Using near-infrared reflectance and biomass allocation patterns of maize plants (Zea mays L.) to characterize soil fertility and productivity response dynamics

Using near-infrared reflectance and biomass allocation patterns of maize plants (Zea mays L.) to characterize soil fertility and productivity response dynamics | 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 Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Using near-infrared reflectance and biomass allocation patterns of maize plants (Zea mays L.) to characterize soil fertility and productivity response dynamics Alex O. Awiti This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-1800544/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract This paper characterizes the variation in soil quality along a forest to cropland conversation chronosequence using (i) VIS/NIR spectral characteristics of leaf samples from maize plants grown on soils from a forest-cropland chronosequence and, (ii) evaluates root: shoot biomass partitioning patterns of maize plants along a chronosequence soil quality gradient. Five hundred and forty-two topsoil samples were retrieved from a forest to cropland conversation chronosequence comprising primary and secondary forest, recently converted and historically converted farmlands. About 200 grams of each of the 542 samples were used in a bioassay in which hybrid maize seeds were planted in plastic pots under controlled conditions. Plants were harvested after 14 days. Shoots and roots were separated, dried and weighed. Discriminant analysis was used to evaluate the underlying spectral differences among the dried, ground shoot samples. Root: shoot biomass allocation patterns were assessed relative to soil spectral condition classes. The difference in reflectance of leaf samples was significant (Wilk’s Lambda = 0.006; F= 18.27; p>0.0001). The reflectance of leaf samples from maize plants grown in nutrient-poor exhibited a right shift in the 0.68-0.74 µm region (red edge position); a phenomenon diagnostic of nutrient stress. Maize plants exhibited plasticity in biomass partitioning patterns consistent with the soil quality gradient defined by fertility classes. Log-transformed root-shoot ratios revealed that maize plants grown in high nutrient soils had a low root/shoot ratio. High root: shoot ratio was observed among maize plants grown in soils classified as low in nutrients. Chronosequence Soil quality gradient Near-infrared reflectance Root: shoot ratio Biomass allocation Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 1. Introduction Land degradation and the associated soil quality decline at the farm household level presents a major challenge to meeting the goal of ending hunger, which has cause-effect links to meeting the goals of ending poverty, quality education and biodiversity conservation. Sustainable management of soil resources for agricultural production and the delivery of other vital environmental services will require rapid and cost-effective methods for collection and interpretation of plant and soil data to generate diagnosis for spatially explicit preventative or remedial management intervention. However, farm and landscape-level agricultural and environmental assessments in sub-Saharan Africa are currently under-resourced and depend invariably on costly and time-consuming chemical and physical-based assays (Shepherd and Walsh 2002 ). In smallholder production systems in sub-Saharan Africa declining crop yield and its impact on livelihood outcomes are strongly correlated with soil quality decline. However, routine monitoring and robust management of soil quality are seldom advanced as a critical determinant of the increasing vulnerability of smallholder farm households to food insecurity (Stocking, 2003 ). Stocking (1988) has shown that maize grain yield decline follows a curvilinear exponential decline in Nitisols in western Kenya. Hence, understanding patterns of maize yield decline is critical to designing appropriate and timely agronomic response interventions. Moreover, a better understanding of the factors that constrain plant growth is an essential milestone to achieving higher crop productivity But can low-resource smallholder farm households access conventional laboratory-based assay methods to maintain soil quality between cropping seasons? Time and financial resources invariably present constraints to adequate sampling. The consequence is a sub-optimal sampling considering spatial coverage and sampling density to support high-resolution assessments at the farm or even landscape level. Researchers in agro-ecosystem must recognize that agricultural farmlands even at the smallholder scale are inherently heterogeneous and hence require more intensive soil sampling than is traditionally practiced under conventional field trails to improve both accuracy and precision of estimates of key measures of soil and plant condition. Such patterns of heterogeneity or fertility gradients are attributed to the inherent variability in soil types owing to landscape position and differential allocation of landscape position, and distance to the homestead hence confirming significant with-farm variability (Tittonell et al., 2013 ). Hence, sweeping advisory on soil quality management is at best inappropriate because they do not consider within-farm variability. Today, visible and near-infrared reflectance spectroscopy (VNIRS) can deal with within-farm variability. VNIRS technology permits high sampling intensity and allows high-throughput. Moreover, VNIRS is inexpensive and allows simultaneous analyses of a wide range of organic and inorganic constituents of plants and soil (Richardson et al., 2003; Stenberg, et al., 2010 , Ely et al., 2019 ). The diagnostic capability of reflectance spectroscopy has been harnessed successfully through the development of spectral libraries for characterization of soil quality (Shepherd and Walsh, 2002 ), spectral screening tests for case definition of low fertility soils (Awiti et al., 2008 ; Vagen et al., 2006 ; Cohen et al., 2004 ) and spectral detection of nutrient deficiencies in growing plants (Osborne, et al., 2002 ; Zhao et al., 2003 ; Bonifas, et al., 2005 ). Plant tissue and soil particles are composed largely of hydrogen, carbon, oxygen, and nitrogen. Thus, the absorption bands observed in reflectance spectra of soil and plant tissue are due to vibrations of C-O, C-H, N-H and O-H bonds as well as overtones, and combinations of these vibrations (Ben-Dor and Bannin, 1995 ; Curran, 1989 ). Using wavelength ranges in the visible wavelength (0.4–0.75 µm), the near-infrared (0.75–2.5 µm) and the mid-infrared (2.5–25 µm) Ben-Dor and Bannin ( 1995 ), Shepherd and Walsh, ( 2002 ), and Awiti et al. ( 2008 ) have reliably predicted (r 2 > 0.9) important attributes of soil quality, especially SOC and TN. Nitrogen is one of six macronutrients that are essential for plant growth. Nitrogen is necessary to produce protein and chlorophyll and these are essential for plant development, re-growth, reproduction and yield (Vickery, 1981 ). Leaf spectral reflectance is a function of pigment production (Gates, 1980) and is thus sensitive to soil nutrient conditions that inhibit plant growth (Carter, 1993 , Carter and Knapp, 2001 ; Zhao et al., 2005 ). Reflectance spectroscopy-based approaches can therefore provide a platform for spatially explicit, and integrated diagnostic assessment, as well as monitoring of soil nutrient status and associated plant physiological condition. Most current advances in methods for measuring, classification, and monitoring soil quality are based on the soil nutrient status, and less on plant growth and productivity measurements (Peng et al., 2019 ; Ning et al., 2018 ). Moreover, approaches that deploy an integrated approach; combining soil spectral response and foliar spectral response are even more limited. The absence of integrated approaches for soil and crop condition assessments is a key impediment to targeted management of soil fertility for agronomic and environmental goals (Shepherd and Walsh, 2007 ). Moreover, early detection and responsive fertilizer application to address within-field and site-specific nitrogen deficiency during the growing season can improve crop yield, and reduce nutrient loss and pollution of vital soil and water resources (Zhao et al., 2003 ; Wetterlind et al., 2008 ). Previous research has evaluated spectral reflectance techniques for estimating the nutrient status of growing crops by determining wavelengths or a combination of wavelengths to characterize specific nutrient deficiencies (Walburg et al., 1982 ;; Zhao et al., 2003 ; Schlemmer et al., 2005 ). Ely et al. ( 2019 ) have demonstrated the use of leaf spectroscopy to estimate key plant metabolite pools associated with source-sink balance and C-N status. Reflectance spectroscopy involving the visible and near-infrared wavelengths has relied on leaf spectral changes to characterize soil nutrient stress and associated biochemical and structural components that influence plant physiological processes and growth (Richardson et al., 2004 ). The overarching assumption here is that leaf spectral patterns can be used as correlates of the chemical makeup of leaf samples, and a reliable basis for determining the degree to which samples drawn from substrates of varying fertility are different or similar without undertaking chemical analysis. The red edge is associated with abrupt reflectance change in the 0.68–0.74 µm region of vegetation spectra due to the combined effects of strong chlorophyll absorption and leaf internal scattering. Experimental and theoretical studies show that the red edge position shifts according to changes in leaf chlorophyll content, which is in turn influenced by soil nutrient status. The red edge is a broad feature. Hence, it is characterized by its maximum slope called the red edge position defined as the wavelength of the inflexion point of the reflectance slope at the red edge (Ruiliang et al., 2003 ). The position of the red edge is diagnostic and has been used for the early detection of plant stress associated with water, disease and nutrients (Carter and Knapp, 2001 ; Fridgen and Varco 2004 ; Eitel et al., 2011 ; Malini, et al.,2022). Nitrogen is one of the most important biological elements for crop growth and yield. Hence, insufficient N supply in the soil is known to reduce crop leaf area, and photosynthetic activity, and is often a limiting nutrient in intensely cultivated tropical soils. The deficiency of N in agricultural soils reduces leaf area development and biomass production, often leading to poor crop yield (Dev and Bhardwaj, 1995). Optimal partitioning theory predicts that plants will optimize their overall growth rate and adjust their biomass partitioning patterns to obtain the most limiting resource (Davidson, 1969 ; Hilbert, 1990 ; McConnaughay and Coleman, 1994 ). Lindquist et al. (2005) showed that corn (maize) will display true plasticity and partition a larger percentage of its biomass to roots and a smaller percentage of its biomass to shoots when the N supply is limited. However, ontogenetic drift can pose problems in the interpretation of data from experiments on biomass partitioning. Using cotton plants ( Gossypium herbaceum L.) Xie, et al. ( 2012 ) have shown that biomass allocation to roots, shoots and leaves was determined by environmental factors while biomass allocation of metabolically non-active organs like stems was controlled by ontogenetic drift. Building on previous work (Awiti et al., 2008 ; Vagen et al., 2006 ) seeks to demonstrate an approach for assessment of soil quality that combines VNIRS of soil and plant shoot as well as root: shoot biomass allocation patterns. This approach permits rapid and integrated evaluation of soil and plant response to soil fertility changes along a forest-cropland chronosequence. The study has two objectives: 1) To establish the relationship between maize shoot optical qualities and biomass allocation patterns on one hand and soil quality on the other; 2) To demonstrate that VNIRS, in combination with plant physiological response, in this case, biomass allocation patterns, can facilitate rapid, inexpensive assessment and management of soil and plant health. 2. Methods 2.1 Study site The study was conducted along a forest-cropland conversation chronosequence in the Kakamega forest area. Kakamega forest in Kenya is the eastern-most remnant of the Guinea-Congolean rainforest. It is located between latitudes 0 o 10 N and 0 o 21 N; longitude: 34 o 47 E and 34 o 58 E. The region is characterized by a bimodal rainfall pattern, with the long rains occurring in March, April and May and the short rains in October, November and December. The mean rainfall is circa 2080 mm. while the mean annual temperature ranges between 18 o C and 21 o C. The soils are predominantly Nitisols (FAO­-UNESCO) or Ultisols (USDA) and are associated with Kavirondian sediments and granite. The soils also and are moderately acidic (pH 5-5.9) with predominantly clay texture (Kenya Soil Survey, 2004). Time series analysis using Landsat TM and ETM imagery (1977-2002) was used to characterize contemporary patterns of land use and land-use change. Kenya Forest Service records provided valuable validation for timelines on recent changes in land use and land cover. Historical timelines of sedentary settlement and cultivation were obtained through oral accounts of community elders. Existing ordinance maps and aerial photographs were used to validate the oral accounts of land-use change. Three chronosequence classes were identified: (i) Forest, comprising primary and secondary forest (circa 200 years ago the parts of the forest were inhabited by slash and burn cultivators); (ii) recently converted (RC) sites; converted from primary of secondary forest to cropland for 17-20 years at the time the study was conducted; and (iii) historically converted (HC) sites, converted from forest to cropland about 70 to 100 years ago. Farmer anecdotes revealed that intensive, settled and regular cultivation was going on for over seven decades years at the time of the study. 2. 2 Soil sampling Soil sampling was implemented in plots of five clusters measuring 64 hectares, in each of the three chronosequence age classes; namely primary/secondary forest, recently converted and historically converted cropland. For each cluster, 13–30 meters transects were laid out and topsoil samples were retrieved 5, 15 and 25 meters in the direction of the dominant slope. Auger holes were kept at uniform volumes, with a sampling depth of 0-20 cm. A subsample of topsoil weighing 200 grams was collected separately, for each sample, for use in the plant bioassay experiment. The remainder of the sample was air-dried and processed for laboratory and spectral analysis. The chemical and physical properties and spectral profiles of soils drawn from the three chronosequence age classes were determined and reported in Awiti, et al. (2008) and are illustrated in Table 1 and Figure 1. Table 1. Statistical summary of the soil chemical and physical properties across a forest-cropland chronosequence analysed using conventional laboratory methods. Soil Property Chronosequence Age Class Forest ( n=65) Recently Cultivated ( n=65) Historically Cultivated ( n=64) Range Mean SE Range Mean SE Range Mean SE pH 5.1-7.4 6.5a 0.05 5.1-7.6 6.2b 0.06 5.1-7.7 5.7c 0.05 Total C,gkg -1 12-73 36.8a 1.34 21-59 34.2b 0.97 9-44 18.7c 1.0 Total N, gkg -1 2.4-5.7 3.8a 0.09 1.4-5.6 3.5b 0.12 0.9-3.2 1.4c 0.08 C:N 7.6-12 9.74a 0.4 7.9-13.1 9.97a 0.53 10-16 13.5b 0.32 Exch.Ca,cmol c kg -1 4.1-23 12.8a 0.4 0.4-19 10.5b 0.54 1.5-12 5.6c 0.32 Exch.Mg,cmol c kg -1 0.8-7 2.6a 0.2 0.1-3.4 1.8b 0.07 0.3-3.2 1.1c 0.09 Exch.K,cmol c kg -1 0.1-1.7 0.47a 0.03 0.1-0.8 0.29b 0.02 0.1-0.7 0.22c 0.02 †Means in a row followed by the same letter were not significantly different at p<0.0001 pH in 1:2:5 soil water suspensions; mean clay content ranged between 35.7 and 52.4 gkg -1 , highest in forest soils and lowest in historically converted soils. Exch.Ca = Exchangeable Calcium; Exch. Mg= Exchangeable Magnesium; Exch.K = Exchangeable K; 2. 