality, indicating an internal force \nof the sample to resist constraints from the outside. All PIabs measurements were multiplied by 10 as a \ncorrection factor for error in the handy PEA software (Shara et al. 2022, PhD dissertation). Then, the \nmiddle section of the lamina was collected from these leaves following Shara et al. (2021), placed in \nan airtight plastic bag, and moved to the AMU laboratory for further analysis of leaf moisture and dry \nmatter content and specific leaf area (SLA). A similar procedure as Garnier et al. (2001) was followed \nwith little modification for SLA. A 30 cm*15 cm (450 cm2) section of the laminal leaf was taken from \nthe middle part , and fresh weight was recorded . Then, these leaves were oven-dried at 60 °C for 24 \nhours, and then the dry mass was weighed. We then calculated specific leaf moisture content (LMoC), \ndry matter content (LDMC) , and specific leaf area (SLA) from the fresh mass and dry mass of the \nsection of a leaf blade. LDMC was calculated as the ratio of the dry mass of a leaf to its fresh mass and \nexpressed as a percentage, while SLA was measured as the leaf area per dry mass of a specific leaf \nsection. \nStatistical analysis \nAll statistical analyses were conducted using the R software version 4.3.3  (R Core Team, 2024) .  A \nsimple GLM was used to assess the variation between tree species in terms of their biometric \ncharacteristics. Tree age was included as a covariate factor as it highly  varies between trees and tree \nspecies, which could  influence the canopy characteristics . The microclimate offset values were \ncalculated as the differences between the temperatures under and outside the tree canopy, as well as the \ndifferences in soil moisture content. First, the raw microclimate data, recorded at 15-minute intervals, \nwere processed to obtain the daily minimum and maximum values. Then, the offsets were calculated \nfrom these values.  Next, we analyzed the effects of tree species on the offsets  using a GLM, by \nincluding the crown area as a covariate. Then, a linear mixed-effect model was employed to model leaf \nmoisture and leaf dry matter content, chlorophyll content, specific leaf area, quantum efficiency, and \nperformance index against tree species, planting position relative to the tree canopy (hereafter referred \nto as radial distance) , and their interaction , as fixed factors  by individual farms as random factor. \nMultiple comparisons were performed with Tukey’s post hoc test using the ‘emmeans’ function in the \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n7 \n \n‘emmeans’ R package. The visualization of the results was performed using the ‘ggplot2’ and ‘ggpubr’ \nfunctions in R software.  \n3. Results \n3.1. Tree biometric characteristics \nThe tree species significantly differed in mean age (F  = 3.65; p < 0.05), with F. sur having the oldest \ntrees, whereas the youngest trees observed were C. macrostachyus (Table 2). The GLM result showed \nno significant difference in DBH, CD, and CA between the tree species. However, significant \ndifferences were observed in CO (F = 3.71; p < 0.05) among the species.  The highest CO (18.2%) was \nrecorded for C. macrostachyus, which was significantly higher than the CO of E. brucei (13.9%) but \nnot different from that of F . sur (14.7%), C. africana (14.8%), and H. abyssinica (17.1%) (Table 2). \nTable 2. Comparison of biometric characteristics (mean±SE) of trees belonging to five species, solitarily present \ninside enset farms in homegardens of the Gamo highlands of South Ethiopia.  \nTree species Age (years) DBH (cm) CD (m) CA (m²) CO (%) \nC. macrostachyus 9.8±8.0b 50.9±10.6a 9.6±1.2a 79.3±18.9a 18.2±1.3a \nF. sur 44.9±6.9a 59.0±9.5a 10.2±1.1a 91.0±17.0a 14.7±1.2ab \nE. brucei 33.8±6.9ab 68.0±8.8a 10.1±1.0a 89.0±15.6a 13.9±1.1b \nH. abyssinica 18.9±6.9ab 42.2±8.8a 7.4±1.0a 48.9±15.6a 17.1±1.1ab \nC. africana 20.4±6.9ab 46.6±8.7a 8.5±1.0a 61.0±15.6a 14.8±1.1ab \nDifferent letters  in the same column indicate significant differences among tree species . Significances were \nderived from a GLM with tree age as a covariate.   \n3.2. Tree influences on microclimate offsets in enset homegardens \nThe daily offsets of maximum air, surface, and soil temperatures were highly significantly influenced \nby tree species (Table 3). The highest maximum air temperature offset was recorded under F. sur (-1.9 \n°C), while the lowest offset was recorded under H. abyssinica (-0.5 °C) (Figure 3, Table S1). Similarly, \nthe highest and lowest cooling of the maximum surface temperature was recorded under F. sur (-2.1 °C) \nand H. abyssinica (-0.4 oC), respectively. On the other hand, t he highest maximum soil temperature \noffset was recorded under F. sur (-1.0 °C), which was not significantly different from the offset by C. \nafricana (-0.9 °C) and C. macrostachyus (-0.8 °C). The lowest offset (+0.4 °C) was recorded under H. \nabyssinica (Figure 3, Table S1). Tree species significantly influenced the minimum volumetric soil \nmoisture content (VMC) offsets ( p < 0.001, Table 3). The highest soil moisture offset was observed \nunder E. brucei (+5.7 %), while the lowest minimum soil moisture offset (+0.8%) was recorded under \nH. abyssinica (Figure 3, Table S1). On the other hand, the soil moisture offsets observed by C. africana, \nC. macrostachyus, and F . sur were statistically similar (Figure 3).  \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n8 \n \n \nFigure 3. Microclimate offsets under solitary trees of five tree species in enset homegardens of the Gamo highlands \nof South Ethiopia. Bars represent the offsets of maximum air, surface, and soil temperature, and soil volumetric \nmoisture content offsets. Different letters in each figure indicate significant differences among the tree species' \neffects on offsets based on a Tukey’s HSD post hoc test. VMC: soil volumetric moisture content. \nAll the microclimate offsets were also significantly affected by the interaction between tree species and \ncrown area (p < 0.001) (Table 3), indicating that the positive effects of increasing crown area on offsets \nwere highly dependent on the tree species (Fig S1).  An increase in crown area increased the temperature \noffsets, with C. africana  providing the strongest temperature offsets. Exceptionally, an inverse \ncorrelation was observed for soil temperature offset by H. abyssinica. Regarding the soil volumetric \nmoisture content, the offsets by C. africana , C. macrostachyus , and H. abyssinica were positively \ncorrelated with crown area, whereas negative offsets were recorded from E. brucei and F. sur.  \nTable 3. The effect of tree species on offsets of microclimate parameters (maximum air, surface, and soil \ntemperature, and soil moisture content) with the tree's crown area is considered a covariate. \nParameters Factors F value P value \nMax air temperature offset \n Tree species 114.88 <0.001 \n Crown area 122.97 <0.001 \n Tree species * Crown area 17.06 <0.001 \nMax surface temperature offset \n Tree species 52.04 <0.001 \n Crown area 92.39 <0.001 \n Tree species * Crown area 19.93 <0.001 \nMax soil temperature offset \n Tree species 26.85 <0.001 \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n9 \n \n Crown area 11.07 <0.001 \n Tree species * Crown area 34.69 <0.001 \nMinVMC offset  \nTree species 32.28 <0.001  \nCanopy area 9.93 0.002  \nTree species * Canopy area 135.87 <0.001 \nVMC: soil volumetric moisture content \n3.3. Tree canopy cover effect on enset morpho-physiological traits  \nLeaf moisture content of enset varied significantly by tree species (F = 2.50, p = 0.06) and radial distance \n(F = 9.13, p < 0.001), but not by their interaction (F = 1.37 , p = 0.21) (Table 4). Enset under the tree \ncanopy had the highest moisture content (85.5%), significantly higher than outside the canopy (84.5%), \nbut not different from the canopy edge (85.3%) (Table S2). Despite the non -significant interaction, \npairwise com parisons showed that tree species affected leaf moisture in the middle  of the canop y. \nSimilarly, the interaction of tree species and radial distance had no significant effect on leaf dry matter \ncontent (F = 1.37, p = 0.21), but radial distance alone did (F = 9.13 , p < 0.001) (Table 4). The highest \ndry matter content (15.5%) was recorded from enset outside the canopy, significantly higher than under \nthe canopy (14.5%) (Table S2). Tree species significantly affected dry matter content only in the middle \nof the canopy, where it was lower under E. brucei than C. africana (Figure 5). \n \nFigure 5. Box plots comparing tree species and enset radial distance (at the middle, edge, and outside canopy) \neffects on LDMC of enset grown in homegardens of Gamo highlands of Ethiopia. Different letters within each \ndistance indicate significant differences among tree species based on a Tukey HSD post hoc test. The differences \nbetween planting positions are indicated by ‘ns’: p-value>0.05, **: p-value<0.01, ***: p-value <0.001. \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n10 \n \nRadial distance significantly affected the specific leaf area (SLA) of enset (F = 21.68, p < 0.001; Table \n4). Enset outside the canopy had a lower SLA (200 cm²/g) than those at the edge (220 cm²/g) and under \nthe canopy (229 cm²/g) (Table 4, Figure 6A), showing a decreasing trend with distance from the tree. \nHowever, SLA did not differ significantly between enset at the middle and the edge of the canopy. Tree \nspecies had no significant effect on SLA at any distance, except at the canopy edge, where enset at the \nedge of E. brucei had significantly lower SLA than those at the edge of  C. africana, F . sur, and H. \nabyssinica (Figure 6A). \nLeaf chlorophyll content was not significantly influenced by the interaction of tree species and radial \ndistance (F = 2.09, p = 0.12) or by tree species alone (F = 1.63, p = 0.11) (Table 4). It was marginally \naffected by radial distance (F = 2.7, p = 0.06, Table 4). This marginal difference showed a slight gradient \nalong the radial distance, with the highest chlorophyll content index (18.7) in the middle of the tree \ncanopy, followed by the edge (17.5) and outside (17.0) (Table S2). The pairwise comparison revealed a \nsignificant effect of tree identity only at the edge of the canopy, where enset at the edge of H. abyssinica \nhad lower chlorophyll content than at the edge of C. africana (Figure 6B).  \n \nFigure 6. Box plots comparing tree species and radial distance (at the middle, edge, and outside canopy) effects \non SLA (A) and leaf chlorophyll content index (B) of enset grown in homegardens of Gamo highlands of Ethiopia. \nDifferent letters within each distance indicate significant differences among tree species based on a Tukey HSD \npost hoc test. The differences between planting positions are indicated by ‘ns’: p -value>0.05, **: p-value<0.01, \n***: p-value <0.001. \nEnset leaf chlorophyll fluorometric variables varied significantly across planting positions relative to \nthe tree canopy ( Table 4). The maximum quantum efficiency was significantly affected by the radial \ndistance (F = 14.76, p < 0.001, Table 4), but not by tree species alone (F = 1.80, p = 0.15, Table 4) or \ntheir interaction (F = 1.24, p = 0.28, Table 4). The highest quantum efficiency (0.82) was recorded under \nthe tree canopy, significantly higher than outside the tree canopy (0.79) (Table S2; Figure 7A). Similarly, \nthe Fv/Fm outside the tree canopy was also significantly lower than at the edge of the tree canopy \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n11 \n \n(Figure 7A). Within the radial distances, tree species had no significant effect on Fv/Fm except a  \nmarginal difference outside the canopy, where Fv/Fm outside C. macrostachyus was significantly lower \nthan outside C. africana (Figure 7A).  \n \nFigure 7. Box plots representing tree species and radial distance (at the middle, edge, and outside canopy) effects \non the Fv/Fm (A) and PIabs(B) of enset grown in homegardens of Gamo highlands of Ethiopia. Different letters \nwithin each distance indicate significant differences among tree species based on a Tukey HSD post-hoc test. The \ndifferences between planting positions are indicated by ‘ns’:  p-value > 0.05, **: p -value<0.01, ***: p-value \n<0.001. \nSimilarly, the performance index of enset was not significantly affected by the interaction of tree species \nand radial distance (F = 0.25, p = 0.98) but was significantly influenced by the tree species (F = 2.79, p \n< 0.05) and radial distance (F  = 15.74, p < 0.001) (Table 4). The highest index (26.0) was recorded \nunder the tree canopy, significantly higher than at the edge (23.4) and outside the tree canopy ( 20.0) \n(Table S2, Figure 7B). Generally, both Fv/Fm and PIabs showed a decreasing trend with increasing \ndistance from the tree (Table S2, Figure 8). Considering the tree species’ effect in each radial distance, \na significant difference  was observed at the edge and outside the canopy, with significantly lower \nperformance indexes of enset planted at the edge and outside the canopy of C. macrostachyus compared \nto C. africana (Figure 7B).  \nTable 4. Effects of tree species, radial distance (middle, edge, and outside canopy), and their interaction on LMoC, \nLDMC, SLA, leaf chlorophyll content (CCI), (Fv/Fm and PIabs inferred by linear mixed model with farm included \nas a random factor.  \nResponse variables  F value DF P value R2m/R2c \nLMoC   \n Tree species 2.50 4 0.06 13.35/37.65 \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n12 \n \n Radial distance 9.13 2 <0.001 \n Tree species x Radial distance 1.37 8 0.21  \nLDMC   \n Tree species 2.50 4 0.06 \n13.35/37.65  Radial distance 9.13 2 <0.001 \n Tree species x Radial distance 1.37 8 0.21 \nSLA   \n Tree species 1.62 4 0.19 \n15.28/39.32  Radial distance 21.68 2 <0.001 \n Tree species x Radial distance 1.94 8 0.05 \nCCI \n Tree species 2.09 4 0.11 \n13.14/30.87  Radial distance 2.70 2 0.06 \n Tree species x Radial distance 1.63 8 0.12 \nFv/Fm  \n Tree species 1.80 4 0.15 \n12.08/22.18  Radial distance 14.76 2 <0.001 \n Tree species x Radial distance 1.24 8 0.28 \nPIabs  \n Tree species 2.79 4 0.04 \n13.98/31.33  Radial distance 15.74 2 <0.001 \n  Tree species x Radial distance 0.25 8 0.98 \n4. Discussion \n4.1. Tree species impact on microclimate offsets \nWe found that woody trees are only sparingly incorporated into the enset-based farming systems of the \nGamo highlands. The common tree species identified include C. macrostachyus, F. sur, E. brucei, H. \nabyssinica, and C. africana, either within enset farms or in the outfields (outside the enset farms). These \ntrees have been identified as vital native species in the agroforestry systems of Ethiopia before (Lelamo, \n2021; Molla et al., 2023). Some have also been acknowledged as the most  common tree species in \nbanana agroforestry systems in central Uganda (Mpiira et al ., 2013) . The trees in the study area \nexhibited significant variation in age and canopy structure, particularly in canopy openness, which may \ninfluence understory enset performance. Specifically, the mean canopy openness followed the order: C. \nmacrostachyus > H. abyssinica > C. africana > F . sur > E. brucei. This aligns with prior findings; for \nexample, Lemenih et al . (2004)  reported variability in canopy openness among plantation species \n(Cordia africana, Eucalyptus saligna, Cupressus lusitanica, and Pinus patula) in southern Ethiopia.  \nLikewise, Kohl et al. (2024) found that overstory tree species significantly influenced light transmission \nin agroforestry systems in Ghana.  \nOur findings indicate that tree identity plays a key role in regulating the microclimate, closely linked to \ncanopy characteristics such as  openness and crown area . All studied species reduced both air \ntemperature at 15cm (−0.5 °C to −1.9 °C) and surface temperature (−0.4 °C to −2.1 °C) beneath their \ncanopies. While soil temperature reductions at 8cm depth were modest (−0.6 °C to −1.0 °C), more \npronounced effects may occur under denser canopies. Notably, soil moisture at 1 4 cm depth was \nconsistently higher beneath trees (+0.8% to +5.7%), highlighting their role in enhancing topsoil \nmoisture retention. The tree buffering capacities could be beyond the recorded values,  as all \nmeasurements were taken under the enset plants' canopy, which itself ameliorates microclimate through \nshading with their broad leaves (Jilo et al., 2021; Senbeta et al., 2022).  \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n13 \n \nMicroclimate regulation varied significantly among tree species. F. sur exhibited the strongest buffering \ncapacity, achieving the highest reductions in maximum daily air and soil surface temperatures. C. \nafricana and C. macrostachyus showed similar effectiveness in moderating soil surface temperatures. \nIn contrast, H. abyssinica demonstrated the weakest temperature regulation across all measured \nvariables (air, surface, and soil temperatures). On the other hand, the highest offset of low soil moisture \ncontent was observed under E. brucei, while the lowest buffering effect was recorded under H. \nabyssinica. Trees’ structural attributes play a key role in influencing microclimate buffers (Zhang et al., \n2022), with higher basal area and canopy cover improving buffering capacity (Zellweger et al., 2020; \nZhang et al., 2022). Similarly, Sharmin et al. (2023) reported that trees with higher leaf area index and \nbroader canopies provided the greatest cooling benefits in Australia. In an agroforestry system, Kohl et \nal. (2024) also observed substantial variation in microclimate buffering capacity among common shade \ntree species in Ghana. This capacity is often reflected through reducing the daily maximum temperature \nand increasing the daily minimum temperature (Merle et al., 2022). The tree height-to-crown-base also \naffects the overall understory environment (Blaser-Hart et al., 2021).   \nThe influence of trees on microclimate regulation in our study was strongly correlated with crown area, \nthough species effects varied across different microclimate variables. Increasing the crown area of all \ntree species consistently reduced the air and surface temperatures. However, the effects on soil \ntemperature and volumetric moisture content were less distinct. For instance, soil temperature offset \nwas strongly correlated with the crown area expansion in C. africana, F . sur, and E. brucei, but unclear \nfor C. macrostachyus. In contrast, H. abyssinica exhibited an inverse relationship with soil temperature \noffset, likely because soil temperature is affected not only by canopy cover but also by the amount of \nleaf litter deposited on the ground  (Hou et al., 2020). These canopy-driven microclimate regulations \ncan enhance broader ecosystem functioning. For instance, banana-based agroforestry systems support \ntermite survival (Godfrey et al., 2017), key decomposers in tropical ecosystems, thereby improving soil \nfertility through accelerated litter decomposition (Anbessa & Utaile, 2024; Seidelmann et al., 2016). In \ncoffee agroforestry systems, shade canopies buffer temperature extremes and humidity fluctuations, \nenhancing pollinator diversity (Jha et al., 2014). Similarly, changing environmental conditions, such as \nsoil temperature and moisture, may also alter fungal symbionts critical for nutrient cycling  (Kivlin et \nal., 2011; Tedersoo et al., 2020).  \n4.2. Enset functional trait responses to tree canopy cover/shade \nA critical factor in evaluating a crop's suitability for agroforestry systems is its ability to thrive under a \n(dappled) shade. Our findings revealed that tree species identity had negligible effects on most enset \nleaf traits, whereas radial distance from the trunk exerted a stronger influence. All measured traits varied \nsignificantly with radial distance s, indicating that microclimate gradients , rather than tree species \nidentity, are the primary drivers of enset’s phenotypic responses. Similar results were reported for other \ncrops, such as banana (Senevirathna et al., 2008), cocoa (Isaac et al., 2007; Kohl et al., 2024), sorghum \n(Kessler, 1994), and wheat (Yang et al., 2019). \nMost enset leaf traits exhibited a decreasing trend with increasing distance from the tree trunk, except \nfor leaf dry matter content, which showed an inverse relationship. Notably, the specific leaf area (SLA) \nof enset was significantly higher under the tree canopies than in the open area, suggesting a shade-\nacclimation response, a common adaptive strategy in plants growing under  low-light environments \n(Poorter et al ., 2009) . This morphological adjustment, characterized by thinner, broader leaves, \nenhances light capture efficiency through increased photosynthetic surface area per unit leaf mass  \n(Valladares & Niinemets, 2008). The elevated SLA in shaded conditions is typically associated with \nimproved relative growth rates (Hunt & Cornelissen, 1997; Villar et al., 2005), indicating that enset may \noptimize carbon assimilation under shade. While no prior studies have explicitly examined SLA in \nenset, our results align with observations in other crops, such as banana (Muhidin et al ., 2021; \nSenevirathna et al., 2008) and cacao (Isaac et al., 2007). These findings support the well-documented \nassociation between high SLA and shade tolerance across plant species (Janse-Ten Klooster et al., 2007; \nSánchez-Gómez et al., 2006; Liu et al., 2016), typically involving reduced leaf thickness and stomatal \ndensity (Israeli et al., 1995).    \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n14 \n \nThis morphological ad aptation was accompanied by distinct  physiological changes in leaf tissue \ncomposition. We observed higher leaf moisture content and lower LDMC under shaded conditions \ncompared to open areas, consistent with reports that high -SLA species tend to maintain greater leaf \nwater content (V endramini et al., 2002). The increased hydration under tree canopies likely reflects \nreduced evaporative demand due to moderated microclimate conditions (V alladares et al ., 2008) . \nConversely, the lower dry matter content suggests a strategic trade-off, where shaded leaves prioritize \nlight interception over investment in structural compounds like lignin and cellulose (Wright et al ., \n2004). While thinner leaves enhance light capture, they may increase susceptibility to physical damage \n(He et al., 2019), a critical consideration for optimization of agroforestry systems. \nInterestingly, chlorophyll content showed only marginal variation across radial distance, with a slight \ndecreasing trend with increasing distance. This stability parallels observations in Musa spp.  (Thomas \n& Turner, 2001) , suggesting some species maintain consistent chlorophyll levels despite light \nvariations. However, while absolute chlorophyll content remained stable, we cannot rule out potential \nadjustments in chlorophyll a/b ratios - a common adaptation observed in Musa spp . and Hevea \nbrasiliensis (Senevirathna et al ., 2003, 2008)  that optimizes the photosynthetic apparatus for shade \nconditions. \nFluorometric measurements provided further insight into enset's photochemical adaptations. Both \nmaximum quantum efficiency (Fv/Fm) and Performance Index (PIabs) were significantly higher under \ntree canopies, with values declining toward open areas. These results align with findings in the banana \n(Senevirathna et al ., 2008; Thomas & Turner, 2001) , where direct sunlight caused greater \nphotochemical damage than diffuse light. The elevated Fv/Fm and PIabs under shade suggest that lower \nirradiance mitigates light stress while maintaining sufficient energy capture (Baker, 2008). Notably, the \nhigher photochemical efficiency alongside the stable chlorophyll content implies that enset prioritizes \nphotosynthetic quality (efficiency) over quantity  (light absorption ) under shade condition s. This \nstrategy appears effective for early growth establishment, as reduced photoinhibition in shade promotes \nseedling performance (Senevirathna et al., 2003). \nFurthermore, our findings support the idea that moderate shade in agroforestry systems does not \nnecessarily compromise productivity but may instead induce beneficial physiological adjustments. \nSimilar patterns have been reported in other perennial crops, w here shaded environments enhance \nwater-use efficiency and reduce heat stress (Lin, 2007) . In addition,  other research  suggests that \nmoderate densities of tall trees in agroforestry systems can be compatible with high er productivity of \nbanana and cacao (Salazar-Díaz & Tixier, 2019) . Similarly, enhanced plant height, leaf length, leaf \nnumber, and leaf width  of banana were observed under natural shading  (Muhidin et al ., 2021) . \nAdditionally, low-density shade trees improved nutrient uptake and biomass in cocoa  (Isaac et al ., \n2007). Therefore, given the enset’s long-standing adaptation to agroforestry systems, its plasticity in \nleaf morphology likely contributes to sustained productivity under variable light conditions, a key trait \nfor climate -resilient cropping systems.  In light of alarming global climate change, efforts of \nreforestation in the region  need to prioritize  enset-based homegarden agroforestry as a resilient,  \nsustainable production system.  \n5. Conclusion  \nThe scattered tree species in the enset homegardens of the Gamo highlands regulated extreme climate \nconditions, reducing maximum air, soil surface, and soil temperatures , while keeping minimum soil \nmoisture content higher under the tree canopy. Notably, this buffering capacity may be higher than the \nrecorded values compared to open areas outside the enset plot.  Tree species identity had a negligible \neffect on enset phenology, but radial distance  significantly influenced responses. All \nmorphophysiological parameters (leaf moisture content, SLA, chlorophyll content, Fv/Fm, and \nperformance index) increased closer to the canopy, though leaf dry matter content was significantly \nlower under canopies than at edges or outside. Thus, we conclude that enset’s phenotypic plasticity to \nadapt to environmental changes suggests it may be a shade-loving crop and, more specifically, acclimate \nto mild shade for optimal growth.  \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n15 \n \nGiven the threats of global climate change and regional vulnerability to land degradation, adopting \nenset-based agroforestry with strategic tree spacing to enhance enset physiological performance \nprovides smallholders a resilient and sustainable adaptation strategy. Future work should assess tree -\nmediated soil fertility improvements, cultivar-specific shade tolerance thresholds, yield estimation, and \nnutritional composition of enset products . Moreover, further research on quantifying carbon \nsequestration in woody tree-enset agroforestry systems may provide insight for future scaling of enset \ncultivation towards outfields from its current confinement to homegardens. \nAcknowledgments \nWe thank the Flemish Inter-University Council for Development Collaboration (VLIR-UOS), Belgium, \nfor funding research activities through the VLIR-IUC (inter-university cooperation) program with Arba \nMinch University, Ethiopia.  We also want to acknowledge the support of IUC coordinators Dr. Fassil \nEshetu and Prof. Roel Merckx; P5 project leader Prof. Yisehak Kechero; and AMU-IUC program team \nmembers who were instrumental in facilitating logistics. We want to extend our gratitude to the farmers \nfor sharing information and allowing access to their farms for the study, as well as Sintayehu Tomas, \nLegesse Bode, Beyene Kushe, and Zelalem Aniley for their assistance in gathering field data and \nconducting laboratory analysis. Finally, ChatGPT was instrumental in language improvement and \ncoding assistance. \nFunding \nThis work was supported by the VLIR-UOS (grant number: ET22IUC035A101) \nDisclosure statement \nThe authors report there are no competing interests to declare. \nReference  \nAbebe, T. (2018). Land-use changes in the enset -based agroforestry systems of Sidama, Southern Ethiopia, and \nits implications for agricultural sustainability. Ethiopian Journal of Biological Sciences. \nAbebe, T., & Bongers, F. (2012). Land-use dynamics in enset-based agroforestry homegardens in Ethiopia. In B. \nArts, S. van Bommel, M. Ros -Tonen, & G. Verschoor (Eds.), Forest-people interfaces: Understanding \ncommunity forestry and biocultural diversity  (pp. 69 –85). Academic Publishers. \nhttps://doi.org/10.3920/978-90-8686-749-3_4 \nAbebe, T., Wiersum, K. F., Bongers, F., & Sterck, F. (2006). Diversity and dynamics in homegardens of Southern \nEthiopia. In B. M. Kumar & P. K. R. Nair (Eds.), Tropical Homegardens: A Time-Tested Example of \nSustainable Agroforestry  (pp. 123 –142). Springer Netherlands. https://doi.org/10.1007/978 -1-4020-\n4948-4_8 \nAfrican Orphan Crops Consortium. (2013). Ensete ventricosum – African Orphan Crops Consortium . \nhttps://africanorphancrops.org/ensete-ventricosum/ \nAmede, T., & Taboge, E. (2007). Optimizing soil fertility gradients in the enset (Ensete ventricosum) systems of \nthe Ethiopian Highlands: Trade -offs and local innovations. In A. Bationo, B. Waswa, J. Kihara, & J. \nKimetu (Eds.), Advances in integrated soil fertility management in sub-Saharan Africa: Challenges and \nopportunities (pp. 289–297). Springer Netherlands. https://doi.org/10.1007/978-1-4020-5760-1_26 \nAnbessa, G. N., & Utaile, Y . U. (2024). Scattered trees in smallholder farms improve soil properties and litter \ndecomposition in humid agroecosystems in Ethiopia. Agroforestry Systems . \nhttps://doi.org/10.1007/s10457-024-00982-z \nAssefa, E., & Bork, H. -R. (2014). Deforestation and forest management in Southern Ethiopia: Investigations in \nthe Chencha and Arba Minch areas. Environmental Management , 53(2), 284 –299. \nhttps://doi.org/10.1007/s00267-013-0182-x \nAssefa, E., & Bork, H.