Keith Shepherd | World Agroforestry Centre (original) (raw)
Papers by Keith Shepherd
New Zealand is undertaking a new phase of soil mapping (S-map) with the goal of national soil map... more New Zealand is undertaking a new phase of soil mapping (S-map) with the goal of national soil map coverage at 1:50 000 scale. One S-map objective is to develop a national capability in digital soil mapping (DSM). Concurrent development of the global soil map (GSM) project provides an opportunity for the Smap project to align its work with the GSM initiative both to achieve an enhanced national soil map and contribute New Zealand data to the GSM project. Consequently we have applied GSM method guidelines nationally. We matched the generic GSM methods to 193 New Zealand soilscapes. Soilscapes with similar levels of legacy soil information were grouped into eight soilscape groups, and DSM methods applied to them. Four soilscape groups were in low relief land where intensification of land use generates high demand for new soil information. The other four method classes were in rolling, hilly and mountainous land. The most extensive method in low relief land requires establishment of bet...
European Journal of Soil Science, 2020
Simple measures of appropriate levels of soil organic matter are needed for soil evaluation, mana... more Simple measures of appropriate levels of soil organic matter are needed for soil evaluation, management and monitoring, based on readily measurable soil properties. We test an index of soil organic matter based on the soil organic carbon (SOC) to clay ratio, defined by thresholds of SOC/clay ratio for specified levels of soil structural quality. The thresholds were originally delineated for a small number of Swiss soils. We assess the index using data from the initial sampling (1978–83) of the National Soil Inventory of England and Wales, covering 3,809 sites under arable land, grassland and woodland. Land use, soil type, annual precipitation and soil pH together explained 21% of the variance in SOC/clay ratio in the dataset, with land use the most important variable. Thresholds of SOC/clay ratio of 1/8, 1/10 and 1/13 indicated the boundaries between “very good”, “good”, “moderate” and “degraded” levels of structural condition. On this scale, 38.2, 6.6 and 5.6% of arable, grassland ...
Journal of the University of Ruhuna, 2016
Hyperion hyperspectral (VNIR-SWIR) satellite image data was used to map the spatial distribution ... more Hyperion hyperspectral (VNIR-SWIR) satellite image data was used to map the spatial distribution of mineral precipitates at Lake Magadi area, in the southernmost part of the eastern Kenya rift, Kenya. Mapping was coupled with laboratory analysis, including visible near infrared diffuse reflectance spectroscopy (VNIR) measurements and X-ray diffraction for selected rock and soil samples. The VNIR spectral responses of 92 rock and soil samples including trona, chert, diatomite, basalt/trachyte, erionite, Green bed and High Magadi bed were studied and identified. The spectral signatures of Chert samples show the broad Si-OH absorption feature at 2.2 μm while, Green bed, High Magadi bed and diatomite exhibit carbonate absorption feature at 2.35 μm with broad Si-OH absorption feature at 2.2 μm. Trona exhibits six common absorption features at 1.50, 1.74, 1.94, 2.03, 2.22 and 2.39 μm. These characteristics spectral absorption features together with the general shape of the spectral curve ...
Soil Science Society of America Journal, 1998
Abstract We hypothesized that trees can rapidly root into subsoil and capture NO 3, which can acc... more Abstract We hypothesized that trees can rapidly root into subsoil and capture NO 3, which can accumulate in the subsoil of agricultural soils with high anion sorption. The vertical distribution of root length and inorganic N (NO 3 and NH 4) to 3.95-m soil depth was compared for 11-mo-old stands of eucalyptus (Eucalyptus grandis W. Hill ex Maiden), sesbania [Sesbania sesban (L.) Merr.], calliandra (Calliandra calothyrsus Meissner), markhamia [Markhamia lutea (Benth.) Schumann], and grevillea (Grevillea robusta A. ...
Forest Ecology and Management, 1999
Phosphorus de®ciency is widespread in the subhumid highlands of eastern Africa but there are few ... more Phosphorus de®ciency is widespread in the subhumid highlands of eastern Africa but there are few data on the effect of P de®ciency on the growth of agroforestry tree species. We studied the effect of P application on growth, nutrient uptake and dry matter partitioning in young trees of Calliandra calothyrsus, Cedrela serrulata, Eucalyptus grandis, Grevillea robusta, Markhamia lutea, Senna spectabilis, and Sesbania sesban on a P-de®cient soil (Kandiudal®c Eutrudox, bicarbonate-EDTA extractable P 1 mg kg À1) in western Kenya. The trees were grown at two P levels (control and 500 kg added P ha À1) at 1 m 2 spacing in a randomized complete block design with three replications. Leaf K concentrations were in the low range for all species (5±9 mg g À1) and K de®ciency may have limited responses to P. Averaged over species, P addition increased aboveground shoot dry matter by a factor of 2.6 at 62 and 124 days, but the response decreased to 1.3 at 325 days. The increases at 62 days were large in sesbania (5.4) and eucalyptus (3.2) but small in calliandra (1.4) and markhamia (1.1). Relative response to P was more strongly correlated with shoot growth rate per unit root length among species than with shoot growth rate alone. Calliandra, which had high early growth rate but low response to added P, had an exceptionally high root length (6.0 km m À2) compared with the other species (0.3±2.1 km m À2). P addition increased N and P content but decreased ®nal shoot K content in sesbania and calliandra, and had little effect on K content in the other species. The high-yielding species (eucalyptus, sesbania and calliandra) accumulated more than 30 g N and 2 g P m À2 in shoots in 325 days of growth. The proportion of total shoot N in wood (branch stem) was in a higher range (67±75%) in the shrubby species (sesbania, calliandra, senna) than in the upperstorey tree species (38±43%). Slow early shoot growth relative to total root length, and high speci®c root length (root length per unit root mass) are proposed as criteria for the selection of species and provenances that are well adapted to P de®cient soils.