3 Plant culture and growing conditions Five hundred and forty-two pre-weighed hybrid maize (Zea mays L.) seeds of HB-1451 variety were planted in plastic pots containing topsoil retrieved from the forest, recently converted and historically converted sites. The plants were grown in a greenhouse (poly-house) constructed using a 720-gauge Clear Polythene (light transmission ca. 90 %). The pots were arranged on a wooden bed placed along the centre line of the greenhouse. Plants were irrigated with tap water at 0800hrs daily over a 14-day growing period. Each pot had five one cm diameter holes at the bottom to allow for drainage. Average ambient day and night temperatures in the greenhouse were 30 o C and 22 o C respectively during the growing period. After 14 days, all plants were harvested. The soil was separated from plant roots using a gentle stream of tap water after which the biomass was separated into roots and shoots. Fresh samples of maize root and shoot were weighed using an electronic analytical balance, Scientech™. Leaf samples were removed from the plant shoot and oven-dried for 24 hours at 60 o C. Oven-dried leaf samples were ground using a Cyclotec 1093 TM sample mill and passed through a 1-mm sieve in preparation for spectral analysis. 2.4 Spectroscopy of topsoil and dried leaf samples FieldSpec TM FR spectroradiometer (Analytical Spectral Devices Inc, Boulder, Colorado) was used to collect reflectance spectra. Spectra from 542 topsoil and dried leaf samples from the three chronosequence classes (Forest, Recently Converted and Historically Converted cropland) were measured at wavelengths between 0.35 µm to 2.5 µm, and at a spectral sampling interval of 0.01µm. Ground maize leaf samples were scanned through the bottom of a 7.4 cm diameter Duran glass Petri dish using a high-intensity source probe (Analytical Devices, Boulder CO). The spectroradiometer uses a high temperature (3000 ◦ K) tungsten filament bulb for sample illumination (Protocol after Shepherd and Walsh, 2002). Reflected light in 1-nm bandwidths between 350 and 2500 nm is collected by three internal spectrometers (350–1000, 1000–1800, and 1800–2500 nm). Instrument specifications and optical setup are described in detail by Shepherd et al. (2003) and Shepherd and Walsh (2002). To sample within dish variation, reflectance spectra were recorded at two positions to sample within dish variability. This was achieved by successively rotating the sample dish through 90 0 between readings. An average of 25 spectra was recorded at each position to minimize instrument noise It was important to reduce data volume. Therefore, relative reflectance spectra were re-sampled by selecting every 100th-micrometre value from 0.35 to 2.5 µm. More details of spectral measurements are described in Awiti et al. (2008) and Shepherd and Walsh (2002). 2. 5 Spectral Pretreatment of leaf samples Spectral pretreatment reduces the effects of factors such as measurement geometry, and the effect of particle sample size on the optical measurement on reflectance. Derivative transformation is an optimal spectral pretreatment in similar studies and has been described in detail in Shepherd and Walsh ( 2002 ), Vagen et al. ( 2006 ) and Awiti et al. ( 2008 ). To minimize variance between samples caused by grinding and optical set-up, first derivative pretreatment was applied, in addition to multiplicative scatter correction (MSC) to reduce the effects of variable sample particle sizes. Through MSC each spectrum is normalized according to the average spectrum of the calibration set, which was calculated by the regression of each spectrum with respect to the average spectrum and by removing the slope and offset effects (Awiti et al., 2008 ). Wavelengths in regions of low signal/noise ratio or displaying noise due to splicing between individual spectrometers were omitted (Analytical Spectral Devices,1997); these were 0.35–0.40 µm, 0.97–1.01 µm, 1.75 − 1.8 µm and 2.45–2.50 µm, leaving 198 wavelength predictors between 0.4 and 2.40 µm. 2.6 The red edge position (REP) The variation in the position of the chlorophyll red-edge among leaf samples from forest, recently cultivated and historically cultivated sites were examined. The first derivatives of leaf reflectance (d R /d λ ) in the red edge wavelength, were calculated based on (Dawson and Curran, 1998 ; Lamb et al., 2002 ). $${D}_{\lambda \left(i\right)}=\frac{({R}_{\lambda \left(j+1\right)}-{R}_{\lambda \left(j\right)})}{\varDelta \lambda }$$ Where \({D}_{\lambda \left(i\right)}\) is the first-difference transformation at wavelength \(i\) between \(j\) and \(\left(j+1\right)\) and \({R}_{\lambda \left(j\right)} \text{a}\text{n}\text{d} {R}_{\lambda \left(j+1\right)}\) are the reflectance at wavelength at \(j\) and \(\left(j+1\right)\) respectively and \(\varDelta \lambda\) is the difference in wavelength between \(\left(j\right)\) and \(\left(j+1\right)\) . 2.7 Discriminant analysis of maize leaf spectra Discriminant analysis (DA) was used to determine the spectral variation among dried, ground maize shoot samples grown for 14-days in a greenhouse plant bioassay experimental set-up. The soil planting media as described earlier was from forest, recently cultivated and historically cultivated sites, along a forest-cropland conversation chronosequence. The objective of DA is to determine the relationship between spectral characteristics of dried, ground maize leaf samples or the independent discriminating variable and the categorical variable, the conversation age classes described in section 2.1 . DA extracts linear combinations of the quantitative variables, also known as canonical variables, which maximizes variation between classes and minimizes the difference within chronosequence age classes (Awiti, et al., 2008 ; Levi et al.,2020). Previous applications of discriminant analysis have included classifying plant material from closely related species, subspecies or growth environments based on leaf spectral properties (Kemsley et al., 1995 ; Atkinson et al., 1997 ; Richardson et al., 2004 ) and top soil spectral differences across a forest-cropland chronosequence soil quality gradient (Awiti et al., 2008 ). Two criteria: i) The average square canonical correlation (ASCC) and; Wilk’s Lambda were used to evaluate the spectral differences among the maize leaf samples. ASCC approaches one if the different groups are well separated and is zero if the groups are distinguished. In contrast, Wilk’s Lambda has a discriminating power ranging between 0 and 1; the lower the value the higher the discriminating power. Discriminant analysis was run in PROC DISCRIM in SAS. 2.8 Biomass allocation patterns of maize across a spectrally defined fertility gradient The study examined variations in root and shoot biomass allocation patterns of maize across a spectrally defined, ordinal fertility gradient of low, medium and high fertility. The objective was to determine whether topsoil spectral characteristics were consistent with plant growth response patterns due to soil nutrient gradients associated with time since forest conversion. A proportional odds logistic regression model (McCullagh, 1980 ) was applied to uncover the inherent spectral structure of the soils from the forest-cropland chronosequence. Previous studies have used a proportional odds logistic model, to develop a soil fertility index (Vagen et al., 2006 ) and to develop an approach for classification of soil condition applicable for detection of landscape-level changes in soil quality after land use and land cover change (Awiti, et al., 2008 ). The first ten principal components score of Savitsky-Golay transformed (Fearn, 2000) topsoil spectra (explanatory variable) and categorical chronosequence age class (response variable) were used in the proportional odds model. The sum of the product of the model coefficients and the principal component scores was used to compute the log odds (logits) for each soil sample. The logits were arranged in ascending order after which the model intercepts (cut-points) were used to partition the 542 samples into three spectral ordinal classes. The model cut-points represent the inherent segmentation among the samples based on spectral characteristics which are in turn strongly correlated with soil physical, chemical and biological properties (Ben-Dor and Bannin, 1995 ; Shepherd and Walsh, 2002 ). Further details on the proportional odds logistic model and spectral condition classification are presented in Awiti et al. ( 2008 ). Three ordinal classes were identified based on the model cut-off points. The classes were designated into three fertility classes, namely: high; medium; and, low (see Table 2 ) based on key soil chemical properties (pH, TC, TN, Exch.Ca, Exch. Mg and Exch. K). Moreover, the range and mean values of key soil fertility parameters defined under high, medium, and low were consistent with and, matched the range and mean values of forest, recently cultivated and historically cultivated soils respectively (see Table 1.) Table 2 Statistical summary of the soil chemical properties in the three spectrally defined soil fertility classes; High, Medium and Low Soil Fertility Class Soil Property High ( n = 98) Medium ( n = 54) Low ( n = 42) Range Mean SE Range Mean SE Range Mean SE pH 5.1–7.6 6.45a 0.43 5.2–7.7 6.01b 0.44 5.1–6.3 5.65b 0.312 Total C,gkg − 1 20.2–73.1 35.8a 0.91 9.5–51.8 28.1b 1.05 8.8–39.3 17.5c 0.76 Total N, gkg − 1 2.4–5.7 3.68a 0.08 1.4–5.6 2.58b 0.12 0.9–3.2 1.32c 0.06 Exch.Ca,cmol c kg −1 0.4–22.9 11.8a 3.94 1.5–17.0 9.33b 3.83 1.9–11.3 4.88c 2.22 Exch.Mg,cmol c kg −1 0.6-7.0 2.47a 1.38 0.1–3.4 1.36b 0.07 0.3–2.80 0.85c 0.59 Exch.K,cmol c kg −1 0.11–1.7 0.38a 0.23 0.06–0.83 0.32b 0.16 0.11–0.7 0.21c 0.1 †Means in a row followed by the same letter were not significantly different at p < 0.0001 pH in 1:2:5 soil water suspensions; Exch.Ca = Exchangeable Calcium; Exch. Mg = Exchangeable Magnesium; Exch.K = Exchangeable K; A linear mixed-effects model was used to model the relationship between natural log-transformed root: shoot and the ordinal fertility classes (high, medium, low) as a fixed (experimental factor) effect. A natural log transformation converted the root: shoot ratio to a simple range of negative (low root mass) to positive (high root mass) scale. The mixed-effects model was implemented in S-PLUS (Insightful Corp, 2002). 3. Results And Discussion 3.1 Leaf spectral reflectance characteristics Although there were variations among leaf samples, especially in the level of each curve, all reflectance spectra had the same general shape (Fig. 2 A). Spectral patterns of leaf samples from forest, recently converted, and historically converted cropland were characterized by absorption centred at 0.4 µm, broad peak at 0.59 µm, an absorption trough at 0.67 µm, a steep increase in reflectance between 0.69–0.76 µm in the red edge. A gentle plateau was observed in the near-infrared (0.8–1.39 µm) region. The spectral region between 1.4–2.5 µm was characterized by two major water absorption bands around 1.48 and 1.94 µm. Maize leaf samples from forest sites exhibited the lowest overall reflectance in the 0.4–0.69 µm range, with the deepest absorption trough at 0.67 µm (Fig, 2B). Conversely, samples from recently converted and historically converted sites were characterized by increased reflectance in the 0.4–0.69 µm range. Samples from forest sites had the highest reflectance in the near-infrared wavelength region (0.8–1.4 µm), compared to samples from recently converted and historically converted sites (Fig. 2 C). These patterns of reflectance are consistent with those observed in plants growing under different levels of nutrient supply. For instance, Zhao et al. ( 2005 ) observed that sorghum crops receiving 100% nitrogen solution had the lowest reflectance in the 0.4–0.69 µm range and highest reflectance in the 0.8–1.39 range compared to those receiving 20% and 0% nitrogen solution. Therefore, the reflectance at 0.4–0.69 µm, 0.70–0.76 and 0.8–1.39 µm may be a reliable indicator of plant nutrient deficiencies. [Figure 2 A; 2 B; 2 C about here] A distinctive feature of the derivative of dried leaf reflectance was the sharp increase in leaf reflectance in the 0.69–0.74 µm wavelength range, called the red-edge. The derivative reflectance of maize dried leaf samples from forest and recently converted and recently converted sites exhibited a marked shift toward the longer wavelengths in the in the 0.71–0.73 µm wavelength region compared to the derivative reflectance of samples drawn from historically converted sites (Fig. 3 ). Furthermore, the differences among the derivatives of dried, ground maize leaf reflectance in the red edge region was found to be statistically significant (Wilk’s Lambda = 0.1177; F = 85.2; p > 0.0001). The average squared canonical correlation (ASCC) was 0.90. [Figure 3 about here] These observations are consistent with Fridgen and Varco ( 2004 ) and Zhao et al. ( 2005 ), who observed that leaves from plants grown under conditions of nitrogen deficiency caused red edge reflectance to shift towards shorter wavelengths compared to leaves from plants grown under conditions of adequate N supplies. Similarly, Pinar and Curran ( 1996 ) observed that the point of maximum slope of the red edge is displaced towards longer wavelengths with increasing chlorophyll concentration. The leaf spectral characteristics in the red edge are consistent with the variations in soil nutrients across the forest-cropland chronosequence, especially total C and Total N, as well as Exchangeable Ca, Mg and K (see Table 1), This study recognizes that the accuracy of locating the red edge position is sensitive to the spectral resolution of the reflectance data. The reflectance spectra were re-sampled to reduce data volume and to match it more closely to the spectral resolution of the FieldSpec™ FR spectroradiometer. However, these findings suggest that the red edge position is potentially diagnostic of soil nutrient-induced stress in the early establishment phase of maize plants. 3.2 Multivariate analysis of spectral reflectance patterns of dried, ground maize leaf The difference in dried, ground maize leaf reflectance among the chronosequence age classes was statistically significant (Wilk’s Lambda = 0.006; F = 18.27; p > 0.0001). The average squared canonical correlation (ASCC) was 0.90. Leaf samples from forest, recently converted and historically converted sites were distinctly separated on the first and second canonical variates (Fig. 4 ). Pairwise squared distances (Mahalanobis distance D ) between samples were: D 2 = 26.77; p < 0.0001 (forest to recently converted sites); D 2 = 29.37; p < 0.0001 (forest to historically converted sites); and D 2 = 9.33; p < 0.0001 (recently converted to historically converted sites). [Figure 4 about here] The results illustrate that spectral (and thus biochemical) differences are large between leaf samples from the forest and the two cultivated sites, and small between leaf samples from recently converted and historically converted sites. The Discriminant analysis provided a robust technique for estimating the degree of biochemical differences among maize leaf samples. Based on the derived canonical variates, inferences can be made about the overall spectral similarity or dissimilarity of plants/crops under varying conditions of nutrient stress. 3.3 Biomass allocation patterns and ordinal spectral fertility classes. Soil fertility classes had a significant effect (p = 0.006) on the root: shoot ratio. Table 3 shows the mixed-effects model estimates of 95% confidence intervals around means of root: shoot of 14-day-old maize among three soil fertility classes. Allocation of biomass to roots relative to shoot was high among maize plants grown in the low and medium ordinal soil fertility classes, thus yielding positive natural log (root: shoot) ratios. Conversely, plants allocated higher biomass to shoot relative to root in maize plants grown in the high soil fertility class, resulting in a negative (-0.303) ln (root: shoot) ratio. The biomass allocation pattern among the three soil fertility classes is illustrated graphically in Fig. 5 . These results are consistent with the optimal partitioning theory. Table 3 Parameter estimates of linear mixed effects model of the effects of soil spectral class on root: shoot ratio Spectral Class Root: shoot ratio Std.Error DF t-value p-value High -0.303 0.064 149 -4.75329 <.0001 Medium 0.115 0.061 149 1.902168 0.0591 Low 0.164 0.092 149 1.774892 0.078 [Figure 5 about here] Consistent with the optimal partitioning theory, plants growing under conditions of limited supplies of belowground resources will shift biomass partitioning toward more root production and less shoot production (Hilbert, 1990 ). Consistent with this theory, this study shows maize plants exhibited biomass plasticity in allocation patterns, shifting biomass allocations in a pattern that responds to soil nutrient gradients, along the forest to cropland conversation chronosequence. Variations in levels of selected soil properties among the soil fertility classes illustrate a strong nutrient gradient (see Table 2 ). For instance, “mean total C levels were 35.8 g kg − 1 , 28.1 g kg − 1 and 17.5 g kg − 1 for high, medium and low soil fertility respectively, and mean total N levels were 3.68 g kg − 1 , 2.58 g kg − 1 and 1.31g kg − 1 for high, medium, and low soil fertility respectively” ( First reported in Awiti et al.,2008). 