-R. (2016). Dynamics and driving forces of agricultural landscapes in Southern Ethiopia – \na case study of the Chencha and Arba Minch areas. Journal of Land Use Science , 11(3), 278 –293. \nhttps://doi.org/10.1080/1747423X.2014.940613 \nBaker, N. R. (2008). Chlorophyll Fluorescence: A Probe of Photosynthesis In Vivo. Annual Review of Plant \nBiology, 59(V olume 59, 2008), 89–113. https://doi.org/10.1146/annurev.arplant.59.032607.092759 \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n16 \n \nBelay, A., Recha, J. W., Woldeamanuel, T., & Morton, J. F. (2017). Smallholder farmers’ adaptation to climate \nchange and determinants of their adaptation decisions in the Central Rift Valley of Ethiopia. Agriculture \n& Food Security, 6(1), 24. https://doi.org/10.1186/s40066-017-0100-1 \nBerhanu, B., Melesse, A. M., & Seleshi, Y . (2013). GIS -based hydrological zones and soil geo -database of \nEthiopia. CATENA, 104, 21–31. https://doi.org/10.1016/j.catena.2012.12.007 \nBertsch-Hoermann, B., Egger, C., Gaube, V ., & Gingrich, S. (2021). Agroforestry trade -offs between biomass \nprovision and aboveground carbon sequestration in the alpine Eisenwurzen region, Austria. Regional \nEnvironmental Change, 21(3), 77. https://doi.org/10.1007/s10113-021-01794-y \nBirmeta, G., Nybom, H., & Bekele, E. (2004). Distinction between wild and cultivated enset (Ensete ventricosum) \ngene pools in Ethiopia using RAPD markers. Hereditas, 140(2), 139–148. https://doi.org/10.1111/j.1601-\n5223.2004.01792.x \nBlaser-Hart, W. J., Hart, S. P., Oppong, J., Kyereh, D., Yeboah, E., & Six, J. (2021). The effectiveness of cocoa \nagroforests depends on shade-tree canopy height. Agriculture, Ecosystems & Environment, 322, 107676. \nhttps://doi.org/10.1016/j.agee.2021.107676 \nBorrell, J. S., Biswas, M. K., Goodwin, M., Blomme, G., Schwarzacher, T., Heslop -Harrison, J. S. (Pat), \nWendawek, A. M., Berhanu, A., Kallow, S., Janssens, S., Molla, E. L., Davis, A. P., Woldeyes, F., Willis, \nK., Demissew, S., & Wilkin, P. (2019). Enset i n Ethiopia: A poorly characterized but resilient starch \nstaple. Annals of Botany, 123(5), 747–766. https://doi.org/10.1093/aob/mcy214 \nBorrell, J. S., Goodwin, M., Blomme, G., Jacobsen, K., Wendawek, A. M., Gashu, D., Lulekal, E., Asfaw, Z., \nDemissew, S., & Wilkin, P. (2020). Enset-based agricultural systems in Ethiopia: A systematic review of \nproduction trends, agronomy, processing and the wider food security applications of a neglected banana \nrelative. Plants, People, Planet, 2(3), 212–228. https://doi.org/10.1002/ppp3.10084 \nBrandt, S. A., Spring, A., Hiebsch, C., McCabe, J. T., Tabogie, E., Diro, M., Woldemichael, G., Yntiso, G., Shigeta, \nM., & Tesfaye, S. (1997). The “tree against hunger”: Enset -based agricultural systems in Ethiopia. \nAmerican Association for the Advancement of Science , 56. \nhttps://cir.nii.ac.jp/crid/1130282270896766720 \nCameron, H. A., Díaz, G. M., & Beverly, J. L. (2021). Estimating canopy fuel load with hemispherical \nphotographs: A rapid method for opportunistic fuel documentation with smartphones. Methods in \nEcology and Evolution, 12(11), 2101–2108. https://doi.org/10.1111/2041-210X.13708 \nCardell, L., Zereyesus, Y . A., Ajewole, K., Farris, J., Johnson, M. E., Lin, J., Valdes, C., & Zeng, W. (2024). \nInternational Food Security Assessment, 2024 –34. http://www.ers.usda.gov/publications/pub -\ndetails/?pubid=109852 \nChapman, M. A., He, Y ., & Zhou, M. (2022). Beyond a reference genome: Pangenomes and population genomics \nof underutilized and orphan crops for future food and nutrition security. New Phytologist, 234(5), 1583–\n1597. https://doi.org/10.1111/nph.18021 \nChase, R. R., Büchi, L., Rodenburg, J., Roux, N., Wendawek, A., & Borrell, J. S. (2023). Smallholder farmers \nexpand production area of the perennial crop enset as a climate coping strategy in a drought -prone \nindigenous agrisystem. Plants, People, Planet, 5(2), 254–266. https://doi.org/10.1002/ppp3.10339 \nColtorti, M., Pieruccini, P., Arthur, K. J. W., Arthur, J., & Curtis, M. (2019). Geomorphology, soils and palaeosols \nof the Chencha area (Gamo Gofa, south western Ethiopian Highlands). Journal of African Earth \nSciences, 151, 225–240. https://doi.org/10.1016/j.jafrearsci.2018.12.018 \nDegefa, I., & Dawit, E. (2018). Indigenous knowledge on cultivation and consumption of enset (Enset \nventricosum) in Kercha district, West Guji Zone, Oromia Region of Ethiopia. Journal of Plant Breeding \nand Genetics, 6(3), 87–94. https://doi.org/10.33687/pbg.006.03.2584 \nDíaz, G. M. (2022). Optimizing forest canopy structure retrieval from smartphone -based hemispherical \nphotography (p. 2021.03.17.435793). bioRxiv. https://doi.org/10.1101/2021.03.17.435793 \nFAO, IFAD, UNICEF, WFP, & WHO. (2024). The State of Food Security and Nutrition in the World 2024 -\nFinancing to end hunger, food insecurity and malnutrition in all its forms. FAO ; IFAD ; UNICEF ; WFP ; \nWHO ; https://openknowledge.fao.org/handle/20.500.14283/cd1254en \nFeng, X., Wen, H., He, M., & Xiao, Y . (2023). Microclimate effects and influential mechanisms of four urban tree \nspecies underneath the canopy in hot and humid areas. Frontiers in Environmental Science, 11:1108002. \nhttps://doi.org/10.3389/fenvs.2023.1108002 \nFeyisa, A. D., de Mey, Y ., & Maertens, M. (2022). Orphan crops and the vulnerability of rural livelihoods: The \ncase of enset in Ethiopia. Q Open, 2(2), 1–22. https://doi.org/10.1093/qopen/qoac029 \nGarnier, E., Shipley, B., Roumet, C., & Laurent, G. (2001). A standardized protocol for the determination of \nspecific leaf area and leaf dry matter content. Functional Ecology , 15(5), 688 –695. \nhttps://doi.org/10.1046/j.0269-8463.2001.00563.x \nGemeda, B., Tesfaye, G., Simachew, A., Wang, A., Mekonnen, A., Guadie, A., & Andualem, B. (2023). Linking \nindigenous with scientific knowledge about enset ( Ensete ventricosum) disease management in Gamo \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n17 \n \nhighlands of Ethiopia: Evidence from local people response, soil physicochemical and microbial \ndynamics. Agricultural Systems, 212, 103768. https://doi.org/10.1016/j.agsy.2023.103768 \nGodfrey, S., Godfrey, K. H., Kenneth, N., Samuel, M., Tushemereirwe, Wilberforce, K., Eldad, K., & Charles, S. \n(2017). Termite assemblages in the banana agroforestry systems of Kiboga District, Central Uganda. \nInternational Journal of Agroforestry and Silviculture, 4(3), 267–277. \nGonçalves, B., Morais, M. C., Pereira, S., Mosquera -Losada, M. R., & Santos, M. (2021). Tree –crop ecological \nand physiological interactions within climate change contexts: A mini-review. Frontiers in Ecology and \nEvolution, 9:661978. https://doi.org/10.3389/fevo.2021.661978 \nHe, P., Wright, I. J., Zhu, S., Onoda, Y ., Liu, H., Li, R., Liu, X., Hua, L., Oyanoghafo, O. O., & Ye, Q. (2019). \nLeaf mechanical strength and photosynthetic capacity vary independently across 57 subtropical forest \nspecies with contrasting light requiremen ts. New Phytologist , 223(2), 607 –618. \nhttps://doi.org/10.1111/nph.15803 \nHemp, A. (2006). The Banana Forests of Kilimanjaro: Biodiversity and Conservation of the Chagga \nHomegardens. Biodiversity and Conservation , 15(4), 1193–1217. https://doi.org/10.1007/s10531 -004-\n8230-8 \nHou, D., Liu, C., Qiao, X., & Guo, K. (2020). Asymmetric effects of litter accumulation on soil temperature and \ndominant plant species in fenced grasslands. Ecosphere, 11(11), e03289. \nhttps://doi.org/10.1002/ecs2.3289 \nHunt, R., & Cornelissen, J. H. C. (1997). Components of relative growth rate and their interrelations in 59 \ntemperate plant species. The New Phytologist , 135(3), 395 –417. https://doi.org/10.1046/j.1469 -\n8137.1997.00671.x \nIntergovernmental Panel on Climate Change (IPCC). (2023). Climate Change 2022 – Impacts, Adaptation and \nVulnerability: Working Group II Contribution to the Sixth Assessment Report of the Intergovernmental \nPanel on Climate Change (1st ed.). Cambridge University Press. https://doi.org/10.1017/9781009325844 \nIsaac, M. E., Timmer, V . R., & Quashie-Sam, S. J. (2007). Shade tree effects in an 8 -year-old cocoa agroforestry \nsystem: Biomass and nutrient diagnosis of Theobroma cacao by vector analysis. Nutrient Cycling in \nAgroecosystems, 78(2), 155–165. https://doi.org/10.1007/s10705-006-9081-3 \nIsraeli, Y ., Plaut, Z., & Schwartz, A. (1995). Effect of shade on banana morphology, growth and production. \nScientia Horticulturae, 62(1–2), 45–56. https://doi.org/10.1016/0304-4238(95)00763-J \nJanse-Ten Klooster, S. H., Thomas, E. J. P., & Sterck, F. J. (2007). Explaining interspecific differences in sapling \ngrowth and shade tolerance in temperate forests. Journal of Ecology , 95(6), 1250 –1260. \nhttps://doi.org/10.1111/j.1365-2745.2007.01299.x \nJennings, S., Brown, N., & Sheil, D. (1999). Assessing forest canopies and understorey illumination: Canopy \nclosure, canopy cover and other measures. Forestry: An International Journal of Forest Research, 72(1), \n59–74. https://doi.org/10.1093/forestry/72.1.59 \nJha, S., Bacon, C. M., Philpott, S. M., Ernesto Méndez, V ., Läderach, P ., & Rice, R. A. (2014). Shade Coffee: \nUpdate on a Disappearing Refuge for Biodiversity. BioScience, 64(5), 416 –428. \nhttps://doi.org/10.1093/biosci/biu038 \nJilo, B. F., Tiruneh, Z. A., & Fida, G. T. (2021). On-farm demonstration of homegarden agroforestry design and \nits role in improving livelihood of small holder farmers at West Arsi Zone, Oromia, Ethiopia. American \nJournal of Science, Engineering and Technology , 6(3), 59 –63. \nhttps://doi.org/10.11648/j.ajset.20210603.12 \nKassie, G. W. (2017). Agroforestry and farm income diversification: Synergy or trade -off? The case of Ethiopia. \nEnvironmental Systems Research, 6(1), 8. https://doi.org/10.1186/s40068-017-0085-6 \nKessler, J.-J. (1994). The Influence of Trees in Parklands on Sorghum Yields. In P. C. Struik, W. J. Vredenberg, J. \nA. Renkema, & J. E. Parlevliet (Eds.), Plant Production on the Threshold of a New Century: Proceedings \nof the International Conference at the Occasion of the 75th Anniversary of the Wageningen Agricultural \nUniversity, Wageningen, The Netherlands, held June 28 – July 1, 1993  (pp. 419 –421). Springer \nNetherlands. https://doi.org/10.1007/978-94-011-1158-4_52 \nKidane, S. A., Haukeland, S., Meressa, B. H., Hvoslef -Eide, A. K., & Coyne, D. L. (2021). Planting material of \nenset (Ensete ventricosum), a key food security crop in Southwest Ethiopia, ss a key element in the \ndissemination of plant -parasitic nematode infection. Frontiers in Plant Science , 12:664155. \nhttps://doi.org/10.3389/fpls.2021.664155 \nKivlin, S. N., Hawkes, C. V ., & Treseder, K. K. (2011). Global diversity and distribution of arbuscular mycorrhizal \nfungi. Soil Biology and Biochemistry, 43(11), 2294–2303. https://doi.org/10.1016/j.soilbio.2011.07.012 \nKoch, O., Mengesha, W. A., Pironon, S., Pagella, T., Ondo, I., Rosa, I., Wilkin, P., & Borrell, J. S. (2021). \nModelling potential range expansion of an underutilised food security crop in Sub -Saharan Africa. \nEnvironmental Research Letters, 17(1), 014022. https://doi.org/10.1088/1748-9326/ac40b2 \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n18 \n \nKohl, T., Niether, W., & Abdulai, I. (2024). Impact of common shade tree species on microclimate and cocoa \ngrowth in agroforestry systems in Ghana. Agroforestry Systems , 98(6), 1579 –1590. \nhttps://doi.org/10.1007/s10457-024-01029-z \nKudama, G., Tolera, T., & Gebeyehu, L. (2022). Good farm practices and improved processing technology of \nenset for sustainable hunger solution in Ethiopia. Journal of Innovation and Entrepreneurship, 11(1), 17. \nhttps://doi.org/10.1186/s13731-022-00210-x \nKuria, A., Dawit, S., Endale, Y ., Derero, A., Iiyama, Muthuri, C., Hadgu, K., Kinuthia, R., Betemariam, E., \nAsantewaa, M., Muriuki, J., Kindt, R., Dedefo, K., Senbeto, M., Njenga, M., Lamond, J., Pagella, T., & \nSinclair, F. (2017). Suitable tree species selection and management tool for Ethiopia [Data base]. World \nAgroforestry Centre (ICRAF). https://apps.worldagroforestry.org/suitable-tree/ethiopia \nKuyah, S., Whitney, C. W., Jonsson, M., Sileshi, G. W., Öborn, I., Muthuri, C. W., & Luedeling, E. (2019). \nAgroforestry delivers a win-win solution for ecosystem services in sub-Saharan Africa. A meta-analysis. \nAgronomy for Sustainable Development, 39(5), 1–18. https://doi.org/10.1007/s13593-019-0589-8 \nLalejini, A., Ferguson, A. J., Grant, N. A., & Ofria, C. (2021). Adaptive phenotypic plasticity stabilizes evolution \nin fluctuating environments. Frontiers in Ecology and Evolution , 9, 1 –16. \nhttps://doi.org/10.3389/fevo.2021.715381 \nLelamo, L. L. (2021). A review on the indigenous multipurpose agroforestry tree species in Ethiopia: \nManagement, their productive and service roles and constraints. Heliyon, 7(9), 1 –8. \nhttps://doi.org/10.1016/j.heliyon.2021.e07874 \nLemenih, M., Gidyelew, T., & Teketay, D. (2004). Effects of canopy cover and understory environment of tree \nplantations on richness, density, and size of colonizing woody species in Southern Ethiopia. Forest \nEcology and Management, 194(1), 1–10. https://doi.org/10.1016/j.foreco.2004.01.050 \nLin, B. B. (2007). Agroforestry management as an adaptive strategy against potential microclimate extremes in \ncoffee agriculture. Agricultural and Forest Meteorology , 144(1), 85 –94. \nhttps://doi.org/10.1016/j.agrformet.2006.12.009 \nLiu, Y ., Dawson, W., Prati, D., Haeuser, E., Feng, Y ., & van Kleunen, M. (2016). Does greater specific leaf area \nplasticity help plants to maintain a high performance when shaded? Annals of Botany , 118(7), 1329–\n1336. https://doi.org/10.1093/aob/mcw180 \nMabhaudhi, T., Chimonyo, V . G. P., Hlahla, S., Massawe, F., Mayes, S., Nhamo, L., & Modi, A. T. (2019). \nProspects of orphan crops in climate change. Planta, 250(3), 695–708. https://doi.org/10.1007/s00425-\n019-03129-y \nMellisse, B. T., van de Ven, G. W. J., Giller, K. E., & Descheemaeker, K. (2018). Home garden system dynamics \nin Southern Ethiopia. Agroforestry Systems , 92(6), 1579 –1595. https://doi.org/10.1007/s10457 -017-\n0106-5 \nMerle, I., Villarreyna -Acuña, R., Ribeyre, F., Roupsard, O., Cilas, C., & Avelino, J. (2022). Microclimate \nestimation under different coffee-based agroforestry systems using full -sun weather data and shade tree \ncharacteristics. European Journal of Agronomy, 132, 126396. https://doi.org/10.1016/j.eja.2021.126396 \nMolla, T., Asfaw, Z., Muluneh, M. G., & Worku, B. B. (2023). Diversity of woody species in traditional \nagroforestry practices in Wondo district, south -central Ethiopia. Heliyon, 9(2), 1 –18. \nhttps://doi.org/10.1016/j.heliyon.2023.e13549 \nMpiira, S., Staver, C., Kagezi, G. H., Wesiga, J., Nakyeyune, C., Ssebulime, G., Kabirizi, J., Nowakunda, K., \nKaramura, E., & Tushemereirwe, W. K. (2013). The use of trees and shrubs to improve banana \nproductivity and production in central Uganda: An analy sis of the current situation. In Banana systems \nin the humid highlands of sub -Saharan Africa: Enhancing resilience and productivity  (pp. 150–157). \nhttps://doi.org/10.1079/9781780642314.0150 \nMuhidin, Nurmas, A., Sadimantara, G., Leomo, S., & Yusuf, D. N. (2021). The growth performance of dwarf \nbanana Cavendish from SE Sulawesi under natural shading. IOP Conference Series: Earth and \nEnvironmental Science, 807(4), 1–4. https://doi.org/10.1088/1755-1315/807/4/042038 \nMuthuri, C. W., Kuyah, S., Njenga, M., Kuria, A., Öborn, I., & van Noordwijk, M. (2023). Agroforestry’s \ncontribution to livelihoods and carbon sequestration in East Africa: A systematic review. Trees, Forests \nand People, 14, 100432. https://doi.org/10.1016/j.tfp.2023.100432 \nNational Research Council. (2006). Lost Crops of Africa: Volume II: Vegetables. The National Academies Press. \nhttps://doi.org/10.17226/11763 \nNegash, A. (2001). Diversity and conservation of enset (Ensete ventricosum Welw. Cheesman) and its relation to \nhousehold food and livelihood security in south-western Ethiopia. 