Experimental Agriculture, 1985
SUMMARYThe development and influence of crop drought stress was examined in winter-planted small-... more SUMMARYThe development and influence of crop drought stress was examined in winter-planted small-grain cereals under Mediterranean environmental conditions. In two average or wetter than average years crop drought stress in the grain-filling period was greatly exacerbated by the addition of fertilizer (N and P) and this usually resulted in significantly lower kernel weights. Yet this reduction was not so substantial as to seriously affect crop yields. Total grain yields were largest in treatments receiving fertilizer, due to greater spike numbers and numbers of kernels per spike. But heavy nitrogen applications may promote the risk of a large proportion of small or shrivelled grains in a dry year.
Biology and Fertility of Soils, 1996
A greenhouse experiment was conducted to assess the arbuscular mycorrhizal (AM) fungi inoculum po... more A greenhouse experiment was conducted to assess the arbuscular mycorrhizal (AM) fungi inoculum potential for Leucaena leucocephala in a range of nutrientdepleted farm soils in western Kenya. Leucaena was grown in 12 uninoculated farm soils with pH 5.0-6.7, with or without rock P and farmyard manure. Root infection, nodulation, and shoot and root weight were determined 25, 46, and 69 days after planting. Spore concentrations in the farm soils at the start of the experiment ranged from 44 to 126 live spores per 100 g dry soil with 6-10 species per soil, principally ScuteIlospora spp. and Acaulospora spp. Nodulation was absent or poor in all soils, indicating the need for rhizobial inoculation of species belonging to the leucaena crossinoculation group in this agrosystem. Rock-P alone increased final shoot dry weight by a factor of 1.4, manure alone by 1.8, and rock-P plus manure by 1.9, compared with no ameliorant. Root infection with AM fungi was detected in all soils 25 days after planting and increased linearly in the different farm soils to values of 33-65% 69 days after planting. Soil pH and root infection 25 days after planting accounted for much of the variation in final shoot weight among soils with no amelioront (87%). As early root infection increased from 20 to 40% at soil pH 5.0, the predicted final shoot dry weight was doubled, and the response to ameliorants was reduced by two-thirds. The growth responses to increased infection became smaller as pH increased from 5.0 to 6.5.
Agriculture, Ecosystems & Environment, 1998
A farm simulation model was designed to assess the long-term impact of existing soil management s... more A farm simulation model was designed to assess the long-term impact of existing soil management strategies, on farm productivity, pro®tability and sustainability. The model, which runs in time units of 1 year, links soil management practices, nutrient availability, plant and livestock productivity, and farm economics A case study is presented of the application of the model to existing, mixed farm systems in Vihiga district, in the highlands of western Kenya. Three representative farm types were developed using participatory techniques to re¯ect differences in resource endowments and constraints faced by farmers. The model was used to assess the sustainability of the existing systems for the three farm types as a basis for recommending improved practices for each. A summary model for calculating new sustainability indicators of soil productivity is presented. The low (LRE) and medium (MRE) resource endowment farms, which comprise about 90% of the farms in the area, have declining soil organic matter and low productivity and pro®tability. In contrast, the high resource endowment category of farms (HRE) have increasing soil organic matter, low soil nutrient losses and are productive and pro®table. Crop nutrient yields were 17, 19 and 86 kg N ha À1 year À1 on LRE, MRE and HRE farms, respectively. Soil C, N and P budgets were negative in LRE and MRE but positive in HRE. Farm revenue in LRE and MRE was 2±13% of farm revenue in HRE. It comprised 7% of household income in LRE compared with 25% in MRE and 63% in HRE. It is concluded that low land and capital resources constrain the adoption of ecologically and economically sustainable soil management practices on the majority of farms in the area. Strategies are needed to (i) increase the value of farm output (ii) increase high quality nutrient inputs at low cash and labour costs to the farmer, and (iii) increase off-farm income.
Analysis of spatial variation of soil fertility gradients in smallholdings in tropical Africa usi... more Analysis of spatial variation of soil fertility gradients in smallholdings in tropical Africa using geostatical techniques : Case studies from Vihiga and Siaya Districts of Western Kenya [abstract] [on line]. DSpace/Manakin Repository. ...
African Crop Science Journal, 2009
In many farming landscapes across Sub-Saharan Africa (SSA), soil fertility has been on the declin... more In many farming landscapes across Sub-Saharan Africa (SSA), soil fertility has been on the decline with significant implications on crop productivity. However, even with such a decline, soil nutrient levels still differ significantly between farms, fields or within the same field. Knowledge of such spatial variability and relationships among soil properties is important in implementation of agricultural land management practices. In this study, the spatial variability of soil organic carbon (SOC) in two districts of western Kenya was ...