3.4 An approach for integrated assessment of soil quality. This study demonstrates an integrated approach to soil quality assessment based on near-infrared reflectance of both leaf tissue and topsoil as well as biomass allocation patterns of young maize plants. The core of this approach is near-infrared reflectance spectroscopy, a tool for the rapid and inexpensive acquisition of high-density spatial data. The integrated assessment is related to three key decision points based on plant response namely: (i) right or left shift in the red edge position? (Ruiliang et al., 2003 ; Fridgen and Varco, 2004 ); (ii) root: shoot ratio positive or negative? (Davidson, 1969 ; Hilbert, 1990 ; McConnaughay and Coleman, 1994 ; Lindquist et al. 2005); and (iii) is the soil degraded (low fertility) or not degraded (Shepherd and Walsh, 2007 ; Awiti et al., 2008 ). The analytical efficiency, sample throughput, cost-saving and repeatability of this approach exceeds the levels that currently exist from decades of agronomic research using time-consuming and expensive wet chemistry analysis for soil and plant tissue analyses (Batten, 1998 ; Viscarra Rossel et al., 2006 ; Shepherd and Walsh, 2007 ). This approach is particularly suitable for Sub-Saharan Africa where on-station and on-farm crop response trials still need to be conducted on a large scale but where resources for these trials are often limited. This approach holds the potential for evaluation of soil quality based on a rapid assessment of crop response. This provides opportunities for precise targeting of nutrient management at the farm level, over broad spatial scales across diverse land use land cover, and land management regimes. With modest resource investments, this approach can be executed across multiple sites in different agro-ecological zones to simultaneously determine soil fertility status and crop response characteristics. Shepherd and Walsh ( 2007 ) predicted that in ten years, case definitions of fertile or non-fertile soils will have been developed for agronomic and environmental purposes. Once the spectral case definition of infertile soils and the corresponding poor crop health is established, infrared spectroscopy and crop trials can then be implemented routinely to diagnose to measure prevalence, incidence and risk factors (Shepherd and Walsh, 2007 ) associated with different land use and land management scenarios. This study focused on plant response to a soil fertility gradient across a forest-cropland chronosequence. However, spectral and biomass response trials to specific soil nutrient deficiencies, especially those that constrain agricultural productivity, such as N, P, and K need to be conducted to better characterize plant nutrient stress and target management response to alleviate specific nutrient constraints. This approach, therefore, offers scope for a reliable and inexpensive approach to precision, spatially explicit management of soil nutrients in rural smallholder systems in sub-Saharan Africa, where resources are scarce. This includes the potential for within season assessment of plant and soil conditions and site/crop-specific management, such as targeted application of fertilizer/organic amendments, and water management. Judicious use of agricultural inputs is essential to balance the factors of soil nutrient depletion, crop nutrient requirements, increasing cost of mineral fertilizer, and the need to minimize environmental impacts of inappropriate fertilizer use and application. 4. Conclusion This study demonstrates that diffuse infrared reflectance responds to plant tissue biochemical composition as well as soil physical and chemical properties and can therefore support an integrated approach for evaluating changes in soil quality and the corresponding plant productivity responses. The study shows that the relative red edge position was generally diagnostic of nutrient-induced stress in young maize plants. Moreover, 14-day-old maize plants exhibited plasticity in biomass allocation patterns, responding optimally to soil nutrient concentrations by shifting biomass allocations in response to soil nutrient gradients, along the forest-cropland conversation chronosequence. Rapid screening of plant and soil health using reflectance spectroscopy techniques holds great promise for guiding spatially explicit, high precision soil nutrient management and plant performance monitoring in low resource settings such as sub-Saharan Africa. This is critical for targeting fertilizer application and reducing the risk of environmental pollution owing to inappropriate and excessive application of fertilizers. Future efforts should be directed to the development of an integrated diagnostic and predictive models based on spectral characteristics and biomass allocation patterns to develop and deploy rapid approaches for testing and monitoring soil quality parameters, within growing season plant responses, at the farm and landscape level, which are important for agricultural productivity and the maintenance of vital ecosystem functions and services. Moreover, while the study recognizes that real-time NIRS applications are now available for fresh leaves, their use is still greatly limited by low investments in spectral devices in sub-Saharan Africa. More research and innovation will be needed to develop affordable and portable devices that could be used in smallholder agricultural systems, where extension and scientific support services are also the weakest. Declarations Acknowledgements I am grateful to the BIOTA Africa project for logistical support in the field. Also thank ou field and laboratory team; Elvis Weullow, Sila Andrew, Wilson Ondiala, Isaac Learamo, Luka Anjeho. Funding This work was supported by the The World Agroforestry Centre (ICRAF), German Federal Ministry of Education and Research (BMBF) and the Rockefeller Foundation, Nairobi. References Atkinson, M.D., Jervis, A.P., Sangha, R.S., 1997. Discrimination between Betulapendula , Betula pubescens , and their hybrids using near-infrared reflectance spectroscopy. Canadian Journal of Forest Research 27, 1896-1900. https://doi.org/10.1139/x97- 141 Awiti, A.O., Walsh, M.G., Shepherd, K.D., Kinyamario, J. (2008). Soil condition classification using infrared spectroscopy: A proposition for assessment of soil condition along a tropical forest-cropland chronosequence. Geoderma 143, 73-84. https://doi.org/10.1016/j.geoderma.2007.08.021 Batten, G.D. (1998). Plant analysis using infrared near-infrared reflectance spectroscopy: the potential and limitations. Australian Journal of Experimental 38, 697-706. DOI:10.1071/EA97146 Ben-Dor, E., Bannin, A., (1995). Near infrared analysis (NIRA) is a rapid method to simultaneously evaluate several soil properties. Soil Science Society of America Journal 59, 364-372. https://doi.org/10.2136/sssaj1995.03615995005900020014x Carter, G.A., (1993) Responses of leaf reflectance to plant stress. American Journal of Botany 80, 239-243. https://doi.org/10.2307/2445346 Carter, G.A., Knapp, A.K. (2001). Leaf optical properties in higher plants: linking spectral characteristics to stress and chlorophyll concentration. American Journal of Botany 88, 677-684. PMID: 11302854. Malini, RC., Christopher, J., Das, S., Apan, A., Menzies, N.W., Chapman, S., Vincent Mellor, V., Dang, YP. (2022). Detection of Calcium, magnesium, and chlorophyll variations of wheat genotypes on sodic soils using hyperspectral red edge parameters, Environmental Technology & Innovation, Volume 27, 102469. Https://doi.org/10.1016/j.eti.2022.102469. Chang, C-W., Laird D.A., Mausbach, M.J., Hurburgh Jr. C.R., (2001). Near-Infrared reflectance spectroscopy-principal components regression analysis of soil properties. Soil Science Society of America Journal 65, 480-490. https://doi.org/10.2136/sssaj2001.652480x Chun L., Mi, G., Li, J., Chen, F., Zhang, F., (2005). Genetic analysis of maize root characteristics in response to low nitrogen stress. Plant and Soil 276, 369-382. DOI 10.1007/s11104-005-5876-2 Cohen, M.J., Shepherd, K.D., Walsh, M.G., (2004). Empirical reformulation of the universal soil loss equation for soil erosion risk assessment in a tropical watershed. Geoderma 124, 235-252.https://doi.org/10.1016/j.geoderma.2004.05.003 Coleman, J. S., McConnaughay, K. D. M., (1994). Interpreting phenotypic variation in plants. Trends in Ecology and Evolution 9, 951-954. https://doi.org/10.1016/0169-5347(94)90087-6 Curran, P.J., (1989). Remote sensing of foliar chemistry. Remote Sensing of Environment 30, 271-278. https://doi.org/10.1016/0034-4257(89)90069-2 Davidson, R. L., (1969). Effects of root/leaf temperature differentials on root/shoot ratios in some pasture grasses and clover. Annals of Botany 33, 561-569. https://www.jstor.org/stable/42908746 Dawson, TP, Curran, PJ. (1998). Technical note A new technique for interpolating the reflectance red edge position. Int. J. Remote Sensing, 19: 2133-2139. https://doi.org/10.1080/014311698214910 Dev, S. P, KKR, Bhardwaj K K R. (1995). Effect of crop wastes and nitrogen levels of biomass production and nitrogen uptake in wheat-maize sequence. Ann. Agric. Res. 16, 264–267. Ely, KS., Burnett, AC., Liberman-Cribbin, W., Serbin, SP., Rogers, A. (2019). Spectroscopy can predict key leaf traits associated with source–sink balance and carbon–nitrogen status. Journal of Experimental Botany, Vol. 70 (6): 1789-1799. DOI: 10.1093/jxb/erz061 Eitel, UH., Vierling, LA., Litvak, ME., Long, DS., Schulthess, U., Ager, AA., Krofcheck, DJ., Stoscheck, L. (2011). Broadband, red-edge information from satellites improves early stress detection in a New Mexico conifer woodland. Remote Sensing of Environment 115: 3640-3646. https://doi.org/10.1016/j.rse.2011.09.002 Fridgen, J.L., Varco, J.J. (2004). Dependency of cotton leaf nitrogen, chlorophyll, and reflectance on nitrogen and potassium availability. Agronomy Journal 96, 63-69 https://doi.org/10.2134/agronj2004.6300 Gates, D. M., Keegan H. J., Schlecter, J. C., Weidner, V. R. (1965). Spectral Properties of Plants. Applied Optics 4(1), 11-20. https://doi.org/10.1364/AO.4.000011 Hilbert, D. W. (1990). Optimization of plant root: shoot ratios and internal nitrogen concentrations. Annals of Botany 66, 91-99. https://doi.org/10.1093/oxfordjournals.aob.a088005 Jingchao, T., Sun, B., Cheng, R., Shi, Z., Luo, D., Luil, S., Centritto, M. (2019). Effects of soil nitrogen (N) deficiency on photosynthetic N-use efficiency in N-fixing and non-N-fixing tree seedlings in subtropical China. Nature, Scientific Reports, 9:4604 https://doi.org/10.1038/s41598-019-41035-1 Kemsley, E.K., Ruault, S., Wilson, R.H., (1995) . Discrimination between Coffea arabica and Coffea canephora variant robusta beans using infrared spectroscopy. Food Chemistry, 54, 321–326. https://doi.org/10.1016/0308-8146(95)00030-M Kenya Soil Survey. (2004). Characteristics of eight soil profiles in Kakamega. Miscellaneous report No. M58. Lamb, D. W., Steyn-Ross, M., Schaare, P., Hanna, M. M., Silvester W., Steyn-Ross, A., (2002). Estimating leaf nitrogen concentration in ryegrass ( Lolium spp.) pasture using the chlorophyll red-edge: theoretical modeling and experimental observations. International Journal. Remote Sensing 23, 3619-3648.,https://doi.org/10.1080/01431160110114529 Bonifas, K.D., Walters, D.T., Cassman, K.G, Lindquist, J.L. (2005). Nitrogen supply affects root: shoot ratio in corn and velvetleaf ( Abutilon theophrasti ). Weed Science 53, 670-675. DOI:10.1614/WS-05-002R.1 Levi, N, Karnieli, A, Paz-Kagan. T. (2020. Using reflectance spectroscopy for detecting land-use effects on soil quality in drylands, Soil and Tillage Research, Volume 199. https://doi.org/10.1016/j.still.2020.104571 McCullagh, P., (1980). Regression models for ordinal data. Journal of Royal Statistical Society B 42 (2), 109-142. Ning, J., Sheng, M., Yi, X., Wang, Y., Hou, Z., Zhang, Z., Gu, X. (2018). Rapid evaluation of soil fertility in tea plantation based on near-infrared spectroscopy, Spectroscopy Letters, 51:9, 463-471, DOI: 10.1080/00387010.2018.1475398 Osborne, S. L., Schepers, J. S., Francis, D. D., Schlemmer, M. R. (2002). Detection of phosphorus and nitrogen deficiencies in corn using spectral radiance measurements. Agronomy Journal 94, 1215-1221. https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1006&context=agronomy facpub Prananto, JA., Minansy, B., Weaver, T. (2021). Rapid cost-effective nutrient content analysis of cotton leaves using near-infrared spectroscopy (NIRS). Peer J 9: e11042DOI 10.7717/peerj.11042. Peng, Y., Zhao, L., Hu, Y., Wang, G., Wang, L., Liu, Z. (2019). Prediction of Soil Nutrient Contents using Visible and Near-Infrared reflectance spectroscopy. SPRS Int. J. Geo-Inf. 2019, 8, 437; doi:10.3390/ijgi8100437. Pinheiro, J.C., Bates, D.M. (2000). Mixed-Effects Models in S-PLUS. Springer, New York. Pinar, A., Curran PJ. (1996). Grass chlorophyll and the reflectance red-edge. International Journal or Remote Sensing, 17:351-357. https://doi.org/10.1080/01431169608949010 Richardson, A. D., Reeves, J.B., Gregorie, T.G. (2004). Multivariate analyses of visible/near infrared (VIS/NIR) absorbance spectra reveal underlying spectral differences among dried, ground conifer needle samples from different growth environments. New Phytologist 161, 291-301. DOI:10.1046/j.1469-8137.2003.00913.x Ruiliang P., Gong, P., Biging, G.S., Larrieu, M.R. (2003). Extraction of red edge optical parameters from hyperion data for estimation of forest leaf area index. IEEE Transactions on Geoscience and Remote Sensing Vol. 41(4), 916-921.DOI:10.1109/TGRS.2003.813555 SAS Institute Inc. (2016). SAS/STAT® 14.2 User’s Guide: Higher Performance Procedures. Cary, NC: SAS Institute Inc. Sanchez, P.A., Villachica, J.H., Bandy, D. E. 1(983). Soil fertility dynamics after clearing a tropical rainforest in Peru. Soil Science Society of America Journal 47, 1171-1178. https://doi.org/10.2136/sssaj1983.03615995004700060023x Schlemmer, M.R., Francis, D.D., Shanahan, J.F., Schepe, J.S. (2005). Remotely measuring chlorophyll content in corn leaves with differing nitrogen levels and relative water content. Agronomy Journal 97, 106-112. https://doi.org/10.2134/agronj2005.0106 Shepherd, K.D., Walsh, M.G., 2002. Development of reflectance spectral libraries for characterization of soil properties. Soil Science Society America Journal 66, 988-998. Shepherd, K.D., Palm, C.A., Gachengo, C.N., Vanlauwe, B., (2003). Rapid characterization of residue quality for soil and livestock management in tropical agroecosystems using near-infrared spectroscopy. Agronomy Journal 95, 1314–1322.DOI:10.2134/agronj2003.1314 Shepherd, K.D., Walsh, M.G., (2007). Infrared spectroscopy – enabling an evidence-based diagnostic surveillance approach to agricultural and environmental management in developing countries. Journal of Near Infrared Spectroscopy 15, 1-19.https://doi.org/10.1255/jnirs.716 Stenberg, B., Viscarra Rossel, R A., Mouazen, AM., Wetterlind, J. (2010). Visible and Near Infrared Spectroscopy in Soil Science. In Donald L. Sparks, editor: Advances in Agronomy, Vol. 107, Burlington: Academic Press, 2010, pp. 163-215 Stocking MA. (2003). Tropical Soils and Food Security: The Next 50 Years. Science, Vol 302: 1-5. DOI:10.1126/science.1088579 Tittonell, P, Muruiki, A, Klapwijk, CJShepherd, KD., Coe, R Vanlauwe, B. (2013). Soil Heterogeneity and Soil Fertility Gradients in Smallholder Farms of the East African Highland. Soil Sci. Soc. Am. J. 77:525–538. https://doi.org/10.2136/sssaj2012.0250 Vagen, T-G., Shepherd, K.D., Walsh, MG., (2006). Sensing landscape level change in soil fertility following deforestation and conversion in the highlands of Madagascar using Vis-NIR spectroscopy. Geoderma133, 281-294. https://doi.org/10.1016/j.geoderma.2005.07.014 van Noordwjik, M, Cerri, C, Woomer, P Nugroho, M. Bernoux, M. (1997). Soil carbon dynamics in the humid tropical forest zone. Geoderma 79, 187-225. https://doi.org/10.1016/S0016-7061(97)00042-6 Vickery, P. J., 1981. Pasture growth under grazing. In Grazing Animals , edited by F. H. W. Morley (Amsterdam: Elsevier), pp. 55–78. Viscarra Rossel, R.A., Walvoort, D.J.J., McBratney, A.B., Janik, L.J., Skjemstad, J.O., (2006). Visible, near-infrared or combined diffuse reflectance spectroscopy for simultaneous assessment of various soil properties. Geoderma 131, 59-75. https://doi.org/10.1016/j.geoderma.2005.03.007 Walburg, G., Bauer, M.E., Daughtry, C.S.T., Housley, T.L. (1982). Effects of nitrogen nutrition on the growth, yield and reflectance characteristics of corn canopies. Agronomy Journal 74, 677-683. https://doi.org/10.2134/agronj1982.00021962007400040020x Wetterlind, J., B. Stenberg, Johnson, A (2008). Near infrared reflectance spectroscopy compared with soil clay and organic matter content for estimating within-field variation in N uptake in cereals. Plant Soil, 302:317–327 DOI 10.1007/s11104- 007-9489-9 Xie J, Tang L, Wang Z, Xu G, Li Y. (2012). Distinguishing the Biomass Allocation Variance Resulting from Ontogenetic Drift or Acclimation to Soil Texture. PLoS ONE 7(7): e41502. doi: 10.1371/journal.pone.0041502 Zhao, D., Reddy, K. R., Kakani, V. G., Read, J. J., Carter, GA (2003). Corn ( Zea mays L.) growth, leaf pigment concentration, photosynthesis and leaf hyperspectral reflectance properties as affected by nitrogen supply. Plant Soil 257, 205-217. https://doi.org/10.1023/A:1026233732507 Zhao D., Reddy K. R., Kakani, V. G., Reddy, V. R., (2005). Nitrogen deficiency effects plant growth, leaf photosynthesis, and hyperspectral reflectance properties of Sorghum. European Journal of Agronomy 22, 391-403. https://doi.org/10.1016/j.eja.2004.06.005 Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted 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 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-1800544","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":116938382,"identity":"50c73096-5607-4f69-9b6a-350cb1f9cb8a","order_by":0,"name":"Alex O. Awiti","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAs0lEQVRIiWNgGAWjYJCCAwwGNkCKsfEA8VoOGKSBtDQQrwVozWEoTQzgbz9jePhDwXm7te2HgbbU2EQT1CJxJscA6LDbydvOJAK1HEvLbSDsqNwNYC1mB4BaGBsOE9Yif/4tSMu5ZLPzD4nUYnADbMsBO7MbxNpieOP9hwNnDJITzG4AbUkgxi9y59OSP1T8sbM3O5/+8MGHGhsivA8FiWCVCcQqBwF7UhSPglEwCkbBCAMAYIlPsbQOnJoAAAAASUVORK5CYII=","orcid":"","institution":"The Aga Khan University-Kenya","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Alex","middleName":"O.","lastName":"Awiti","suffix":""}],"badges":[],"createdAt":"2022-06-27 15:44:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-1800544/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-1800544/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":23526301,"identity":"c2ce7c38-522c-4a5f-a069-79c7472e3fd0","added_by":"auto","created_at":"2022-07-06 14:39:49","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":22636,"visible":true,"origin":"","legend":"\u003cp\u003eMean reflectance spectra of soils from Forest, Recently Converted RC) and Historically Converted sites.