Spatial predictions of soil macro and micro-nutrient content across Sub-Saharan Africa at 250 m spatial resolution and for 0-30 cm depth interval are presented. Predictions were produced for 15 target nutrients: organic carbon (C) and total (organic) nitrogen (N), total phosphorus (P), and extractable-phosphorus (P), potassium (K), calcium (Ca), magnesium (Mg), sulfur (S), sodium (Na), iron (Fe), manganese (Mn), zinc (Zn), copper (Cu), aluminum (Al) and boron (B). Model training was performed using soil samples from ca. 59,000 locations (a compilation of soil samples from the AfSIS, EthioSIS, One Acre Fund, VitalSigns and legacy soil data) and an extensive stack of remote sensing covariates in addition to landform, lithologic and land cover maps. An ensemble model was then created for each nutrient from two machine learning algorithms- random forest and gradient boosting, as implemented in R packages ranger and xgboost-and then used to generate predictions in a fully-optimized computing system. Cross-validation revealed that apart from S, P and B, significant models can be produced for most targeted nutrients (R-square between 40-85%). Further comparison with OFRA field trial database shows that soil nutrients are indeed critical for agricultural development, with Mn, Zn, Al, B and Na, appearing as the most important nutrients for predicting crop yield. A limiting factor for mapping nutrients using the existing point data in Africa appears to be (1) the high spatial clustering of sampling locations, and (2) missing more detailed parent material/geological maps. Logical steps towards improving prediction accuracies include: further collection of input (training) point samples, further harmonization of measurement methods, addition of more detailed covariates specific to Africa, and implementation of a full spatiotemporal statistical modeling framework.

No abstract

Background: Land degradation reduces agricultural productivity and poses a serious threat on food security status of households. In Ethiopia, farmers have been using only urea and di-ammonium phosphate for more than 15 years. Several reports that indicate lack of response to these fertilizers, which could be due to limitation of nutrients other than nitrogen and phosphorus. Therefore, the present study was initiated to evaluate the soil fertility status of ten sites in central highlands Vertisols of Ethiopia and wheat nutrient content. Results:The physico-chemical properties of soils showed that the soils were clayey in texture, neutral to slightly alkaline (pH 7.2-7.9) and low to medium in their organic matter (1.6-3.2 %) content. Total N content was low in 100 % of the samples while 80 % of the soil samples showed P deficiency (<10 mg kg −1 ). Exchangeable K, Ca and Mg in all soil samples were high, while available sulfur was low. On the other hand, K to Mg ratio varied from 0.13:1 to 0.44:1, indicating Mg induced K deficiency. All soil samples were adequate when analyzed by ammonium bicarbonate diethylene tri-amine penta-acetic acid extractable Cu (>0.5 mg kg ) and acid ammonium oxalate extractable Mo (<0.1 mg kg −1 ) were found to be low in all soil samples. The plant analysis data showed that all samples were low in N, P and K, while high in Ca and Mg concentrations. The deficiency of tissue K content was not predicted by the soil exchangeable K test. Plant micronutrient analysis showed that Cu, Fe, Mn and Cl concentrations were within the sufficiency range while Zn was deficient in all of the samples. Conclusions:Soil and/or tissue test results are indicative of deficiency of N, P, K, S, Zn, B and Mo that could be amended by fertilizer application, although more data are needed to thoroughly support this conclusion. The highest correlation (r > 0.90) between soil and plant nutrient content was observed for P, K, Mg, Cu, Fe, Mn and Zn, implying that flag leaves at flowering stage can be used to calibrate soil and plant contents for the deficient nutrients.

Background: Understanding the landscape features of agricultural lands and soil management practices is pertinent to verify the potential and limitations of the soil resources; and devise relevant land management strategies. However, information is lacking in Wolaita, Southern Ethiopia. Thus, this study aimed at investigating the physiographic characteristics of agricultural lands, farmers' soil fertility management practices and their influences on soil fertility and crop productivity was conducted. The survey involved 789 randomly sampled agricultural lands. Soil sample as well as data on slope, topography, land use, crop rotation, fallowing, cropping intensity, crop residue management, fertilizer use and farmers' estimated crop yield were collected and evaluated. Results:The result indicated that agriculture has been practiced under diverse slope positions (1-58%). This significantly influenced most of soil physico-chemical properties in which an increase in soil bulk density, and decrease in available phosphorous, exchangeable calcium, extractable boron, copper, iron and zinc with an increase in slope positions were recorded. Furthermore, continuous cultivation without any fallow periods and complete crop residue removal (96%) were common practices. Fertilizer use was not sufficient. The average use (kg ha −1 ) was 30 and 9 for diammonium phosphate (DAP) and urea, respectively, and about 89% of study area received manure below 1.5 t ha −1 . All these limitations made farmers to experience lower crop yield than the national average estimates. Conclusion:Based on the result, it is possible to conclude that the following soil management interventions such as soil conservation, application of sufficient organic and inorganic fertilizers are recommended to restore the soil fertility and improve crop productivity.

No abstract