Agricultural Systems, 2006
Arable land in western Kenya is under considerable pressure from increasing human population. Rur... more Arable land in western Kenya is under considerable pressure from increasing human population. Rural households depend on farming for at least part of their livelihood, and poverty rates are among the highest in Kenya. Land is often depleted of nutrients, and for most farmers, access to inputs and markets is poor. There is a need to identify options that are manageable within the context of the farmer's resource base and the household's objectives that could improve farm household well-being. In this study we integrated qualitative informal participatory approaches with quantitative mathematical programming and biophysical simulation modelling. Households in four sub-locations in Vihiga District were clustered and pilot cases identified. Meetings were held with farmers to elicit their perceptions of what their ideal farm would look like, and how its performance might compare with their own farm's performance. With farmers' help, a range of scenarios was analysed, relating to changes in current enterprise mixes, changes in current farm sizes, and changes in prices of staples foods and cash crops. A considerable mismatch was found between farmers' estimates of their own farm's performance, and what was actually produced. There seems to be a threshold in farm size of
Agriculture, Ecosystems & Environment, 2005
Strong gradients of decreasing soil fertility are found with increasing distance from the homeste... more Strong gradients of decreasing soil fertility are found with increasing distance from the homestead within smallholder African farms, due to differential resource allocation. As nutrient use efficiency varies strongly along these gradients, such heterogeneity must be considered when designing soil management strategies, aimed at an improved overall resource use efficiency at farm scale. Here, we quantify the magnitude and study the origin of farmer-induced, within-farm soil fertility gradients as affected by biophysical and socioeconomic conditions, and investigate farmers' perceptions of such heterogeneity. Farm transects, participatory resource flow mapping, farmers' classification of land qualities, and soil sampling for both chemical and spectral reflectance analyses were performed across 60 farms in three sub-locations (Emuhaia, Shinyalu, Aludeka) representing the variability found in the highlands of western Kenya. Differences between the various field types of a farm were observed for input use (e.g. 0.7-104 kg N ha À1), food production (e.g. 0.6-2.9 t DM ha À1), partial C (e.g. À570 to 1480 kg ha À1) and N (e.g. À92 to 57 kg ha À1) balances and general soil fertility status, despite strong differences across sub-locations. Concentration of nutrients in the home fields compared with the remote fields were verified for extractable P (e.g. 2.1-19.8 mg kg À1) and secondarily for exchangeable K (e.g. 0.14-0.54 cmol (+) kg À1), on average, whereas differences for soil C and N were only important when considering each individual farm separately. Farmers managed their fields according to their perceived land quality, varying the timing and intensity of management practices along soil fertility gradients. Fields classified by them as poor were planted later (up to 33.6 days of delay), with sparser crops (ca. 30% less plants m À2) and had higher weed infestation levels than those classified as fertile, leading to important differences in maize yield (e.g. 0.9 versus 2.4 t ha À1). The internal heterogeneity in resource allocation varied also between farms of different social classes, according to their objectives and factor constraints. Additionally, the interaction of sub-location-specific socioeconomic (population, markets
Current Opinion in Environmental Sustainability, Jun 1, 2017
Accountability and adaptive management of recent global agreements such as the Sustainable Develo... more Accountability and adaptive management of recent global agreements such as the Sustainable Development Goals and Paris Climate Agreement, will in part rely on the ability to track progress toward the social and environmental targets they set. Current metrics and monitoring systems, however, are not yet up to the task. We argue that there is an imperative to consider principles of coherence (what to measure), standardization (how to measure) and decision-relevance (why to measure) when designing monitoring schemes if they are to be practical and useful. New approaches that have the potential to match the necessary scale of monitoring, with sufficient accuracy and at reasonable cost, are emerging; although, they represent a significant departure from the historical norm in some cases. Iterative review and adaptation of analytical approaches and available technology will certainly be needed to continuously design ways to best track our progress.
PLOS ONE, Feb 2, 2022
Adept use of fertilizers is critical if sustainable development goal two of zero hunger and agroe... more Adept use of fertilizers is critical if sustainable development goal two of zero hunger and agroecosystem resilience are to be achieved for African smallholder agroecosystems. These heterogeneous systems are characterized by poor soil health mainly attributed to soil nutrient depletion. However, conventional methods do not take into account spatial patterns across geographies within agroecosystems, which poses great challenges for targeted interventions of nutrient management. This study aimed to develop a novel populationbased farm survey approach for diagnosing soil nutrient deficiencies. The approach embraces principles of land health surveillance of problem definition and rigorous sampling scheme. The advent of rapid soil testing techniques, like infrared spectroscopy, offers opportune avenues for high-density soil and plant characterization. A farm survey was conducted on 64 maize fields, to collect data on soil and plant tissue nutrient concentration and grain yield (GY) for maize crops, using hierarchical and purposive sampling. Correlations between soil test values with GY and biomass were established. The relationship between GY, soil NPK, and the tissue nutrient concentrations was evaluated to guide the setting up of localized critical soil test values. Diagnosis Recommendation Integrated System (DRIS) indices for total nitrogen (N), total phosphorus (P), and total potassium (K) were used to rank and map the prevalence of nutrient limitations. A positive correlation existed between plant tissue nutrient concentration with GY with R 2 values of 0.089, 0.033, and 0.001 for NPK, respectively. Soil test cutoff values were 0.01%, 12 mg kg-1 , 4.5 cmol c kg-1 for NPK, respectively, which varied slightly from established soil critical values for soil nutrient diagnostics. N and K were the most limiting nutrients for maize production in 67% of sampled fields. The study demonstrates that a population-based farm survey of crop fields can be a useful tool in nutrient diagnostics and setting priorities for site-specific fertilizer recommendations. A larger-scale application of the approach is warranted.