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-1800544/v1/1517a5a0f8fde690f75c4cbc.png"},{"id":23526304,"identity":"2686311a-7426-4183-90fc-c8bad0b0089d","added_by":"auto","created_at":"2022-07-06 14:39:49","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":45924,"visible":true,"origin":"","legend":"\u003cp\u003eMean relative reflectance spectra in the 0.4-2.45 μm range (1A), 0.4- 0.69 μm range (1B) and 0.8-1.4 μm range (1C) for dried ground maize leaf samples from forest, recently converted and historically converted sites.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-1800544/v1/4a28a24e5af42aa0f230da4d.png"},{"id":23526305,"identity":"2d0d8393-97e4-4020-85c1-3064e68c76ba","added_by":"auto","created_at":"2022-07-06 14:39:49","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":46699,"visible":true,"origin":"","legend":"\u003cp\u003eVariations of relative reflectance in (2A) maize leaf reflectance in red edge (0.69–74 μm) and (2B) the first derivative of maize leaf reflectance (d\u003cem\u003eR\u003c/em\u003e/d\u003cem\u003eλ\u003c/em\u003e) in red edge across a forest-cropland chronosequence.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-1800544/v1/63889732761ae5bc226c758d.png"},{"id":23527315,"identity":"63ae3e86-1aa1-4619-9553-2f6b6c0af692","added_by":"auto","created_at":"2022-07-06 14:44:49","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":142628,"visible":true,"origin":"","legend":"\u003cp\u003eCanonical discriminant analysis of spectral reflectance of dried ground maize leaves from forest, recently converted and historically converted sites.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-1800544/v1/5be0b3808fac8d44af1737ab.png"},{"id":23526302,"identity":"74e91dcc-c5c9-44ee-b7ec-cc5b5de3d0f2","added_by":"auto","created_at":"2022-07-06 14:39:49","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":19177,"visible":true,"origin":"","legend":"\u003cp\u003eRoot: shoot biomass allocation patterns among spectral condition classes of good, average and poor soils. Standard errors on the mean estimates are shown in Table 3.\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cp\u003e \u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-1800544/v1/d9587501266afbbbae7462cd.png"},{"id":23527338,"identity":"1d081bc2-2c3a-4e87-b1ff-fcd06fb5b0fd","added_by":"auto","created_at":"2022-07-06 14:44:53","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":683920,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-1800544/v1/31eb0450-f047-4a98-be61-f8c0158e3f8f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Using near-infrared reflectance and biomass allocation patterns of maize plants (Zea mays L.) to characterize soil fertility and productivity response dynamics","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eLand degradation and the associated soil quality decline at the farm household level presents a major challenge to meeting the goal of ending hunger, which has cause-effect links to meeting the goals of ending poverty, quality education and biodiversity conservation. Sustainable management of soil resources for agricultural production and the delivery of other vital environmental services will require rapid and cost-effective methods for collection and interpretation of plant and soil data to generate diagnosis for spatially explicit preventative or remedial management intervention. However, farm and landscape-level agricultural and environmental assessments in sub-Saharan Africa are currently under-resourced and depend invariably on costly and time-consuming chemical and physical-based assays (Shepherd and Walsh \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2002\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn smallholder production systems in sub-Saharan Africa declining crop yield and its impact on livelihood outcomes are strongly correlated with soil quality decline. However, routine monitoring and robust management of soil quality are seldom advanced as a critical determinant of the increasing vulnerability of smallholder farm households to food insecurity (Stocking, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Stocking (1988) has shown that maize grain yield decline follows a curvilinear exponential decline in Nitisols in western Kenya. Hence, understanding patterns of maize yield decline is critical to designing appropriate and timely agronomic response interventions. Moreover, a better understanding of the factors that constrain plant growth is an essential milestone to achieving higher crop productivity But can low-resource smallholder farm households access conventional laboratory-based assay methods to maintain soil quality between cropping seasons?\u003c/p\u003e \u003cp\u003eTime and financial resources invariably present constraints to adequate sampling. The consequence is a sub-optimal sampling considering spatial coverage and sampling density to support high-resolution assessments at the farm or even landscape level. Researchers in agro-ecosystem must recognize that agricultural farmlands even at the smallholder scale are inherently heterogeneous and hence require more intensive soil sampling than is traditionally practiced under conventional field trails to improve both accuracy and precision of estimates of key measures of soil and plant condition. Such patterns of heterogeneity or fertility gradients are attributed to the inherent variability in soil types owing to landscape position and differential allocation of landscape position, and distance to the homestead hence confirming significant with-farm variability (Tittonell et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Hence, sweeping advisory on soil quality management is at best inappropriate because they do not consider within-farm variability.\u003c/p\u003e \u003cp\u003eToday, visible and near-infrared reflectance spectroscopy (VNIRS) can deal with within-farm variability. VNIRS technology permits high sampling intensity and allows high-throughput. Moreover, VNIRS is inexpensive and allows simultaneous analyses of a wide range of organic and inorganic constituents of plants and soil (Richardson et al., 2003; Stenberg, et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, Ely et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). The diagnostic capability of reflectance spectroscopy has been harnessed successfully through the development of spectral libraries for characterization of soil quality (Shepherd and Walsh, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2002\u003c/span\u003e), spectral screening tests for case definition of low fertility soils (Awiti et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Vagen et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Cohen et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2004\u003c/span\u003e) and spectral detection of nutrient deficiencies in growing plants (Osborne, et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Zhao et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Bonifas, et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePlant tissue and soil particles are composed largely of hydrogen, carbon, oxygen, and nitrogen. Thus, the absorption bands observed in reflectance spectra of soil and plant tissue are due to vibrations of C-O, C-H, N-H and O-H bonds as well as overtones, and combinations of these vibrations (Ben-Dor and Bannin, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1995\u003c/span\u003e; Curran, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e1989\u003c/span\u003e). Using wavelength ranges in the visible wavelength (0.4\u0026ndash;0.75 \u0026micro;m), the near-infrared (0.75\u0026ndash;2.5 \u0026micro;m) and the mid-infrared (2.5\u0026ndash;25 \u0026micro;m) Ben-Dor and Bannin (\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1995\u003c/span\u003e), Shepherd and Walsh, (\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2002\u003c/span\u003e), and Awiti et al. (\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2008\u003c/span\u003e) have reliably predicted (r\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.9) important attributes of soil quality, especially SOC and TN. Nitrogen is one of six macronutrients that are essential for plant growth. Nitrogen is necessary to produce protein and chlorophyll and these are essential for plant development, re-growth, reproduction and yield (Vickery, \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e1981\u003c/span\u003e). Leaf spectral reflectance is a function of pigment production (Gates, 1980) and is thus sensitive to soil nutrient conditions that inhibit plant growth (Carter, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e1993\u003c/span\u003e, Carter and Knapp, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Zhao et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e2005\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eReflectance spectroscopy-based approaches can therefore provide a platform for spatially explicit, and integrated diagnostic assessment, as well as monitoring of soil nutrient status and associated plant physiological condition. Most current advances in methods for measuring, classification, and monitoring soil quality are based on the soil nutrient status, and less on plant growth and productivity measurements (Peng et al., \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Ning et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Moreover, approaches that deploy an integrated approach; combining soil spectral response and foliar spectral response are even more limited. The absence of integrated approaches for soil and crop condition assessments is a key impediment to targeted management of soil fertility for agronomic and environmental goals (Shepherd and Walsh, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Moreover, early detection and responsive fertilizer application to address within-field and site-specific nitrogen deficiency during the growing season can improve crop yield, and reduce nutrient loss and pollution of vital soil and water resources (Zhao et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Wetterlind et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePrevious research has evaluated spectral reflectance techniques for estimating the nutrient status of growing crops by determining wavelengths or a combination of wavelengths to characterize specific nutrient deficiencies (Walburg et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e1982\u003c/span\u003e;; Zhao et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Schlemmer et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). Ely et al. (\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) have demonstrated the use of leaf spectroscopy to estimate key plant metabolite pools associated with source-sink balance and C-N status. Reflectance spectroscopy involving the visible and near-infrared wavelengths has relied on leaf spectral changes to characterize soil nutrient stress and associated biochemical and structural components that influence plant physiological processes and growth (Richardson et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). The overarching assumption here is that leaf spectral patterns can be used as correlates of the chemical makeup of leaf samples, and a reliable basis for determining the degree to which samples drawn from substrates of varying fertility are different or similar without undertaking chemical analysis.\u003c/p\u003e \u003cp\u003eThe red edge is associated with abrupt reflectance change in the 0.68\u0026ndash;0.74 \u0026micro;m region of vegetation spectra due to the combined effects of strong chlorophyll absorption and leaf internal scattering. Experimental and theoretical studies show that the red edge position shifts according to changes in leaf chlorophyll content, which is in turn influenced by soil nutrient status. The red edge is a broad feature. Hence, it is characterized by its maximum slope called the red edge position defined as the wavelength of the inflexion point of the reflectance slope at the red edge (Ruiliang et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). The position of the red edge is diagnostic and has been used for the early detection of plant stress associated with water, disease and nutrients (Carter and Knapp, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Fridgen and Varco \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Eitel et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Malini, et al.,2022). Nitrogen is one of the most important biological elements for crop growth and yield. Hence, insufficient N supply in the soil is known to reduce crop leaf area, and photosynthetic activity, and is often a limiting nutrient in intensely cultivated tropical soils. The deficiency of N in agricultural soils reduces leaf area development and biomass production, often leading to poor crop yield (Dev and Bhardwaj, 1995).\u003c/p\u003e \u003cp\u003eOptimal partitioning theory predicts that plants will optimize their overall growth rate and adjust their biomass partitioning patterns to obtain the most limiting resource (Davidson, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e1969\u003c/span\u003e; Hilbert, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e1990\u003c/span\u003e; McConnaughay and Coleman, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1994\u003c/span\u003e). Lindquist et al. (2005) showed that corn (maize) will display true plasticity and partition a larger percentage of its biomass to roots and a smaller percentage of its biomass to shoots when the N supply is limited. However, ontogenetic drift can pose problems in the interpretation of data from experiments on biomass partitioning. Using cotton plants (\u003cem\u003eGossypium herbaceum\u003c/em\u003e L.) Xie, et al. (\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2012\u003c/span\u003e) have shown that biomass allocation to roots, shoots and leaves was determined by environmental factors while biomass allocation of metabolically non-active organs like stems was controlled by ontogenetic drift.