247. https://doi.org/10.18174/139427 \nNegash, M., & Starr, M. (2013). Litterfall production and associated carbon and nitrogen fluxes of seven woody \nspecies grown in indigenous agroforestry systems in the south -eastern Rift Valley escarpment of \nEthiopia. Nutrient Cycling in Agroecosystems, 97(1), 29–41. https://doi.org/10.1007/s10705-013-9590-\n9 \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n19 \n \nNiether, W., Armengot, L., Andres, C., Schneider, M., & Gerold, G. (2018). Shade trees and tree pruning alter \nthroughfall and microclimate in cocoa (Theobroma cacao L.) production systems. Annals of Forest \nScience, 75(2), 1–16. https://doi.org/10.1007/s13595-018-0723-9 \nNiether, W., Jacobi, J., Blaser, W. J., Andres, C., & Armengot, L. (2020). Cocoa agroforestry systems versus \nmonocultures: A multi -dimensional meta -analysis. Environmental Research Letters , 15(10), 1 –12. \nhttps://doi.org/10.1088/1748-9326/abb053 \nNürnberger, B. (2013). Ecological Genetics. In S. M. Scheiner (Ed.), Encyclopedia of Biodiversity (Third Edition) \n(pp. 436–455). Academic Press. https://doi.org/10.1016/B978-0-12-822562-2.00218-8 \nPoorter, H., Niinemets, Ü., Poorter, L., Wright, I. J., & Villar, R. (2009). Causes and consequences of variation in \nleaf mass per area (LMA): A meta -analysis. New Phytologist , 182(3), 565 –588. \nhttps://doi.org/10.1111/j.1469-8137.2009.02830.x \nR Core Team. (2024). R: A Language and Environment for Statistical Computing.  (Version 4.3.3) [Computer \nsoftware]. R Foundation for   Statistical Computing. \nSahle, M., Saito, O., & Demissew, S. (2022). Characterization and mapping of enset -based home -garden \nagroforestry for sustainable landscape management of the Gurage socioecological landscape in Ethiopia. \nEnvironmental Science and Pollution Research International , 29(17), 24894 –24910. \nhttps://doi.org/10.1007/s11356-021-17605-0 \nSalazar-Díaz, R., & Tixier, P. (2019). Effect of plant diversity on income generated by agroforestry systems in \nTalamanca, Costa Rica. Agroforestry Systems , 93(2), 571 –580. https://doi.org/10.1007/s10457 -017-\n0151-0 \nSánchez-Gómez, D., Valladares, F., & Zavala, M. A. (2006). Functional traits and plasticity in response to light in \nseedlings of four Iberian forest tree species. Tree Physiology , 26(11), 1425 –1433. \nhttps://doi.org/10.1093/treephys/26.11.1425 \nScordia, D., Corinzia, S. A., Coello, J., Vilaplana Ventura, R., Jiménez -De-Santiago, D. E., Singla Just, B., \nCastaño-Sánchez, O., Casas Arcarons, C., Tchamitchian, M., Garreau, L., Emran, M., Mohamed, S. Z., \nKhedr, M., Rashad, M., Lorilla, R. S., Parizel, A., Mancini, G., Iurato, A., Ponsá, S., … Testa, G. (2023). \nAre agroforestry systems more productive than monocultures in Mediterranean countries? A meta -\nanalysis. Agronomy for Sustainable Development , 43:73(6), 1–18. https://doi.org/10.1007/s13593-023-\n00927-3 \nSeidelmann, K. N., Scherer-Lorenzen, M., & Niklaus, P. A. (2016). Direct vs. Microclimate-driven effects of tree \nspecies diversity on litter decomposition in young subtropical forest stands. PLoS ONE, 11(8), 1–16. \nhttps://doi.org/10.1371/journal.pone.0160569 \nSenbeta, A. F., Sime, G., & Struik, P. (2022). Enset farming system – a resilient, climate-robust production system \nin South and South -Western Ethiopia. Cogent Food & Agriculture , 8(1), 2074626. \nhttps://doi.org/10.1080/23311932.2022.2074626 \nSenevirathna, A. M. W. K., Stirling, C. M., & Rodrigo, V . H. L. (2003). Growth, photosynthetic performance and \nshade adaptation of rubber (Hevea brasiliensis) grown in natural shade. Tree Physiology, 23(10), 705–\n712. https://doi.org/10.1093/treephys/23.10.705 \nSenevirathna, W., Stirling, C., & Rodrigo, L. (2008). Acclimation of photosynthesis and growth of banana (Musa \nsp.) to natural shade in the humid Tropics. Experimental Agriculture , 44, 301 –312. \nhttps://doi.org/10.1017/S0014479708006364 \nShalishe, A., Bhowmick, A., & Elias, K. (2023). Agricultural drought analysis and its association among land \nsurface temperature, soil moisture and precipitation in Gamo Zone, Southern Ethiopia: A remote sensing \napproach. Natural Hazards, 117(1), 57–70. https://doi.org/10.1007/s11069-023-05849-7 \nShara, S., Swennen, R., Deckers, J., Weldesenbet, F., Vercammen, L., Eshetu, F., Woldeyes, F., Blomme, G., \nMerckx, R., & Vancampenhout, K. (2021). Altitude and management affect soil fertility, leaf nutrient \nstatus and Xanthomonas wilt prevalence in enset gardens. SOIL, 7(1), 1–14. https://doi.org/10.5194/soil-\n7-1-2021 \nSharmin, M., Tjoelker, M. G., Pfautsch, S., Esperón -Rodriguez, M., Rymer, P. D., & Power, S. A. (2023). Tree \ntraits and microclimatic conditions determine cooling benefits of urban trees. Atmosphere, 14(3), 606. \nhttps://doi.org/10.3390/atmos14030606 \nSnowdon, P., Raison, J., Keith, H., Montagu, K., Bi, H., Ritson, P., Grierson, P., Adams, M., Burrows, W., & \nEamus, D. (2001). Protocol for sampling tree and stand biomass. The University of Western Australia. \nSommer, R. J. (2020). Phenotypic Plasticity: From theory and genetics to current and future challenges. Genetics, \n215(1), 1–13. https://doi.org/10.1534/genetics.120.303163 \nSpeak, A., Montagnani, L., Wellstein, C., & Zerbe, S. (2020). The influence of tree traits on urban ground surface \nshade cooling. Landscape and Urban Planning , 197, 103748. \nhttps://doi.org/10.1016/j.landurbplan.2020.103748 \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n20 \n \nTadele, Z., Farrant, J. M., Bull, S. E., & Mumm, R. H. (2024). Editorial: Orphan crops: breeding and biotechnology \nfor sustainable agriculture, food and nutrition. Frontiers in Plant Science , 14. \nhttps://doi.org/10.3389/fpls.2023.1349215 \nTalabi, A. O., Vikram, P., Thushar, S., Rahman, H., Ahmadzai, H., Nhamo, N., Shahid, M., & Singh, R. K. (2022). \nOrphan Crops: A best fit for dietary enrichment and diversification in highly deteriorated marginal \nenvironments. Frontiers in Plant Science, 13:839704. https://doi.org/10.3389/fpls.2022.839704 \nTedersoo, L., Bahram, M., & Zobel, M. (2020). How mycorrhizal associations drive plant population and \ncommunity biology. Science, 367(6480), eaba1223. https://doi.org/10.1126/science.aba1223 \nThomas, D. S., & Turner, D. W. (2001). Banana ( Musa sp.) leaf gas exchange and chlorophyll fluorescence in \nresponse to soil drought, shading and lamina folding. Scientia Horticulturae , 90(1), 93 –108. \nhttps://doi.org/10.1016/S0304-4238(00)00260-0 \nTorres, B., Maza, O. J., Aguirre, P., Hinojosa, L., & Günter, S. (2021). Contribution of traditional agroforestry to \nclimate change adaptation in the Ecuadorian Amazon: The Chakra system. In W. Leal Filho (Ed.), \nHandbook of Climate Change Adaptation  (pp. 1 –19). Springer. https://doi.org/10.1007/978 -3-642-\n40455-9_102-1 \nTschora, H., & Cherubini, F. (2020). Co-benefits and trade-offs