Field Crops Research, Aug 1, 2021
Abstract Determination of the concentration of nutrients in the plant is key information for eval... more Abstract Determination of the concentration of nutrients in the plant is key information for evaluating crop nutrient removal, nutrient use efficiency, fertilizer recommendations guidelines, and in turn for improving food security and reducing environmental footprints of crop production. Diffuse infrared (IR) reflectance spectroscopy is a powerful, rapid, cheap, and less pollutant analytical tool that could be substituted for traditional laboratory methods for the determination of the concentration of nutrients in plants. However, its accuracy for predicting the concentration of nutrients in rice plants is poorly known. This study aimed i) to determine macro- and micronutrients concentration that can be accurately predicted by near-infrared (NIR, 7498–4000 cm−1), mid-infrared (MIR, 4000–600 cm−1), or their combination (NIR-MIR, 7498–600 cm−1) spectra, ii) to identify the most suitable spectral range with the best prediction potential for the simultaneous analysis of nutrients concentrations in rice plants (straw and paddy) and iii) to assess the influence of agro-ecological zone and production system on nutrients concentrations in straw and paddy (unhulled grains) samples. Second-derivative spectra were fitted against plant laboratory reference data using partial least-squares regression (PLSR) to estimate six macronutrients (N, P, K, Ca, Mg, and S) and seven micronutrients (Na, Fe, Mn, B, Cu, Mo, and Zn) concentration in paddy and rice straw samples collected at harvest from 1628 farmers’ fields in 20 sub-Saharan African (SSA) countries. The modeling prediction potential was assessed by coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and ratio of performance to interquartile distance (RPIQ). Good prediction models (0.75
Uncertainty and risks are common elements of all major projects. Yet such uncertainty is rarely e... more Uncertainty and risks are common elements of all major projects. Yet such uncertainty is rarely effectively calculated when analysing project costs and benefits. This paper presents a Bayesian network (BN) modelling framework to calculate the costs, benefits, and return on investment of a project over a specified time period, allowing for changing circumstances and trade-offs. The framework uses hybrid and dynamic BNs containing both discrete and continuous variables over multiple time stages. The BN calculates costs and benefits based on multiple causal factors including the effects of individual risk factors, budget deficits, and time value discounting. The method is illustrated using a case study of an agricultural development project.
PLOS ONE, Jun 5, 2020
Agricultural development projects have a poor track record of success mainly due to risks and unc... more Agricultural development projects have a poor track record of success mainly due to risks and uncertainty involved in implementation. Cost-benefit analysis can help allocate resources more effectively, but scarcity of data and high uncertainty makes it difficult to use standard approaches. Bayesian Networks (BN) offer a suitable modelling technology for this domain as they can combine expert knowledge and data. This paper proposes a systematic methodology for creating a general BN model for evaluating agricultural development projects. Our approach adapts the BN model to specific projects by using systematic review of published evidence and relevant data repositories under the guidance of domain experts. We evaluate a large-scale agricultural investment in Africa to provide a proof of concept for this approach. The BN model provides decision support for project evaluation by predicting the value-measured as net present value and return on investment-of the project under different risk scenarios.
AfSIS Phase I was a collaborative projective funded by the Bill and Melinda Gates Foundation (BMG... more AfSIS Phase I was a collaborative projective funded by the Bill and Melinda Gates Foundation (BMGF), which aimed to provide a consistent baseline of soil information for monitoring soil ecosystem services in sub-Saharan Africa (SSA). Partners included, CIAT-TSBF, ISRIC, CIESIN, The Earth Institute at Columbia University and World Agroforestry (ICRAF). The Land Degradation Surveillance Framework (LDSF), developed at ICRAF, was employed to systematically sample key biopshyical indicators of land and soil health, across landscapes. LDSF sites were randomized using spatial stratification based on Koeppen-Geiger Climate zones across 19 countries in SSA. Soil samples were collected using the LDSF at two depths, 0-20 cm and 20-50 cm. Mid-infrared diffuse refectance spectroscopy (MIR) was a key soil characterization and screening tool in AfSIS. MIR measurements of the ~18,500 samples were centralized at ICRAF's Soil-Plant Spectral Diagnostics Laboratory based in Nairobi Kenya using the ...
This dataset contains the R code and XRPD data used to investigate data pre-treatments for approp... more This dataset contains the R code and XRPD data used to investigate data pre-treatments for appropriate cluster analysis of soil X-ray powder diffraction data. XRPD measurements are sourced from the Africa Soil Information Service (AfSIS).