\u003c/p\u003e \u003cp\u003eBuilding on previous work (Awiti et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Vagen et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2006\u003c/span\u003e) seeks to demonstrate an approach for assessment of soil quality that combines VNIRS of soil and plant shoot as well as root: shoot biomass allocation patterns. This approach permits rapid and integrated evaluation of soil and plant response to soil fertility changes along a forest-cropland chronosequence. The study has two objectives: 1) To establish the relationship between maize shoot optical qualities and biomass allocation patterns on one hand and soil quality on the other; 2) To demonstrate that VNIRS, in combination with plant physiological response, in this case, biomass allocation patterns, can facilitate rapid, inexpensive assessment and management of soil and plant health.\u003c/p\u003e"},{"header":"2. Methods","content":"\u003ch2\u003e2.1 Study site\u003c/h2\u003e\n\u003cp\u003eThe study was conducted along a forest-cropland conversation chronosequence in the Kakamega forest area. Kakamega forest in Kenya is the eastern-most remnant of the Guinea-Congolean rainforest. It is located between latitudes 0\u003csup\u003eo\u003c/sup\u003e10 N and 0\u003csup\u003eo\u003c/sup\u003e21 N; longitude: 34\u003csup\u003eo\u003c/sup\u003e47 E and 34\u003csup\u003eo\u003c/sup\u003e58 E. The region is characterized by a bimodal rainfall pattern, with the long rains occurring in March, April and May and the short rains in October, November and December. The mean rainfall is circa 2080 mm. while the mean annual temperature ranges between 18\u003csup\u003eo\u0026nbsp;\u003c/sup\u003eC and 21\u003csup\u003eo\u003c/sup\u003eC. The soils are predominantly Nitisols (FAO\u0026shy;-UNESCO) or Ultisols (USDA) and are associated with Kavirondian sediments and granite. The soils also and are moderately acidic (pH 5-5.9) with predominantly clay texture (Kenya Soil Survey, 2004).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTime series analysis using Landsat TM and ETM imagery (1977-2002) was used to characterize contemporary patterns of land use and land-use change. Kenya Forest Service records provided valuable validation for timelines on recent changes in land use and land cover. Historical timelines of sedentary settlement and cultivation were obtained through oral accounts of community elders. Existing ordinance maps and aerial photographs were used to validate the oral accounts of land-use change. Three chronosequence classes were identified: (i) Forest, comprising primary and secondary forest (circa 200 years ago the parts of the forest were inhabited by slash and burn cultivators); (ii) recently converted (RC) sites; converted from primary of secondary forest to cropland for 17-20 years at the time the study was conducted; and (iii) historically converted (HC) sites, converted from forest to cropland about 70 to 100 years ago. \u0026nbsp;Farmer anecdotes revealed that intensive, settled and regular cultivation was going on for over seven decades years at the time of the study.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e2. 2 Soil sampling\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eSoil sampling was implemented in plots of five clusters measuring 64 hectares, in each of the three chronosequence age classes; namely primary/secondary forest, recently converted and historically converted cropland. For each cluster, 13\u0026ndash;30 meters transects were laid out and topsoil samples were retrieved 5, 15 and 25 meters in the direction of the dominant slope. Auger holes were kept at uniform volumes, with a sampling depth of 0-20 cm. A subsample of topsoil weighing 200 grams was collected separately, for each sample, for use in the plant bioassay experiment. The remainder of the sample was air-dried and processed for laboratory and spectral analysis. The chemical and physical properties and spectral profiles of soils drawn from the three chronosequence age classes were determined and reported in Awiti, et al. (2008) and are illustrated in Table 1 and Figure 1.\u003c/p\u003e\n\u003cp style='color: rgb(0, 0, 0); font-family: \"Times New Roman\"; font-size: medium; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: center; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial; margin-bottom: 10px !important;'\u003eTable 1.\u0026nbsp;\u003c/p\u003e\n\u003cp style='color: rgb(0, 0, 0); font-family: \"Times New Roman\"; font-size: medium; font-style: normal; font-variant-ligatures: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: 2; text-align: center; text-indent: 0px; text-transform: none; white-space: normal; widows: 2; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration-style: initial; text-decoration-color: initial; margin-bottom: 10px !important;'\u003eStatistical summary of the soil chemical and physical properties across a forest-cropland chronosequence analysed using conventional laboratory methods.\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\" width=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" valign=\"top\" width=\"20.839580209895054%\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eSoil Property\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"9\" valign=\"top\" width=\"79.16041979010495%\"\u003e\n \u003cp\u003eChronosequence Age Class\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"3\" valign=\"top\" width=\"34.09090909090909%\"\u003e\n \u003cp\u003eForest (\u003cem\u003en=65)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" width=\"31.818181818181817%\"\u003e\n \u003cp\u003eRecently Cultivated (\u003cem\u003en=65)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"3\" valign=\"top\" width=\"34.09090909090909%\"\u003e\n \u003cp\u003eHistorically Cultivated (\u003cem\u003en=64)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"13.636363636363637%\"\u003e\n \u003cp\u003eRange\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.363636363636363%\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.090909090909092%\"\u003e\n \u003cp\u003eSE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"12.5%\"\u003e\n \u003cp\u003eRange\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.227272727272727%\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.090909090909092%\"\u003e\n \u003cp\u003eSE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"12.31060606060606%\"\u003e\n \u003cp\u003eRange\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"11.363636363636363%\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.416666666666666%\"\u003e\n \u003cp\u003eSE\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"20.839580209895054%\"\u003e\n \u003cp\u003epH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.794602698650674%\"\u003e\n \u003cp\u003e5.1-7.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e6.5a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.89505247376312%\"\u003e\n \u003cp\u003e5.1-7.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.095952023988007%\"\u003e\n \u003cp\u003e6.2b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.745127436281859%\"\u003e\n \u003cp\u003e5.1-7.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e5.7c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.245877061469265%\"\u003e\n \u003cp\u003e0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"20.839580209895054%\"\u003e\n \u003cp\u003eTotal C,gkg\u003csup\u003e-1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.794602698650674%\"\u003e\n \u003cp\u003e12-73\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e36.8a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e1.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.89505247376312%\"\u003e\n \u003cp\u003e21-59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.095952023988007%\"\u003e\n \u003cp\u003e34.2b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.97\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.745127436281859%\"\u003e\n \u003cp\u003e9-44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e18.7c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.245877061469265%\"\u003e\n \u003cp\u003e1.0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"20.839580209895054%\"\u003e\n \u003cp\u003eTotal N, gkg\u003csup\u003e-1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.794602698650674%\"\u003e\n \u003cp\u003e2.4-5.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e3.8a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.89505247376312%\"\u003e\n \u003cp\u003e1.4-5.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.095952023988007%\"\u003e\n \u003cp\u003e3.5b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.745127436281859%\"\u003e\n \u003cp\u003e0.9-3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e1.4c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.245877061469265%\"\u003e\n \u003cp\u003e0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"20.839580209895054%\"\u003e\n \u003cp\u003eC:N\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.794602698650674%\"\u003e\n \u003cp\u003e7.6-12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e9.74a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.89505247376312%\"\u003e\n \u003cp\u003e7.9-13.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.095952023988007%\"\u003e\n \u003cp\u003e9.97a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.745127436281859%\"\u003e\n \u003cp\u003e10-16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e13.5b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.245877061469265%\"\u003e\n \u003cp\u003e0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"20.839580209895054%\"\u003e\n \u003cp\u003eExch.Ca,cmol\u003csub\u003ec\u003c/sub\u003ekg\u003csup\u003e-1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.794602698650674%\"\u003e\n \u003cp\u003e4.1-23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e12.8a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.89505247376312%\"\u003e\n \u003cp\u003e0.4-19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.095952023988007%\"\u003e\n \u003cp\u003e10.5b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.745127436281859%\"\u003e\n \u003cp\u003e1.5-12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e5.6c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.245877061469265%\"\u003e\n \u003cp\u003e0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"20.839580209895054%\"\u003e\n \u003cp\u003eExch.Mg,cmol\u003csub\u003ec\u003c/sub\u003ekg\u003csup\u003e-1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.794602698650674%\"\u003e\n \u003cp\u003e0.8-7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e2.6a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.89505247376312%\"\u003e\n \u003cp\u003e0.1-3.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.095952023988007%\"\u003e\n \u003cp\u003e1.8b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.745127436281859%\"\u003e\n \u003cp\u003e0.3-3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e1.1c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.245877061469265%\"\u003e\n \u003cp\u003e0.09\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" width=\"20.839580209895054%\"\u003e\n \u003cp\u003eExch.K,cmol\u003csub\u003ec\u003c/sub\u003ekg\u003csup\u003e-1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"10.794602698650674%\"\u003e\n \u003cp\u003e0.1-1.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e0.47a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.89505247376312%\"\u003e\n \u003cp\u003e0.1-0.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.095952023988007%\"\u003e\n \u003cp\u003e0.29b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"7.19640179910045%\"\u003e\n \u003cp\u003e0.02\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"9.745127436281859%\"\u003e\n \u003cp\u003e0.1-0.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.995502248875562%\"\u003e\n \u003cp\u003e0.22c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" width=\"8.245877061469265%\"\u003e\n \u003cp\u003e0.02\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\u0026dagger;Means in a row followed by the same letter were not significantly different at \u003cem\u003ep\u0026lt;0.0001\u003c/em\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003epH in 1:2:5 soil water suspensions; mean clay content ranged between 35.7 and 52.4\u0026nbsp;gkg\u003csup\u003e-1 \u0026nbsp;\u003c/sup\u003e, highest in forest soils and lowest in historically converted soils.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eExch.Ca = Exchangeable Calcium; Exch. Mg= Exchangeable Magnesium; Exch.K = Exchangeable K;\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e2. 3 Plant culture and growing conditions\u003c/h2\u003e\n\u003cp\u003eFive hundred and forty-two pre-weighed hybrid maize (Zea mays L.) seeds of HB-1451 variety were planted in plastic pots containing topsoil retrieved from the forest, recently converted and historically converted sites. The plants were grown in a greenhouse (poly-house) constructed using a 720-gauge Clear Polythene (light transmission ca. 90 %). The pots were arranged on a wooden bed placed along the centre line of the greenhouse. Plants were irrigated with tap water at 0800hrs daily over a 14-day growing period. Each pot had five one cm diameter holes at the bottom to allow for drainage. Average ambient day and night temperatures in the greenhouse were 30\u003csup\u003eo\u003c/sup\u003eC and 22\u003csup\u003eo\u003c/sup\u003eC respectively during the growing period.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAfter 14 days, all plants were harvested. The soil was separated from plant roots using a gentle stream of tap water after which the biomass was separated into roots and shoots. Fresh samples of maize root and shoot were weighed using an electronic analytical balance, Scientech\u0026trade;. Leaf samples were removed from the plant shoot and oven-dried for 24 hours at 60\u003csup\u003eo\u003c/sup\u003eC. Oven-dried leaf samples were ground using a Cyclotec 1093\u003csup\u003e\u0026nbsp;TM\u003c/sup\u003e sample mill and passed through a 1-mm sieve in preparation for spectral analysis.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e2.4 Spectroscopy of topsoil and dried leaf samples\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eFieldSpec\u003csup\u003eTM\u003c/sup\u003e FR spectroradiometer (Analytical Spectral Devices Inc, Boulder, Colorado) was used to collect reflectance spectra. Spectra from 542 topsoil and dried leaf samples from the three chronosequence classes (Forest, Recently Converted and Historically Converted cropland) were measured at wavelengths between 0.35 \u0026micro;m to 2.5 \u0026micro;m, and at a spectral sampling interval of 0.01\u0026micro;m. Ground maize leaf samples were scanned through the bottom of a 7.4 cm diameter Duran glass Petri dish using a high-intensity source probe (Analytical Devices, Boulder CO). The spectroradiometer uses a high temperature (3000\u003csup\u003e◦\u003c/sup\u003eK) tungsten filament bulb for sample illumination (Protocol after Shepherd and Walsh, 2002). \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eReflected light in 1-nm bandwidths between 350 and 2500 nm is collected by three internal spectrometers (350\u0026ndash;1000, 1000\u0026ndash;1800, and 1800\u0026ndash;2500 nm). Instrument specifications and optical setup are described in detail by Shepherd et al. (2003) and Shepherd and Walsh (2002). To sample within dish variation, reflectance spectra were recorded at two positions to sample within dish variability. This was achieved by successively rotating the sample dish through 90\u003csup\u003e0\u003c/sup\u003e between readings. An average of 25 \u0026nbsp;spectra was recorded at each position to minimize instrument noise It was important to reduce data volume. Therefore, \u0026nbsp;relative reflectance spectra were re-sampled by selecting every 100th-micrometre value from 0.35 to 2.5 \u0026micro;m. More details of spectral measurements are described in Awiti et al. (2008) and Shepherd and Walsh (2002). \u0026nbsp;\u003c/p\u003e\n\u003ch2\u003e2. 