of agroforestry for climate change mitigation and \nother sustainability goals in West Africa. Global Ecology and Conservation , 22, e00919. \nhttps://doi.org/10.1016/j.gecco.2020.e00919 \nTsegaye, A., & Struik, P. C. (2001). Enset (Ensete ventricosum (Welw.) Cheesman) kocho yield under different \ncrop establishment methods as compared to yields of other carbohydrate -rich food crops. NJAS: \nWageningen Journal of Life Sciences, 49(1), 81–94. https://doi.org/10.1016/S1573-5214(01)80017-8 \nValladares, F., & Niinemets, Ü. (2008). Shade Tolerance, a Key Plant Feature of Complex Nature and \nConsequences. Annual Review of Ecology, Evolution, and Systematics , 39(V olume 39, 2008), 237–257. \nhttps://doi.org/10.1146/annurev.ecolsys.39.110707.173506 \nValladares, F., Zaragoza-Castells, J., Sánchez-Gómez, D., Matesanz, S., Alonso, B., Portsmuth, A., Delgado, A., \n& Atkin, O. K. (2008). Is Shade Beneficial for Mediterranean Shrubs Experiencing Periods of Extreme \nDrought and Late -winter Frosts? Annals of Botany , 102(6), 923 –933. \nhttps://doi.org/10.1093/aob/mcn182 \nVendramini, F., Díaz, S., Gurvich, D. E., Wilson, P. J., Thompson, K., & Hodgson, J. G. (2002). Leaf traits as \nindicators of resource -use strategy in floras with succulent species. New Phytologist, 154(1), 147–157. \nhttps://doi.org/10.1046/j.1469-8137.2002.00357.x \nVenkatesan, L., Muzemil, S., Fiche, F., Grant, M., & Studholme, D. J. (2022). Genome resources for Ensete \nventricosum (Enset) and related species. In M. A. Chapman (Ed.), Underutilised Crop Genomes  (pp. \n355–371). Springer International Publishing. https://doi.org/10.1007/978-3-031-00848-1_19 \nVillar, R., Marañón, T., Quero, J. L., Panadero, P., Arenas, F., & Lambers, H. (2005). Variation in relative growth \nrate of 20 Aegilops species (Poaceae) in the field: The importance of net assimilation rate or specific leaf \narea depends on the time scale. Plant and Soil, 272(1), 11–27. https://doi.org/10.1007/s11104-004-3846-\n8 \nWest-Eberhard, M. J. (2008). Phenotypic Plasticity. In S. E. Jørgensen & B. D. Fath (Eds.), Encyclopedia of \nEcology (pp. 2701–2707). Academic Press. https://doi.org/10.1016/B978-008045405-4.00837-5 \nWild, J., Kopecký, M., Macek, M., Šanda, M., Jankovec, J., & Haase, T. (2019). Climate at ecologically relevant \nscales: A new temperature and soil moisture logger for long -term microclimate measurement. \nAgricultural and Forest Meteorology, 268, 40–47. https://doi.org/10.1016/j.agrformet.2018.12.018 \nWolka, K., Biazin, B., Martinsen, V ., & Mulder, J. (2021). Soil organic carbon and associated soil properties in \nEnset (Ensete ventricosum Welw. Cheesman)-based homegardens in Ethiopia. Soil and Tillage Research, \n205, 104791. https://doi.org/10.1016/j.still.2020.104791 \nWright, I. J., Reich, P. B., Westoby, M., Ackerly, D. D., Baruch, Z., Bongers, F., Cavender-Bares, J., Chapin, T., \nCornelissen, J. H. C., Diemer, M., Flexas, J., Garnier, E., Groom, P. K., Gulias, J., Hikosaka, K., Lamont, \nB. B., Lee, T., Lee, W., Lusk, C., … Villar, R. (2004). The worldwide leaf economics spectrum. Nature, \n428(6985), 821–827. https://doi.org/10.1038/nature02403 \nYang, T., Duan, Z. P., Zhu, Y ., Gan, Y . W., Wang, B. J., Hao, X. D., Xu, W. L., Zhang, W., & Li, L. H. (2019). \nEffects of distance from a tree line on photosynthetic characteristics and yield of wheat in a jujube \ntree/wheat agroforestry system. Agroforestry Systems , 93(4), 1545 –1555. \nhttps://doi.org/10.1007/s10457-018-0267-x \nYemata, G. (2020). Ensete ventricosum: A Multipurpose Crop against Hunger in Ethiopia. \nTheScientificWorldJournal, 2020, 6431849. https://doi.org/10.1155/2020/6431849 \nYemataw, Z., Bekele, A., Blomme, G., Muzemil, S., Tesfaye, K., & Jacobsen, K. (2018). A review of enset [Ensete \nventricosum (Welw.) Cheesman] diversity and its use in Ethiopia . 301 –309. \nhttps://doi.org/10.17660/th2018/73.6.1 \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n21 \n \nZellweger, F., De Frenne, P ., Lenoir, J., Vangansbeke, P ., Verheyen, K., Bernhardt-Römermann, M., Baeten, L., \nHédl, R., Berki, I., Brunet, J., Van Calster, H., Chudomelová, M., Decocq, G., Dirnböck, T., Durak, T., \nHeinken, T., Jaroszewicz, B., Kopecký, M.,  Máliš, F., … Coomes, D. (2020). Forest microclimate \ndynamics drive plant responses to warming. Science, 368(6492), 772 –775. \nhttps://doi.org/10.1126/science.aba6880 \nZewdie, S., Fetene, M., & Olsson, M. (2008). Fine root vertical distribution and temporal dynamics in mature \nstands of two enset (Enset ventricosum Welw Cheesman) clones. Plant and Soil , 305(1), 227 –236. \nhttps://doi.org/10.1007/s11104-008-9554-z \nZhang, S., Landuyt, D., Verheyen, K., & De Frenne, P. (2022). Tree species mixing can amplify microclimate \noffsets in young forest plantations. Journal of Applied Ecology , 59(6), 1428 –1439. \nhttps://doi.org/10.1111/1365-2664.14158 \nSupplementary materials \nTable S1 . Offsets of the air temperature at 15cm (Max AiT) , the soil surface temperature (Max SuT),  soil \ntemperature at a depth of -8cm (Max SoT), and soil volumetric moisture content (Min VMC) as influenced by five \nwoody tree species sparsely planted in enset homegardens of Gamo highland of Ethiopia.  \nMicroclimate offsets \nTree species \nC.macrostachyus E. brucei F .sur H. abyssinica C.africana \nMax AiT offset (oC) -0.9±0.1b -1.1±0.1b -1.9±0.1c -0.5±0.1a -1.1±0.1b \nMax SuT offset (oC) -1.0±0.1b -1.3±0.1bc -2.1±0.1d -0.4±0.1a -1.5±0.1c \nMax SoT offset (oC) -0.8±0.1bc -0.6±0.1b -1.0±0.8c 0.4±0.1a -0.9±0.1c \nMinVSM offset (%) 3.4±0.3b 5.7±0.3a 2.5±0.3b 0.8±0.4c 2.9±0.2b \nAll values are given as mean ±SE, and different letters in the same row indicate significant mean \ndifferences among tree species.   \nTable S2. Enset leaf morphophysiological features as influenced by tree canopy coverage in enset homegardens \nof Gamo highlands of Ethiopia. \nVariable \nCanopy position (radial distance) \nMiddle Edge  Outside  \nLeaf moisture content (%) 85.5±0.25a 85.3±0.25ab 84.5±0.25b \nLeaf dry matter content (%) 14.5±0.47a 14.7±0.25ab 15.5+0.25b \nSpecific leaf area (cm2/g) 229±4.8a 220±4.8a 200±4.8b \nLeaf chlorophyll content (CCI) 18.7±0.69a 17.5±0.69ab 17.0±0.69b \nFv/Fm 0.82±0.004a 0.81±0.004a 0.79±0.004b \nPIabs 26.0±1.01a 23.4±1.01b 20.0±1.01c \nAll values are given as mean ±SE, and different letters in the same row indicate significant mean differences \nbetween planting positions relative to trees (radial distance). Fv/Fm: maximum quantum efficiency of \nphotosystem II; Pi: performance index  \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n22 \n \n \nFig S1. Offsets of extreme air, surface, and soil temperatures, as well as volumetric soil moisture content, as \ninfluenced by a change in the crown area for five tree species sparsely planted in enset plots of homegardens of \nthe Gamo highlands of Ethiopia. VMC: volumetric moisture content \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint \n\n23 \n \n  \nFig S2. Microclimate offsets (maximum air, surface, and soil temperature, and soil volumetric moisture \ncontent) throughout the monitoring period (February 1, 2024, to July 31, 2024). AiT: Air temperature; \nSuT: Surface temperature; SoT: Soil temperature; and SVMC: Soil volumetric moisture content \n \n.CC-BY 4.0 International licenseperpetuity. It is made available under a \npreprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in \nThe copyright holder for thisthis version posted March 25, 2026. ; https://doi.org/10.64898/2026.03.23.713702doi: bioRxiv preprint","source_license":"CC-BY-4.0","license_restricted":false}