New Zealand is undertaking a new phase of soil mapping (S-map) with the goal of national soil map... more New Zealand is undertaking a new phase of soil mapping (S-map) with the goal of national soil map coverage at 1:50 000 scale. One S-map objective is to develop a national capability in digital soil mapping (DSM). Concurrent development of the global soil map (GSM) project provides an opportunity for the Smap project to align its work with the GSM initiative both to achieve an enhanced national soil map and contribute New Zealand data to the GSM project. Consequently we have applied GSM method guidelines nationally. We matched the generic GSM methods to 193 New Zealand soilscapes. Soilscapes with similar levels of legacy soil information were grouped into eight soilscape groups, and DSM methods applied to them. Four soilscape groups were in low relief land where intensification of land use generates high demand for new soil information. The other four method classes were in rolling, hilly and mountainous land. The most extensive method in low relief land requires establishment of bet...
European Journal of Soil Science, 2020
Simple measures of appropriate levels of soil organic matter are needed for soil evaluation, mana... more Simple measures of appropriate levels of soil organic matter are needed for soil evaluation, management and monitoring, based on readily measurable soil properties. We test an index of soil organic matter based on the soil organic carbon (SOC) to clay ratio, defined by thresholds of SOC/clay ratio for specified levels of soil structural quality. The thresholds were originally delineated for a small number of Swiss soils. We assess the index using data from the initial sampling (1978–83) of the National Soil Inventory of England and Wales, covering 3,809 sites under arable land, grassland and woodland. Land use, soil type, annual precipitation and soil pH together explained 21% of the variance in SOC/clay ratio in the dataset, with land use the most important variable. Thresholds of SOC/clay ratio of 1/8, 1/10 and 1/13 indicated the boundaries between “very good”, “good”, “moderate” and “degraded” levels of structural condition. On this scale, 38.2, 6.6 and 5.6% of arable, grassland ...
Journal of the University of Ruhuna, 2016
Hyperion hyperspectral (VNIR-SWIR) satellite image data was used to map the spatial distribution ... more Hyperion hyperspectral (VNIR-SWIR) satellite image data was used to map the spatial distribution of mineral precipitates at Lake Magadi area, in the southernmost part of the eastern Kenya rift, Kenya. Mapping was coupled with laboratory analysis, including visible near infrared diffuse reflectance spectroscopy (VNIR) measurements and X-ray diffraction for selected rock and soil samples. The VNIR spectral responses of 92 rock and soil samples including trona, chert, diatomite, basalt/trachyte, erionite, Green bed and High Magadi bed were studied and identified. The spectral signatures of Chert samples show the broad Si-OH absorption feature at 2.2 μm while, Green bed, High Magadi bed and diatomite exhibit carbonate absorption feature at 2.35 μm with broad Si-OH absorption feature at 2.2 μm. Trona exhibits six common absorption features at 1.50, 1.74, 1.94, 2.03, 2.22 and 2.39 μm. These characteristics spectral absorption features together with the general shape of the spectral curve ...
Soil Science Society of America Journal, 1998
Abstract We hypothesized that trees can rapidly root into subsoil and capture NO 3, which can acc... more Abstract We hypothesized that trees can rapidly root into subsoil and capture NO 3, which can accumulate in the subsoil of agricultural soils with high anion sorption. The vertical distribution of root length and inorganic N (NO 3 and NH 4) to 3.95-m soil depth was compared for 11-mo-old stands of eucalyptus (Eucalyptus grandis W. Hill ex Maiden), sesbania [Sesbania sesban (L.) Merr.], calliandra (Calliandra calothyrsus Meissner), markhamia [Markhamia lutea (Benth.) Schumann], and grevillea (Grevillea robusta A. ...
Forest Ecology and Management, 1999
Phosphorus de®ciency is widespread in the subhumid highlands of eastern Africa but there are few ... more Phosphorus de®ciency is widespread in the subhumid highlands of eastern Africa but there are few data on the effect of P de®ciency on the growth of agroforestry tree species. We studied the effect of P application on growth, nutrient uptake and dry matter partitioning in young trees of Calliandra calothyrsus, Cedrela serrulata, Eucalyptus grandis, Grevillea robusta, Markhamia lutea, Senna spectabilis, and Sesbania sesban on a P-de®cient soil (Kandiudal®c Eutrudox, bicarbonate-EDTA extractable P 1 mg kg À1) in western Kenya. The trees were grown at two P levels (control and 500 kg added P ha À1) at 1 m 2 spacing in a randomized complete block design with three replications. Leaf K concentrations were in the low range for all species (5±9 mg g À1) and K de®ciency may have limited responses to P. Averaged over species, P addition increased aboveground shoot dry matter by a factor of 2.6 at 62 and 124 days, but the response decreased to 1.3 at 325 days. The increases at 62 days were large in sesbania (5.4) and eucalyptus (3.2) but small in calliandra (1.4) and markhamia (1.1). Relative response to P was more strongly correlated with shoot growth rate per unit root length among species than with shoot growth rate alone. Calliandra, which had high early growth rate but low response to added P, had an exceptionally high root length (6.0 km m À2) compared with the other species (0.3±2.1 km m À2). P addition increased N and P content but decreased ®nal shoot K content in sesbania and calliandra, and had little effect on K content in the other species. The high-yielding species (eucalyptus, sesbania and calliandra) accumulated more than 30 g N and 2 g P m À2 in shoots in 325 days of growth. The proportion of total shoot N in wood (branch stem) was in a higher range (67±75%) in the shrubby species (sesbania, calliandra, senna) than in the upperstorey tree species (38±43%). Slow early shoot growth relative to total root length, and high speci®c root length (root length per unit root mass) are proposed as criteria for the selection of species and provenances that are well adapted to P de®cient soils.