5 Spectral Pretreatment of leaf samples\u003c/h2\u003e\n\u003cp\u003eSpectral pretreatment reduces the effects of factors such as measurement geometry, and the effect of particle sample size on the optical measurement on reflectance. Derivative transformation is an optimal spectral pretreatment in similar studies and has been described in detail in Shepherd and Walsh (\u003cspan class=\"CitationRef\"\u003e2002\u003c/span\u003e), Vagen et al. (\u003cspan class=\"CitationRef\"\u003e2006\u003c/span\u003e) and Awiti et al. (\u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e). To minimize variance between samples caused by grinding and optical set-up, first derivative pretreatment was applied, in addition to multiplicative scatter correction (MSC) to reduce the effects of variable sample particle sizes. Through MSC each spectrum is normalized according to the average spectrum of the calibration set, which was calculated by the regression of each spectrum with respect to the average spectrum and by removing the slope and offset effects (Awiti et al., \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e). Wavelengths in regions of low signal/noise ratio or displaying noise due to splicing between individual spectrometers were omitted (Analytical Spectral Devices,1997); these were 0.35\u0026ndash;0.40 \u0026micro;m, 0.97\u0026ndash;1.01 \u0026micro;m, 1.75 \u0026minus;\u0026thinsp;1.8 \u0026micro;m and 2.45\u0026ndash;2.50 \u0026micro;m, leaving 198 wavelength predictors between 0.4 and 2.40 \u0026micro;m.\u003c/p\u003e\n\u003cdiv class=\"Section2\" id=\"Sec8\"\u003e\n \u003ch2\u003e2.6 The red edge position (REP)\u003c/h2\u003e\n \u003cp\u003eThe variation in the position of the chlorophyll red-edge among leaf samples from forest, recently cultivated and historically cultivated sites were examined. The first derivatives of leaf reflectance (d\u003cem\u003eR\u003c/em\u003e/d\u003cem\u003e\u0026lambda;\u003c/em\u003e) in the red edge wavelength, were calculated based on (Dawson and Curran, \u003cspan class=\"CitationRef\"\u003e1998\u003c/span\u003e; Lamb et al., \u003cspan class=\"CitationRef\"\u003e2002\u003c/span\u003e).\u003c/p\u003e\n \u003cdiv class=\"Equation\" id=\"Equa\"\u003e\n \u003cdiv class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e$${D}_{\\lambda \\left(i\\right)}=\\frac{({R}_{\\lambda \\left(j+1\\right)}-{R}_{\\lambda \\left(j\\right)})}{\\varDelta \\lambda }$$\u003c/div\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n \u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({D}_{\\lambda \\left(i\\right)}\\)\u003c/span\u003e\u003c/span\u003e is the first-difference transformation at wavelength \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(i\\)\u003c/span\u003e\u003c/span\u003e between \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(j\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\left(j+1\\right)\\)\u003c/span\u003e\u003c/span\u003eand \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({R}_{\\lambda \\left(j\\right)} \\text{a}\\text{n}\\text{d} {R}_{\\lambda \\left(j+1\\right)}\\)\u003c/span\u003e\u003c/span\u003e are the reflectance at wavelength at \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(j\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\left(j+1\\right)\\)\u003c/span\u003e\u003c/span\u003e respectively and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\varDelta \\lambda\\)\u003c/span\u003e\u003c/span\u003e is the difference in wavelength between \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\left(j\\right)\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\left(j+1\\right)\\)\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv class=\"Section2\" id=\"Sec9\"\u003e\n \u003ch2\u003e2.7 Discriminant analysis of maize leaf spectra\u003c/h2\u003e\n \u003cp\u003eDiscriminant analysis (DA) was used to determine the spectral variation among dried, ground maize shoot samples grown for 14-days in a greenhouse plant bioassay experimental set-up. The soil planting media as described earlier was from forest, recently cultivated and historically cultivated sites, along a forest-cropland conversation chronosequence. The objective of DA is to determine the relationship between spectral characteristics of dried, ground maize leaf samples or the independent discriminating variable and the categorical variable, the conversation age classes described in section \u003cspan class=\"InternalRef\"\u003e2.1\u003c/span\u003e. DA extracts linear combinations of the quantitative variables, also known as canonical variables, which maximizes variation between classes and minimizes the difference within chronosequence age classes (Awiti, et al., \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e; Levi et al.,2020).\u003c/p\u003e\n \u003cp\u003ePrevious applications of discriminant analysis have included classifying plant material from closely related species, subspecies or growth environments based on leaf spectral properties (Kemsley et al., \u003cspan class=\"CitationRef\"\u003e1995\u003c/span\u003e; Atkinson et al., \u003cspan class=\"CitationRef\"\u003e1997\u003c/span\u003e; Richardson et al., \u003cspan class=\"CitationRef\"\u003e2004\u003c/span\u003e) and top soil spectral differences across a forest-cropland chronosequence soil quality gradient (Awiti et al., \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e). Two criteria: i) The average square canonical correlation (ASCC) and; Wilk\u0026rsquo;s Lambda were used to evaluate the spectral differences among the maize leaf samples. ASCC approaches one if the different groups are well separated and is zero if the groups are distinguished. In contrast, Wilk\u0026rsquo;s Lambda has a discriminating power ranging between 0 and 1; the lower the value the higher the discriminating power. Discriminant analysis was run in PROC DISCRIM in SAS.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv class=\"Section2\" id=\"Sec10\"\u003e\n \u003ch2\u003e2.8 Biomass allocation patterns of maize across a spectrally defined fertility gradient\u003c/h2\u003e\n \u003cp\u003eThe study examined variations in root and shoot biomass allocation patterns of maize across a spectrally defined, ordinal fertility gradient of low, medium and high fertility. The objective was to determine whether topsoil spectral characteristics were consistent with plant growth response patterns due to soil nutrient gradients associated with time since forest conversion. A proportional odds logistic regression model (McCullagh, \u003cspan class=\"CitationRef\"\u003e1980\u003c/span\u003e) was applied to uncover the inherent spectral structure of the soils from the forest-cropland chronosequence. Previous studies have used a proportional odds logistic model, to develop a soil fertility index (Vagen et al., \u003cspan class=\"CitationRef\"\u003e2006\u003c/span\u003e) and to develop an approach for classification of soil condition applicable for detection of landscape-level changes in soil quality after land use and land cover change (Awiti, et al., \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e\n \u003cp\u003eThe first ten principal components score of Savitsky-Golay transformed (Fearn, 2000) topsoil spectra (explanatory variable) and categorical chronosequence age class (response variable) were used in the proportional odds model. The sum of the product of the model coefficients and the principal component scores was used to compute the log odds (logits) for each soil sample. The logits were arranged in ascending order after which the model intercepts (cut-points) were used to partition the 542 samples into three spectral ordinal classes.\u003c/p\u003e\n \u003cp\u003eThe model cut-points represent the inherent segmentation among the samples based on spectral characteristics which are in turn strongly correlated with soil physical, chemical and biological properties (Ben-Dor and Bannin, \u003cspan class=\"CitationRef\"\u003e1995\u003c/span\u003e; Shepherd and Walsh, \u003cspan class=\"CitationRef\"\u003e2002\u003c/span\u003e). Further details on the proportional odds logistic model and spectral condition classification are presented in Awiti et al. (\u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e). Three ordinal classes were identified based on the model cut-off points. The classes were designated into three fertility classes, namely: high; medium; and, low (see Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e) based on key soil chemical properties (pH, TC, TN, Exch.Ca, Exch. Mg and Exch. K). Moreover, the range and mean values of key soil fertility parameters defined under high, medium, and low were consistent with and, matched the range and mean values of forest, recently cultivated and historically cultivated soils respectively (see Table 1.)\u003c/p\u003e\n \u003cdiv align=\"left\" class=\"colspec\"\u003e\u003cbr\u003e\u003c/div\u003e\u0026nbsp;\u0026nbsp;\u003ctable border=\"1\" id=\"Tab1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eStatistical summary of the soil chemical properties in the three spectrally defined soil fertility classes; High, Medium and Low\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colspan=\"9\"\u003e\n \u003cp\u003eSoil Fertility Class\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\u003eSoil Property\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"3\"\u003e\n \u003cp\u003eHigh (\u003cem\u003en\u0026thinsp;=\u0026thinsp;98)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"3\"\u003e\n \u003cp\u003eMedium (\u003cem\u003en\u0026thinsp;=\u0026thinsp;54)\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\" colspan=\"3\"\u003e\n \u003cp\u003eLow (\u003cem\u003en\u0026thinsp;=\u0026thinsp;42)\u003c/em\u003e\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\u003eRange\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRange\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRange\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMean\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSE\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003epH\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5.1\u0026ndash;7.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6.45a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5.2\u0026ndash;7.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6.01b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5.1\u0026ndash;6.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5.65b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.312\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTotal C,gkg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e20.2\u0026ndash;73.1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e35.8a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9.5\u0026ndash;51.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e28.1b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.05\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8.8\u0026ndash;39.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e17.5c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.76\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTotal N, gkg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.4\u0026ndash;5.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.68a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.4\u0026ndash;5.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.58b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.9\u0026ndash;3.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.32c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.06\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eExch.Ca,cmol\u003csub\u003ec\u003c/sub\u003ekg\u003csup\u003e\u0026minus;1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.4\u0026ndash;22.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11.8a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.5\u0026ndash;17.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9.33b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.9\u0026ndash;11.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4.88c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.22\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eExch.Mg,cmol\u003csub\u003ec\u003c/sub\u003ekg\u003csup\u003e\u0026minus;1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.6-7.0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.47a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.1\u0026ndash;3.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.36b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.07\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.3\u0026ndash;2.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.85c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.59\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eExch.K,cmol\u003csub\u003ec\u003c/sub\u003ekg\u003csup\u003e\u0026minus;1\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.11\u0026ndash;1.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.38a\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.06\u0026ndash;0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.32b\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.11\u0026ndash;0.7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.21c\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003ctfoot\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"10\"\u003e\u0026dagger;Means in a row followed by the same letter were not significantly different at \u003cem\u003ep\u0026thinsp;\u0026lt;\u0026thinsp;0.0001\u003c/em\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"10\"\u003epH in 1:2:5 soil water suspensions; Exch.Ca\u0026thinsp;=\u0026thinsp;Exchangeable Calcium; Exch. Mg\u0026thinsp;=\u0026thinsp;Exchangeable Magnesium; Exch.K\u0026thinsp;=\u0026thinsp;Exchangeable K;\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tfoot\u003e\n \u003c/table\u003e\n \u003cp\u003e\u003c/p\u003e\n \u003cp\u003eA linear mixed-effects model was used to model the relationship between natural log-transformed root: shoot and the ordinal fertility classes (high, medium, low) as a fixed (experimental factor) effect. A natural log transformation converted the root: shoot ratio to a simple range of negative (low root mass) to positive (high root mass) scale. The mixed-effects model was implemented in S-PLUS (Insightful Corp, 2002).\u003c/p\u003e\n\u003c/div\u003e"},{"header":"3. Results And Discussion","content":"\u003cdiv class=\"Section2\" id=\"Sec12\"\u003e\n \u003ch2\u003e3.1 Leaf spectral reflectance characteristics\u003c/h2\u003e\n \u003cp\u003eAlthough there were variations among leaf samples, especially in the level of each curve, all reflectance spectra had the same general shape (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eA). Spectral patterns of leaf samples from forest, recently converted, and historically converted cropland were characterized by absorption centred at 0.4 \u0026micro;m, broad peak at 0.59 \u0026micro;m, an absorption trough at 0.67 \u0026micro;m, a steep increase in reflectance between 0.69\u0026ndash;0.76 \u0026micro;m in the red edge. A gentle plateau was observed in the near-infrared (0.8\u0026ndash;1.39 \u0026micro;m) region. The spectral region between 1.4\u0026ndash;2.5 \u0026micro;m was characterized by two major water absorption bands around 1.48 and 1.94 \u0026micro;m.\u003c/p\u003e\n \u003cp\u003eMaize leaf samples from forest sites exhibited the lowest overall reflectance in the 0.4\u0026ndash;0.69 \u0026micro;m range, with the deepest absorption trough at 0.67 \u0026micro;m (Fig, 2B). Conversely, samples from recently converted and historically converted sites were characterized by increased reflectance in the 0.4\u0026ndash;0.69 \u0026micro;m range. Samples from forest sites had the highest reflectance in the near-infrared wavelength region (0.8\u0026ndash;1.4 \u0026micro;m), compared to samples from recently converted and historically converted sites (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eC). These patterns of reflectance are consistent with those observed in plants growing under different levels of nutrient supply. For instance, Zhao et al. (\u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e) observed that sorghum crops receiving 100% nitrogen solution had the lowest reflectance in the 0.4\u0026ndash;0.69 \u0026micro;m range and highest reflectance in the 0.8\u0026ndash;1.39 range compared to those receiving 20% and 0% nitrogen solution. Therefore, the reflectance at 0.4\u0026ndash;0.69 \u0026micro;m, 0.70\u0026ndash;0.76 and 0.8\u0026ndash;1.39 \u0026micro;m may be a reliable indicator of plant nutrient deficiencies.\u003c/p\u003e\n \u003cp\u003e[Figure \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eA; \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eB; \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eC about here]\u003c/p\u003e\n \u003cp\u003eA distinctive feature of the derivative of dried leaf reflectance was the sharp increase in leaf reflectance in the 0.69\u0026ndash;0.74 \u0026micro;m wavelength range, called the red-edge. The derivative reflectance of maize dried leaf samples from forest and recently converted and recently converted sites exhibited a marked shift toward the longer wavelengths in the in the 0.71\u0026ndash;0.73 \u0026micro;m wavelength region compared to the derivative reflectance of samples drawn from historically converted sites (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). Furthermore, the differences among the derivatives of dried, ground maize leaf reflectance in the red edge region was found to be statistically significant (Wilk\u0026rsquo;s Lambda\u0026thinsp;=\u0026thinsp;0.1177; F\u0026thinsp;=\u0026thinsp;85.2; p\u0026thinsp;\u0026gt;\u0026thinsp;0.0001). The average squared canonical correlation (ASCC) was 0.90.