Experimental Agriculture, 1985
SUMMARYThe development and influence of crop drought stress was examined in winter-planted small-... more SUMMARYThe development and influence of crop drought stress was examined in winter-planted small-grain cereals under Mediterranean environmental conditions. In two average or wetter than average years crop drought stress in the grain-filling period was greatly exacerbated by the addition of fertilizer (N and P) and this usually resulted in significantly lower kernel weights. Yet this reduction was not so substantial as to seriously affect crop yields. Total grain yields were largest in treatments receiving fertilizer, due to greater spike numbers and numbers of kernels per spike. But heavy nitrogen applications may promote the risk of a large proportion of small or shrivelled grains in a dry year.
Biology and Fertility of Soils, 1996
A greenhouse experiment was conducted to assess the arbuscular mycorrhizal (AM) fungi inoculum po... more A greenhouse experiment was conducted to assess the arbuscular mycorrhizal (AM) fungi inoculum potential for Leucaena leucocephala in a range of nutrientdepleted farm soils in western Kenya. Leucaena was grown in 12 uninoculated farm soils with pH 5.0-6.7, with or without rock P and farmyard manure. Root infection, nodulation, and shoot and root weight were determined 25, 46, and 69 days after planting. Spore concentrations in the farm soils at the start of the experiment ranged from 44 to 126 live spores per 100 g dry soil with 6-10 species per soil, principally ScuteIlospora spp. and Acaulospora spp. Nodulation was absent or poor in all soils, indicating the need for rhizobial inoculation of species belonging to the leucaena crossinoculation group in this agrosystem. Rock-P alone increased final shoot dry weight by a factor of 1.4, manure alone by 1.8, and rock-P plus manure by 1.9, compared with no ameliorant. Root infection with AM fungi was detected in all soils 25 days after planting and increased linearly in the different farm soils to values of 33-65% 69 days after planting. Soil pH and root infection 25 days after planting accounted for much of the variation in final shoot weight among soils with no amelioront (87%). As early root infection increased from 20 to 40% at soil pH 5.0, the predicted final shoot dry weight was doubled, and the response to ameliorants was reduced by two-thirds. The growth responses to increased infection became smaller as pH increased from 5.0 to 6.5.
Agriculture, Ecosystems & Environment, 1998
A farm simulation model was designed to assess the long-term impact of existing soil management s... more A farm simulation model was designed to assess the long-term impact of existing soil management strategies, on farm productivity, pro®tability and sustainability. The model, which runs in time units of 1 year, links soil management practices, nutrient availability, plant and livestock productivity, and farm economics A case study is presented of the application of the model to existing, mixed farm systems in Vihiga district, in the highlands of western Kenya. Three representative farm types were developed using participatory techniques to re¯ect differences in resource endowments and constraints faced by farmers. The model was used to assess the sustainability of the existing systems for the three farm types as a basis for recommending improved practices for each. A summary model for calculating new sustainability indicators of soil productivity is presented. The low (LRE) and medium (MRE) resource endowment farms, which comprise about 90% of the farms in the area, have declining soil organic matter and low productivity and pro®tability. In contrast, the high resource endowment category of farms (HRE) have increasing soil organic matter, low soil nutrient losses and are productive and pro®table. Crop nutrient yields were 17, 19 and 86 kg N ha À1 year À1 on LRE, MRE and HRE farms, respectively. Soil C, N and P budgets were negative in LRE and MRE but positive in HRE. Farm revenue in LRE and MRE was 2±13% of farm revenue in HRE. It comprised 7% of household income in LRE compared with 25% in MRE and 63% in HRE. It is concluded that low land and capital resources constrain the adoption of ecologically and economically sustainable soil management practices on the majority of farms in the area. Strategies are needed to (i) increase the value of farm output (ii) increase high quality nutrient inputs at low cash and labour costs to the farmer, and (iii) increase off-farm income.
Analysis of spatial variation of soil fertility gradients in smallholdings in tropical Africa usi... more Analysis of spatial variation of soil fertility gradients in smallholdings in tropical Africa using geostatical techniques : Case studies from Vihiga and Siaya Districts of Western Kenya [abstract] [on line]. DSpace/Manakin Repository. ...
African Crop Science Journal, 2009
In many farming landscapes across Sub-Saharan Africa (SSA), soil fertility has been on the declin... more In many farming landscapes across Sub-Saharan Africa (SSA), soil fertility has been on the decline with significant implications on crop productivity. However, even with such a decline, soil nutrient levels still differ significantly between farms, fields or within the same field. Knowledge of such spatial variability and relationships among soil properties is important in implementation of agricultural land management practices. In this study, the spatial variability of soil organic carbon (SOC) in two districts of western Kenya was ...