\u003c/p\u003e\n \u003cp\u003e[Figure \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e about here]\u003c/p\u003e\n \u003cp\u003eThese observations are consistent with Fridgen and Varco (\u003cspan class=\"CitationRef\"\u003e2004\u003c/span\u003e) and Zhao et al. (\u003cspan class=\"CitationRef\"\u003e2005\u003c/span\u003e), who observed that leaves from plants grown under conditions of nitrogen deficiency caused red edge reflectance to shift towards shorter wavelengths compared to leaves from plants grown under conditions of adequate N supplies. Similarly, Pinar and Curran (\u003cspan class=\"CitationRef\"\u003e1996\u003c/span\u003e) observed that the point of maximum slope of the red edge is displaced towards longer wavelengths with increasing chlorophyll concentration. The leaf spectral characteristics in the red edge are consistent with the variations in soil nutrients across the forest-cropland chronosequence, especially total C and Total N, as well as Exchangeable Ca, Mg and K (see Table\u0026nbsp;1),\u003c/p\u003e\n \u003cp\u003eThis study recognizes that the accuracy of locating the red edge position is sensitive to the spectral resolution of the reflectance data. The reflectance spectra were re-sampled to reduce data volume and to match it more closely to the spectral resolution of the FieldSpec\u0026trade; FR spectroradiometer. However, these findings suggest that the red edge position is potentially diagnostic of soil nutrient-induced stress in the early establishment phase of maize plants.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv class=\"Section2\" id=\"Sec13\"\u003e\n \u003ch2\u003e3.2 Multivariate analysis of spectral reflectance patterns of dried, ground maize leaf\u003c/h2\u003e\n \u003cp\u003eThe difference in dried, ground maize leaf reflectance among the chronosequence age classes was statistically significant (Wilk\u0026rsquo;s Lambda\u0026thinsp;=\u0026thinsp;0.006; F\u0026thinsp;=\u0026thinsp;18.27; p\u0026thinsp;\u0026gt;\u0026thinsp;0.0001). The average squared canonical correlation (ASCC) was 0.90. Leaf samples from forest, recently converted and historically converted sites were distinctly separated on the first and second canonical variates (Fig. \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e). Pairwise squared distances (Mahalanobis distance \u003cem\u003eD\u003c/em\u003e) between samples were: \u003cem\u003eD\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;26.77; p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001 (forest to recently converted sites); \u003cem\u003eD\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;29.37; p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001 (forest to historically converted sites); and \u003cem\u003eD\u003c/em\u003e\u003csup\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;9.33; p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001 (recently converted to historically converted sites).\u003c/p\u003e\n \u003cp\u003e[Figure \u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e about here]\u003c/p\u003e\n \u003cp\u003eThe results illustrate that spectral (and thus biochemical) differences are large between leaf samples from the forest and the two cultivated sites, and small between leaf samples from recently converted and historically converted sites. The Discriminant analysis provided a robust technique for estimating the degree of biochemical differences among maize leaf samples. Based on the derived canonical variates, inferences can be made about the overall spectral similarity or dissimilarity of plants/crops under varying conditions of nutrient stress.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv class=\"Section2\" id=\"Sec14\"\u003e\n \u003ch2\u003e3.3 Biomass allocation patterns and ordinal spectral fertility classes.\u003c/h2\u003e\n \u003cp\u003eSoil fertility classes had a significant effect (p\u0026thinsp;=\u0026thinsp;0.006) on the root: shoot ratio. Table \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e shows the mixed-effects model estimates of 95% confidence intervals around means of root: shoot of 14-day-old maize among three soil fertility classes. Allocation of biomass to roots relative to shoot was high among maize plants grown in the low and medium ordinal soil fertility classes, thus yielding positive natural log (root: shoot) ratios. Conversely, plants allocated higher biomass to shoot relative to root in maize plants grown in the high soil fertility class, resulting in a negative (-0.303) ln (root: shoot) ratio. The biomass allocation pattern among the three soil fertility classes is illustrated graphically in Fig. \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e. These results are consistent with the optimal partitioning theory.\u0026nbsp;\u003c/p\u003e\n \u003cp style=\"text-align: center;\"\u003e\u003cstrong\u003eTable 3\u003c/strong\u003e\u003c/p\u003e\n \u003cp style=\"text-align: center;\"\u003e\u0026nbsp;Parameter estimates of linear mixed effects model of the effects of soil spectral class on root: shoot ratio\u003c/p\u003e\n \u003ctable border=\"1\" cellpadding=\"0\" cellspacing=\"0\" style=\"border-collapse: collapse; margin: 0px auto;\" width=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"20.571428571428573%\"\u003e\n \u003cp\u003eSpectral Class\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"25.142857142857142%\"\u003e\n \u003cp\u003eRoot: shoot ratio\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"14.095238095238095%\"\u003e\n \u003cp\u003eStd.Error\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"12.761904761904763%\"\u003e\n \u003cp\u003eDF\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"14.666666666666666%\"\u003e\n \u003cp\u003et-value\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"12.761904761904763%\"\u003e\n \u003cp\u003ep-value\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"20.571428571428573%\"\u003e\n \u003cp\u003eHigh\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"25.142857142857142%\"\u003e\n \u003cp\u003e-0.303\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"14.095238095238095%\"\u003e\n \u003cp\u003e0.064\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"12.761904761904763%\"\u003e\n \u003cp\u003e149\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"14.666666666666666%\"\u003e\n \u003cp\u003e-4.75329\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"12.761904761904763%\"\u003e\n \u003cp\u003e\u0026lt;.0001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"20.571428571428573%\"\u003e\n \u003cp\u003eMedium\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"25.142857142857142%\"\u003e\n \u003cp\u003e0.115\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"14.095238095238095%\"\u003e\n \u003cp\u003e0.061\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"12.761904761904763%\"\u003e\n \u003cp\u003e149\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"14.666666666666666%\"\u003e\n \u003cp\u003e1.902168\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"12.761904761904763%\"\u003e\n \u003cp\u003e0.0591\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" width=\"20.571428571428573%\"\u003e\n \u003cp\u003eLow\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"25.142857142857142%\"\u003e\n \u003cp\u003e0.164\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"14.095238095238095%\"\u003e\n \u003cp\u003e0.092\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"12.761904761904763%\"\u003e\n \u003cp\u003e149\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"14.666666666666666%\"\u003e\n \u003cp\u003e1.774892\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" width=\"12.761904761904763%\"\u003e\n \u003cp\u003e0.078\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003ctable border=\"1\" id=\"Tab2\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003e\u003cbr\u003e\u003c/div\u003e\n \u003c/caption\u003e\n \u003c/table\u003e\n \u003cp\u003e[Figure \u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e about here]\u003c/p\u003e\n \u003cp\u003eConsistent with the optimal partitioning theory, plants growing under conditions of limited supplies of belowground resources will shift biomass partitioning toward more root production and less shoot production (Hilbert, \u003cspan class=\"CitationRef\"\u003e1990\u003c/span\u003e). Consistent with this theory, this study shows maize plants exhibited biomass plasticity in allocation patterns, shifting biomass allocations in a pattern that responds to soil nutrient gradients, along the forest to cropland conversation chronosequence. Variations in levels of selected soil properties among the soil fertility classes illustrate a strong nutrient gradient (see Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). For instance, \u0026ldquo;mean total C levels were 35.8 g kg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, 28.1 g kg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e and 17.5 g kg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e for high, medium and low soil fertility respectively, and mean total N levels were 3.68 g kg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, 2.58 g kg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e and 1.31g kg\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e for high, medium, and low soil fertility respectively\u0026rdquo; ( First reported in Awiti et al.,2008).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv class=\"Section2\" id=\"Sec15\"\u003e\n \u003ch2\u003e3.4 An approach for integrated assessment of soil quality.\u003c/h2\u003e\n \u003cp\u003eThis study demonstrates an integrated approach to soil quality assessment based on near-infrared reflectance of both leaf tissue and topsoil as well as biomass allocation patterns of young maize plants. The core of this approach is near-infrared reflectance spectroscopy, a tool for the rapid and inexpensive acquisition of high-density spatial data. The integrated assessment is related to three key decision points based on plant response namely: (i) right or left shift in the red edge position? (Ruiliang et al., \u003cspan class=\"CitationRef\"\u003e2003\u003c/span\u003e; Fridgen and Varco, \u003cspan class=\"CitationRef\"\u003e2004\u003c/span\u003e); (ii) root: shoot ratio positive or negative? (Davidson, \u003cspan class=\"CitationRef\"\u003e1969\u003c/span\u003e; Hilbert, \u003cspan class=\"CitationRef\"\u003e1990\u003c/span\u003e; McConnaughay and Coleman, \u003cspan class=\"CitationRef\"\u003e1994\u003c/span\u003e; Lindquist et al. 2005); and (iii) is the soil degraded (low fertility) or not degraded (Shepherd and Walsh, \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e; Awiti et al., \u003cspan class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e\n \u003cp\u003eThe analytical efficiency, sample throughput, cost-saving and repeatability of this approach exceeds the levels that currently exist from decades of agronomic research using time-consuming and expensive wet chemistry analysis for soil and plant tissue analyses (Batten, \u003cspan class=\"CitationRef\"\u003e1998\u003c/span\u003e; Viscarra Rossel et al., \u003cspan class=\"CitationRef\"\u003e2006\u003c/span\u003e; Shepherd and Walsh, \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e). This approach is particularly suitable for Sub-Saharan Africa where on-station and on-farm crop response trials still need to be conducted on a large scale but where resources for these trials are often limited. This approach holds the potential for evaluation of soil quality based on a rapid assessment of crop response. This provides opportunities for precise targeting of nutrient management at the farm level, over broad spatial scales across diverse land use land cover, and land management regimes.\u003c/p\u003e\n \u003cp\u003eWith modest resource investments, this approach can be executed across multiple sites in different agro-ecological zones to simultaneously determine soil fertility status and crop response characteristics. Shepherd and Walsh (\u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e) predicted that in ten years, case definitions of fertile or non-fertile soils will have been developed for agronomic and environmental purposes. Once the spectral case definition of infertile soils and the corresponding poor crop health is established, infrared spectroscopy and crop trials can then be implemented routinely to diagnose to measure prevalence, incidence and risk factors (Shepherd and Walsh, \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e) associated with different land use and land management scenarios.\u003c/p\u003e\n \u003cp\u003eThis study focused on plant response to a soil fertility gradient across a forest-cropland chronosequence. However, spectral and biomass response trials to specific soil nutrient deficiencies, especially those that constrain agricultural productivity, such as N, P, and K need to be conducted to better characterize plant nutrient stress and target management response to alleviate specific nutrient constraints. This approach, therefore, offers scope for a reliable and inexpensive approach to precision, spatially explicit management of soil nutrients in rural smallholder systems in sub-Saharan Africa, where resources are scarce. This includes the potential for within season assessment of plant and soil conditions and site/crop-specific management, such as targeted application of fertilizer/organic amendments, and water management. Judicious use of agricultural inputs is essential to balance the factors of soil nutrient depletion, crop nutrient requirements, increasing cost of mineral fertilizer, and the need to minimize environmental impacts of inappropriate fertilizer use and application.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"4. Conclusion","content":"\u003cp\u003eThis study demonstrates that diffuse infrared reflectance responds to plant tissue biochemical composition as well as soil physical and chemical properties and can therefore support an integrated approach for evaluating changes in soil quality and the corresponding plant productivity responses. The study shows that the relative red edge position was generally diagnostic of nutrient-induced stress in young maize plants. Moreover, 14-day-old maize plants exhibited plasticity in biomass allocation patterns, responding optimally to soil nutrient concentrations by shifting biomass allocations in response to soil nutrient gradients, along the forest-cropland conversation chronosequence.\u003c/p\u003e \u003cp\u003eRapid screening of plant and soil health using reflectance spectroscopy techniques holds great promise for guiding spatially explicit, high precision soil nutrient management and plant performance monitoring in low resource settings such as sub-Saharan Africa. This is critical for targeting fertilizer application and reducing the risk of environmental pollution owing to inappropriate and excessive application of fertilizers. Future efforts should be directed to the development of an integrated diagnostic and predictive models based on spectral characteristics and biomass allocation patterns to develop and deploy rapid approaches for testing and monitoring soil quality parameters, within growing season plant responses, at the farm and landscape level, which are important for agricultural productivity and the maintenance of vital ecosystem functions and services.\u003c/p\u003e \u003cp\u003eMoreover, while the study recognizes that real-time NIRS applications are now available for fresh leaves, their use is still greatly limited by low investments in spectral devices in sub-Saharan Africa. More research and innovation will be needed to develop affordable and portable devices that could be used in smallholder agricultural systems, where extension and scientific support services are also the weakest.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u0026nbsp; \u0026nbsp;\u003c/p\u003e\n\u003cp\u003eI am grateful to the BIOTA Africa project for logistical support in the field. Also thank ou field and laboratory team; Elvis Weullow, Sila Andrew, Wilson Ondiala, Isaac Learamo, Luka Anjeho.\u003cstrong\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the The World Agroforestry Centre (ICRAF), German Federal Ministry of Education and Research (BMBF) and the Rockefeller Foundation, Nairobi. \u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAtkinson, M.D., Jervis, A.P., Sangha, R.S., 1997.