Agricultural Systems, 2006
Arable land in western Kenya is under considerable pressure from increasing human population. Rur... more Arable land in western Kenya is under considerable pressure from increasing human population. Rural households depend on farming for at least part of their livelihood, and poverty rates are among the highest in Kenya. Land is often depleted of nutrients, and for most farmers, access to inputs and markets is poor. There is a need to identify options that are manageable within the context of the farmer's resource base and the household's objectives that could improve farm household well-being. In this study we integrated qualitative informal participatory approaches with quantitative mathematical programming and biophysical simulation modelling. Households in four sub-locations in Vihiga District were clustered and pilot cases identified. Meetings were held with farmers to elicit their perceptions of what their ideal farm would look like, and how its performance might compare with their own farm's performance. With farmers' help, a range of scenarios was analysed, relating to changes in current enterprise mixes, changes in current farm sizes, and changes in prices of staples foods and cash crops. A considerable mismatch was found between farmers' estimates of their own farm's performance, and what was actually produced. There seems to be a threshold in farm size of
Agriculture, Ecosystems & Environment, 2005
Strong gradients of decreasing soil fertility are found with increasing distance from the homeste... more Strong gradients of decreasing soil fertility are found with increasing distance from the homestead within smallholder African farms, due to differential resource allocation. As nutrient use efficiency varies strongly along these gradients, such heterogeneity must be considered when designing soil management strategies, aimed at an improved overall resource use efficiency at farm scale. Here, we quantify the magnitude and study the origin of farmer-induced, within-farm soil fertility gradients as affected by biophysical and socioeconomic conditions, and investigate farmers' perceptions of such heterogeneity. Farm transects, participatory resource flow mapping, farmers' classification of land qualities, and soil sampling for both chemical and spectral reflectance analyses were performed across 60 farms in three sub-locations (Emuhaia, Shinyalu, Aludeka) representing the variability found in the highlands of western Kenya. Differences between the various field types of a farm were observed for input use (e.g. 0.7-104 kg N ha À1), food production (e.g. 0.6-2.9 t DM ha À1), partial C (e.g. À570 to 1480 kg ha À1) and N (e.g. À92 to 57 kg ha À1) balances and general soil fertility status, despite strong differences across sub-locations. Concentration of nutrients in the home fields compared with the remote fields were verified for extractable P (e.g. 2.1-19.8 mg kg À1) and secondarily for exchangeable K (e.g. 0.14-0.54 cmol (+) kg À1), on average, whereas differences for soil C and N were only important when considering each individual farm separately. Farmers managed their fields according to their perceived land quality, varying the timing and intensity of management practices along soil fertility gradients. Fields classified by them as poor were planted later (up to 33.6 days of delay), with sparser crops (ca. 30% less plants m À2) and had higher weed infestation levels than those classified as fertile, leading to important differences in maize yield (e.g. 0.9 versus 2.4 t ha À1). The internal heterogeneity in resource allocation varied also between farms of different social classes, according to their objectives and factor constraints. Additionally, the interaction of sub-location-specific socioeconomic (population, markets
Current Opinion in Environmental Sustainability, Jun 1, 2017
Accountability and adaptive management of recent global agreements such as the Sustainable Develo... more Accountability and adaptive management of recent global agreements such as the Sustainable Development Goals and Paris Climate Agreement, will in part rely on the ability to track progress toward the social and environmental targets they set. Current metrics and monitoring systems, however, are not yet up to the task. We argue that there is an imperative to consider principles of coherence (what to measure), standardization (how to measure) and decision-relevance (why to measure) when designing monitoring schemes if they are to be practical and useful. New approaches that have the potential to match the necessary scale of monitoring, with sufficient accuracy and at reasonable cost, are emerging; although, they represent a significant departure from the historical norm in some cases. Iterative review and adaptation of analytical approaches and available technology will certainly be needed to continuously design ways to best track our progress.
PLOS ONE, Feb 2, 2022
Adept use of fertilizers is critical if sustainable development goal two of zero hunger and agroe... more Adept use of fertilizers is critical if sustainable development goal two of zero hunger and agroecosystem resilience are to be achieved for African smallholder agroecosystems. These heterogeneous systems are characterized by poor soil health mainly attributed to soil nutrient depletion. However, conventional methods do not take into account spatial patterns across geographies within agroecosystems, which poses great challenges for targeted interventions of nutrient management. This study aimed to develop a novel populationbased farm survey approach for diagnosing soil nutrient deficiencies. The approach embraces principles of land health surveillance of problem definition and rigorous sampling scheme. The advent of rapid soil testing techniques, like infrared spectroscopy, offers opportune avenues for high-density soil and plant characterization. A farm survey was conducted on 64 maize fields, to collect data on soil and plant tissue nutrient concentration and grain yield (GY) for maize crops, using hierarchical and purposive sampling. Correlations between soil test values with GY and biomass were established. The relationship between GY, soil NPK, and the tissue nutrient concentrations was evaluated to guide the setting up of localized critical soil test values. Diagnosis Recommendation Integrated System (DRIS) indices for total nitrogen (N), total phosphorus (P), and total potassium (K) were used to rank and map the prevalence of nutrient limitations. A positive correlation existed between plant tissue nutrient concentration with GY with R 2 values of 0.089, 0.033, and 0.001 for NPK, respectively. Soil test cutoff values were 0.01%, 12 mg kg-1 , 4.5 cmol c kg-1 for NPK, respectively, which varied slightly from established soil critical values for soil nutrient diagnostics. N and K were the most limiting nutrients for maize production in 67% of sampled fields. The study demonstrates that a population-based farm survey of crop fields can be a useful tool in nutrient diagnostics and setting priorities for site-specific fertilizer recommendations. A larger-scale application of the approach is warranted.