\u003cstrong\u003e \u003c/strong\u003eDiscrimination between \u003cem\u003eBetulapendula\u003c/em\u003e, \u003cem\u003eBetula pubescens\u003c/em\u003e, and their hybrids using near-infrared reflectance spectroscopy. Canadian Journal of Forest Research\u003cem\u003e \u003c/em\u003e27, 1896-1900. https://doi.org/10.1139/x97- 141\u003c/li\u003e\n\u003cli\u003eAwiti, A.O., Walsh, M.G., Shepherd, K.D., Kinyamario, J. (2008). Soil condition classification using infrared spectroscopy: A proposition for assessment of soil condition along a tropical forest-cropland chronosequence. Geoderma 143, 73-84. https://doi.org/10.1016/j.geoderma.2007.08.021\u003c/li\u003e\n\u003cli\u003eBatten, G.D. (1998). Plant analysis using infrared near-infrared reflectance spectroscopy: the potential and limitations. Australian Journal of Experimental 38, 697-706. DOI:10.1071/EA97146\u003c/li\u003e\n\u003cli\u003eBen-Dor, E., Bannin, A., (1995). Near infrared analysis (NIRA) is a rapid method to simultaneously evaluate several soil properties. Soil Science Society of America Journal 59, 364-372. https://doi.org/10.2136/sssaj1995.03615995005900020014x\u003c/li\u003e\n\u003cli\u003eCarter, G.A., (1993) Responses of leaf reflectance to plant stress. American Journal of Botany 80, 239-243. https://doi.org/10.2307/2445346\u003c/li\u003e\n\u003cli\u003eCarter, G.A., Knapp, A.K. (2001). Leaf optical properties in higher plants: linking spectral characteristics to stress and chlorophyll concentration. American Journal of Botany 88, 677-684. PMID: 11302854.\u003c/li\u003e\n\u003cli\u003eMalini, RC., Christopher, J., Das, S., Apan, A., Menzies, N.W., Chapman, S., Vincent Mellor, V., Dang, YP. (2022). Detection of Calcium, magnesium, and chlorophyll variations of wheat genotypes on sodic soils using hyperspectral red edge parameters, Environmental Technology \u0026amp; Innovation, Volume 27, 102469. Https://doi.org/10.1016/j.eti.2022.102469.\u003c/li\u003e\n\u003cli\u003eChang, C-W., Laird D.A., Mausbach, M.J., Hurburgh Jr. C.R., (2001). Near-Infrared reflectance spectroscopy-principal components regression analysis of soil properties. Soil Science Society of America Journal 65, 480-490. https://doi.org/10.2136/sssaj2001.652480x\u003c/li\u003e\n\u003cli\u003eChun L., Mi, G., Li, J., Chen, F., Zhang, F., (2005). Genetic analysis of maize root characteristics in response to low nitrogen stress. Plant and Soil 276, 369-382. DOI 10.1007/s11104-005-5876-2\u003c/li\u003e\n\u003cli\u003eCohen, M.J., Shepherd, K.D., Walsh, M.G., (2004). Empirical reformulation of the universal soil loss equation for soil erosion risk assessment in a tropical watershed. Geoderma 124, 235-252.https://doi.org/10.1016/j.geoderma.2004.05.003\u003c/li\u003e\n\u003cli\u003eColeman, J. S., McConnaughay, K. D. M., (1994). Interpreting phenotypic variation in plants. Trends in Ecology and Evolution 9, 951-954. https://doi.org/10.1016/0169-5347(94)90087-6\u003c/li\u003e\n\u003cli\u003eCurran, P.J., (1989). Remote sensing of foliar chemistry. Remote Sensing of Environment\u003cem\u003e \u003c/em\u003e30, 271-278. https://doi.org/10.1016/0034-4257(89)90069-2\u003c/li\u003e\n\u003cli\u003eDavidson, R. L., (1969). Effects of root/leaf temperature differentials on root/shoot ratios in some pasture grasses and clover. Annals of Botany 33, 561-569. https://www.jstor.org/stable/42908746\u003c/li\u003e\n\u003cli\u003eDawson, TP, Curran, PJ. (1998). Technical note A new technique for interpolating the reflectance red edge position. Int. J. Remote Sensing, 19: 2133-2139. https://doi.org/10.1080/014311698214910\u003c/li\u003e\n\u003cli\u003eDev, S. P, KKR, Bhardwaj K K R. (1995). Effect of crop wastes and nitrogen levels of biomass production and nitrogen uptake in wheat-maize sequence. Ann. Agric. Res. 16, 264\u0026ndash;267.\u003c/li\u003e\n\u003cli\u003eEly, KS., Burnett, AC., Liberman-Cribbin, W., Serbin, SP., Rogers, A. (2019). Spectroscopy can predict key leaf traits associated with source\u0026ndash;sink balance and carbon\u0026ndash;nitrogen status. Journal of Experimental Botany, Vol. 70 (6): 1789-1799. DOI: 10.1093/jxb/erz061 \u003c/li\u003e\n\u003cli\u003eEitel, UH., Vierling, LA., Litvak, ME., Long, DS., Schulthess, U., Ager, AA., Krofcheck, DJ., Stoscheck, L. (2011). Broadband, red-edge information from satellites improves early stress detection in a New Mexico conifer woodland. Remote Sensing of Environment 115: 3640-3646. https://doi.org/10.1016/j.rse.2011.09.002\u003c/li\u003e\n\u003cli\u003eFridgen, J.L., Varco, J.J. (2004). Dependency of cotton leaf nitrogen, chlorophyll, and reflectance on nitrogen and potassium availability. Agronomy Journal 96, 63-69 https://doi.org/10.2134/agronj2004.6300\u003c/li\u003e\n\u003cli\u003eGates, D. M., Keegan H. J., Schlecter, J. C., Weidner, V. R. (1965). Spectral Properties of Plants. Applied Optics 4(1), 11-20. https://doi.org/10.1364/AO.4.000011\u003c/li\u003e\n\u003cli\u003eHilbert, D. W. (1990). Optimization of plant root: shoot ratios and internal nitrogen concentrations. Annals of Botany 66, 91-99. https://doi.org/10.1093/oxfordjournals.aob.a088005\u003c/li\u003e\n\u003cli\u003eJingchao, T., Sun, B., Cheng, R., Shi, Z., Luo, D., Luil, S., Centritto, M. (2019). Effects of soil nitrogen (N) deficiency on photosynthetic N-use efficiency in N-fixing and non-N-fixing tree seedlings in subtropical China. Nature, Scientific Reports, 9:4604 https://doi.org/10.1038/s41598-019-41035-1\u003c/li\u003e\n\u003cli\u003eKemsley, E.K., Ruault, S., Wilson, R.H., (1995)\u003cstrong\u003e. \u003c/strong\u003eDiscrimination between \u003cem\u003eCoffea arabica \u003c/em\u003eand \u003cem\u003eCoffea canephora \u003c/em\u003evariant \u003cem\u003erobusta \u003c/em\u003ebeans using infrared spectroscopy. Food Chemistry, 54, 321\u0026ndash;326. https://doi.org/10.1016/0308-8146(95)00030-M\u003c/li\u003e\n\u003cli\u003eKenya Soil Survey. (2004). Characteristics of eight soil profiles in Kakamega. Miscellaneous report No. M58.\u003c/li\u003e\n\u003cli\u003eLamb, D. W., Steyn-Ross, M., Schaare, P., Hanna, M. M., Silvester W., Steyn-Ross, A., (2002). Estimating leaf nitrogen concentration in ryegrass (\u003cem\u003eLolium \u003c/em\u003espp.) pasture using the chlorophyll red-edge: theoretical modeling and experimental observations. International Journal. Remote Sensing 23, 3619-3648.,https://doi.org/10.1080/01431160110114529\u003c/li\u003e\n\u003cli\u003eBonifas, K.D., Walters, D.T., Cassman, K.G, Lindquist, J.L. (2005). Nitrogen supply affects root: shoot ratio in corn and velvetleaf (\u003cem\u003eAbutilon theophrasti\u003c/em\u003e). Weed Science 53, 670-675. DOI:10.1614/WS-05-002R.1\u003c/li\u003e\n\u003cli\u003eLevi, N, Karnieli, A, Paz-Kagan. T. (2020. Using reflectance spectroscopy for detecting land-use effects on soil quality in drylands, Soil and Tillage Research, Volume 199. https://doi.org/10.1016/j.still.2020.104571\u003c/li\u003e\n\u003cli\u003eMcCullagh, P., (1980). Regression models for ordinal data. Journal of Royal Statistical Society B 42 (2), 109-142.\u003c/li\u003e\n\u003cli\u003eNing, J., Sheng, M., Yi, X., Wang, Y., Hou, Z., Zhang, Z., Gu, X. (2018). Rapid evaluation of soil fertility in tea plantation based on near-infrared spectroscopy, Spectroscopy Letters, 51:9, 463-471, DOI: 10.1080/00387010.2018.1475398\u003c/li\u003e\n\u003cli\u003eOsborne, S. L., Schepers, J. S., Francis, D. D., Schlemmer, M. R. (2002). Detection of phosphorus and nitrogen deficiencies in corn using spectral radiance measurements. Agronomy Journal 94, 1215-1221. https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1006\u0026amp;context=agronomy facpub\u003c/li\u003e\n\u003cli\u003ePrananto, JA., Minansy, B., Weaver, T. (2021). Rapid cost-effective nutrient content analysis of cotton leaves using near-infrared spectroscopy (NIRS). Peer J 9: e11042DOI 10.7717/peerj.11042.\u003c/li\u003e\n\u003cli\u003ePeng, Y., Zhao, L., Hu, Y., Wang, G., Wang, L., Liu, Z. (2019). Prediction of Soil Nutrient Contents using Visible and Near-Infrared reflectance spectroscopy. SPRS Int. J. Geo-Inf. 2019, 8, 437; doi:10.3390/ijgi8100437.\u003c/li\u003e\n\u003cli\u003ePinheiro, J.C., Bates, D.M. (2000). Mixed-Effects Models in S-PLUS. Springer, New York.\u003c/li\u003e\n\u003cli\u003ePinar, A., Curran PJ. (1996). Grass chlorophyll and the reflectance red-edge. International Journal or Remote Sensing, 17:351-357. https://doi.org/10.1080/01431169608949010\u003c/li\u003e\n\u003cli\u003eRichardson, A. D., Reeves, J.B., Gregorie, T.G. (2004). Multivariate analyses of visible/near infrared (VIS/NIR) absorbance spectra reveal underlying spectral differences among dried, ground conifer needle samples from different growth environments. New Phytologist 161, 291-301. DOI:10.1046/j.1469-8137.2003.00913.x\u003c/li\u003e\n\u003cli\u003eRuiliang P., Gong, P., Biging, G.S., Larrieu, M.R. (2003). Extraction of red edge optical parameters from hyperion data for estimation of forest leaf area index. IEEE Transactions on Geoscience and Remote Sensing Vol. 41(4), 916-921.DOI:10.1109/TGRS.2003.813555\u003c/li\u003e\n\u003cli\u003eSAS Institute Inc. (2016). SAS/STAT\u0026reg; 14.2 User\u0026rsquo;s Guide: Higher Performance Procedures. Cary, NC: SAS Institute Inc.\u003c/li\u003e\n\u003cli\u003eSanchez, P.A., Villachica, J.H., Bandy, D. E. 1(983). Soil fertility dynamics after clearing a tropical rainforest in Peru. Soil Science Society of America Journal 47, 1171-1178. https://doi.org/10.2136/sssaj1983.03615995004700060023x\u003c/li\u003e\n\u003cli\u003eSchlemmer, M.R., Francis, D.D., Shanahan, J.F., Schepe, J.S. (2005). Remotely measuring chlorophyll content in corn leaves with differing nitrogen levels and relative water content. Agronomy Journal 97, 106-112. https://doi.org/10.2134/agronj2005.0106\u003c/li\u003e\n\u003cli\u003eShepherd, K.D., Walsh, M.G., 2002. Development of reflectance spectral libraries for characterization of soil properties. Soil Science Society America Journal 66, 988-998.\u003c/li\u003e\n\u003cli\u003eShepherd, K.D., Palm, C.A., Gachengo, C.N., Vanlauwe, B., (2003). Rapid characterization of residue quality for soil and livestock management in tropical agroecosystems using near-infrared spectroscopy. Agronomy Journal 95, 1314\u0026ndash;1322.DOI:10.2134/agronj2003.1314\u003c/li\u003e\n\u003cli\u003eShepherd, K.D., Walsh, M.G., (2007). Infrared spectroscopy \u0026ndash; enabling an evidence-based diagnostic surveillance approach to agricultural and environmental management in developing countries. Journal of Near Infrared Spectroscopy 15, 1-19.https://doi.org/10.1255/jnirs.716\u003c/li\u003e\n\u003cli\u003eStenberg, B., Viscarra Rossel, R A., Mouazen, AM., Wetterlind, J. (2010). Visible and Near Infrared Spectroscopy in Soil Science. In Donald L. Sparks, editor: Advances in Agronomy, Vol. 107, Burlington: Academic Press, 2010, pp. 163-215\u003c/li\u003e\n\u003cli\u003eStocking MA. (2003). Tropical Soils and Food Security: The Next 50 Years. Science, Vol 302: 1-5. DOI:10.1126/science.1088579\u003c/li\u003e\n\u003cli\u003eTittonell, P, Muruiki, A, Klapwijk, CJShepherd, KD., Coe, R Vanlauwe, B. (2013). Soil Heterogeneity and Soil Fertility Gradients in Smallholder Farms of the East African Highland. Soil Sci. Soc. Am. J. 77:525\u0026ndash;538. https://doi.org/10.2136/sssaj2012.0250 \u003c/li\u003e\n\u003cli\u003eVagen, T-G., Shepherd, K.D., Walsh, MG., (2006). Sensing landscape level change in soil fertility following deforestation and conversion in the highlands of Madagascar using Vis-NIR spectroscopy. Geoderma133, 281-294. https://doi.org/10.1016/j.geoderma.2005.07.014\u003c/li\u003e\n\u003cli\u003evan Noordwjik, M, Cerri, C, Woomer, P Nugroho, M. Bernoux, M. (1997). Soil carbon dynamics in the humid tropical forest zone. Geoderma 79, 187-225. https://doi.org/10.1016/S0016-7061(97)00042-6\u003c/li\u003e\n\u003cli\u003eVickery, P. J., 1981. Pasture growth under grazing. In \u003cem\u003eGrazing Animals\u003c/em\u003e, edited by F. H. W. Morley (Amsterdam: Elsevier), pp. 55\u0026ndash;78.\u003c/li\u003e\n\u003cli\u003eViscarra Rossel, R.A., Walvoort, D.J.J., McBratney, A.B., Janik, L.J., Skjemstad, J.O., (2006). Visible, near-infrared or combined diffuse reflectance spectroscopy for simultaneous assessment of various soil properties. Geoderma 131, 59-75. https://doi.org/10.1016/j.geoderma.2005.03.007\u003c/li\u003e\n\u003cli\u003eWalburg, G., Bauer, M.E., Daughtry, C.S.T., Housley, T.L. (1982). Effects of nitrogen nutrition on the growth, yield and reflectance characteristics of corn canopies. Agronomy Journal 74, 677-683. https://doi.org/10.2134/agronj1982.00021962007400040020x \u003c/li\u003e\n\u003cli\u003eWetterlind, J., B. Stenberg, Johnson, A (2008). Near infrared reflectance spectroscopy compared with soil clay and organic matter content for estimating within-field variation in N uptake in cereals. Plant Soil, 302:317\u0026ndash;327 DOI 10.1007/s11104- 007-9489-9\u003c/li\u003e\n\u003cli\u003eXie J, Tang L, Wang Z, Xu G, Li Y. (2012). Distinguishing the Biomass Allocation Variance Resulting from Ontogenetic Drift or Acclimation to Soil\u003cbr\u003e Texture. PLoS ONE 7(7): e41502. doi: 10.1371/journal.pone.0041502\u003c/li\u003e\n\u003cli\u003eZhao, D., Reddy, K. R., Kakani, V. G., Read, J. J., Carter, GA (2003). Corn (\u003cem\u003eZea mays \u003c/em\u003eL.) growth, leaf pigment concentration, photosynthesis and leaf hyperspectral reflectance properties as affected by nitrogen supply. Plant Soil 257, 205-217. https://doi.org/10.1023/A:1026233732507 \u003c/li\u003e\n\u003cli\u003eZhao D., Reddy K. R., Kakani, V. G., Reddy, V. R., (2005). Nitrogen deficiency effects plant growth, leaf photosynthesis, and hyperspectral reflectance properties of Sorghum. European Journal of Agronomy 22, 391-403. https://doi.org/10.1016/j.eja.2004.06.005\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Chronosequence, Soil quality gradient, Near-infrared reflectance, Root: shoot ratio, Biomass allocation","lastPublishedDoi":"10.21203/rs.3.rs-1800544/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-1800544/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis paper characterizes the variation in soil quality along a forest to cropland conversation chronosequence using (i) VIS/NIR spectral characteristics of leaf samples from maize plants grown on soils from a forest-cropland chronosequence and, (ii) evaluates root: shoot biomass partitioning patterns of maize plants along a chronosequence soil quality gradient. Five hundred and forty-two topsoil samples were retrieved from a forest to cropland conversation chronosequence comprising primary and secondary forest, recently converted and historically converted farmlands. About 200 grams of each of the 542 samples were used in a bioassay in which hybrid maize seeds were planted in plastic pots under controlled conditions. Plants were harvested after 14 days. Shoots and roots were separated, dried and weighed. Discriminant analysis was used to evaluate the underlying spectral differences among the dried, ground shoot samples. Root: shoot biomass allocation patterns were assessed relative to soil spectral condition classes. The difference in reflectance of leaf samples was significant (Wilk’s Lambda = 0.006; F= 18.27; p\u0026gt;0.0001). The reflectance of leaf samples from maize plants grown in nutrient-poor exhibited a right shift in the 0.68-0.74 µm region (red edge position); a phenomenon diagnostic of nutrient stress. Maize plants exhibited plasticity in biomass partitioning patterns consistent with the soil quality gradient defined by fertility classes. Log-transformed root-shoot ratios revealed that maize plants grown in high nutrient soils had a low root/shoot ratio. High root: shoot ratio was observed among maize plants grown in soils classified as low in nutrients.\u0026nbsp;\u003c/p\u003e","manuscriptTitle":"Using near-infrared reflectance and biomass allocation patterns of maize plants (Zea mays L.) to characterize soil fertility and productivity response dynamics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2022-07-06 14:39:47","doi":"10.21203/rs.3.rs-1800544/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"999d194d-0e0f-45f0-b7e6-f3493912ab12","owner":[],"postedDate":"July 6th, 2022","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2022-07-06T14:39:49+00:00","versionOfRecord":[],"versionCreatedAt":"2022-07-06 14:39:47","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-1800544","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-1800544","identity":"rs-1800544","version":["v1"]},"buildId":"369fNeqWncA4NS6XSWjrt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}