Field Crops Research, Aug 1, 2021
Abstract Determination of the concentration of nutrients in the plant is key information for eval... more Abstract Determination of the concentration of nutrients in the plant is key information for evaluating crop nutrient removal, nutrient use efficiency, fertilizer recommendations guidelines, and in turn for improving food security and reducing environmental footprints of crop production. Diffuse infrared (IR) reflectance spectroscopy is a powerful, rapid, cheap, and less pollutant analytical tool that could be substituted for traditional laboratory methods for the determination of the concentration of nutrients in plants. However, its accuracy for predicting the concentration of nutrients in rice plants is poorly known. This study aimed i) to determine macro- and micronutrients concentration that can be accurately predicted by near-infrared (NIR, 7498–4000 cm−1), mid-infrared (MIR, 4000–600 cm−1), or their combination (NIR-MIR, 7498–600 cm−1) spectra, ii) to identify the most suitable spectral range with the best prediction potential for the simultaneous analysis of nutrients concentrations in rice plants (straw and paddy) and iii) to assess the influence of agro-ecological zone and production system on nutrients concentrations in straw and paddy (unhulled grains) samples. Second-derivative spectra were fitted against plant laboratory reference data using partial least-squares regression (PLSR) to estimate six macronutrients (N, P, K, Ca, Mg, and S) and seven micronutrients (Na, Fe, Mn, B, Cu, Mo, and Zn) concentration in paddy and rice straw samples collected at harvest from 1628 farmers’ fields in 20 sub-Saharan African (SSA) countries. The modeling prediction potential was assessed by coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and ratio of performance to interquartile distance (RPIQ). Good prediction models (0.75
Uncertainty and risks are common elements of all major projects. Yet such uncertainty is rarely e... more Uncertainty and risks are common elements of all major projects. Yet such uncertainty is rarely effectively calculated when analysing project costs and benefits. This paper presents a Bayesian network (BN) modelling framework to calculate the costs, benefits, and return on investment of a project over a specified time period, allowing for changing circumstances and trade-offs. The framework uses hybrid and dynamic BNs containing both discrete and continuous variables over multiple time stages. The BN calculates costs and benefits based on multiple causal factors including the effects of individual risk factors, budget deficits, and time value discounting. The method is illustrated using a case study of an agricultural development project.
PLOS ONE, Jun 5, 2020
Agricultural development projects have a poor track record of success mainly due to risks and unc... more Agricultural development projects have a poor track record of success mainly due to risks and uncertainty involved in implementation. Cost-benefit analysis can help allocate resources more effectively, but scarcity of data and high uncertainty makes it difficult to use standard approaches. Bayesian Networks (BN) offer a suitable modelling technology for this domain as they can combine expert knowledge and data. This paper proposes a systematic methodology for creating a general BN model for evaluating agricultural development projects. Our approach adapts the BN model to specific projects by using systematic review of published evidence and relevant data repositories under the guidance of domain experts. We evaluate a large-scale agricultural investment in Africa to provide a proof of concept for this approach. The BN model provides decision support for project evaluation by predicting the value-measured as net present value and return on investment-of the project under different risk scenarios.
AfSIS Phase I was a collaborative projective funded by the Bill and Melinda Gates Foundation (BMG... more AfSIS Phase I was a collaborative projective funded by the Bill and Melinda Gates Foundation (BMGF), which aimed to provide a consistent baseline of soil information for monitoring soil ecosystem services in sub-Saharan Africa (SSA). Partners included, CIAT-TSBF, ISRIC, CIESIN, The Earth Institute at Columbia University and World Agroforestry (ICRAF). The Land Degradation Surveillance Framework (LDSF), developed at ICRAF, was employed to systematically sample key biopshyical indicators of land and soil health, across landscapes. LDSF sites were randomized using spatial stratification based on Koeppen-Geiger Climate zones across 19 countries in SSA. Soil samples were collected using the LDSF at two depths, 0-20 cm and 20-50 cm. Mid-infrared diffuse refectance spectroscopy (MIR) was a key soil characterization and screening tool in AfSIS. MIR measurements of the ~18,500 samples were centralized at ICRAF's Soil-Plant Spectral Diagnostics Laboratory based in Nairobi Kenya using the ...
This dataset contains the R code and XRPD data used to investigate data pre-treatments for approp... more This dataset contains the R code and XRPD data used to investigate data pre-treatments for appropriate cluster analysis of soil X-ray powder diffraction data. XRPD measurements are sourced from the Africa Soil Information Service (AfSIS).