Spatial Variation Of Soil Acidification,Nutrient Balance And Their Improvements In Sub-Saharan Africa Cropland

Feeding the world’s growing population is one of the most pressing problems of this century where global population is expected to be 9.4–10.1 billion in 2050 and 9.4–12.7billion by 2100.Soil acidification and soil nutrient depletion is the main problem of soil fertility of Sub Saharan Africa cropland（SSA）,where considerable high percentage of the people derives their living directly from the soil.Therefore,objectives of this study were:（a）to predict spatial variation of soil nutrient balance in East Africa highland（Rwanda）,（b）to predict spatio temporal change of soil pH from 1980 to 2050,（c）to determine drivers of soil acidification in SSA cropland,（d）to develop the framework for determining the exact spatial lime requirement for different cropping systems in SSA cropland and（e）to assess the combined effect of nitrogen（N）,phosphorus（P）,and potassium（K）chemical fertilizers and Biochar on soil fertility and maize growth in highly weathered soils（moderately and strongly acidic soils）.Integration of machine learning（Random Forest（"rf"）,Gradient Boosting（"gbm"）,e Xtreme Gradient Boosting（"xgb DART"）,and Support Vector Machines（"svm Radial"）algorithms）and existing methods for soil nutrient balance（soil nutrient balance equation）,soil acidification（soil acidity budget）and lime requirement（lime requirement based on the adjustment of aluminium saturation）were used.The pot experiment was used to assess the effect of different fertilizers and Biochar on soil fertility and maze growth.The main conclusions include:1.The results of soil nutrient balance for 2019–2020 growing season in Rwanda revealed that N balance ranged from–234 to 21.37 kg N ha^-1 yr^-1 with mean N balance of–33.60 kg N ha^-1 yr^-1,P balance ranged from–30 to–21.71 kg P ha^-1 yr^-1 with mean P balance of 2.31 kg P ha^-1yr^-1 and K balance ranged from–254.11 kg K ha^-1 yr^-1 to–6.02 kg K ha^-1 yr^-1 with mean K balance of–71.00 kg K ha^-1 yr^-1.High soil nutrient uptake,high soil nutrient loss due to erosion and leaching were main drivers of NPK depletion.Spatial variations of NPK balance were influenced by soil nutrient stocks,soil erosion,elevation,rainfall,soil texture,and soil bulk density.The 10-fold cross-validation showed that coefficients of determination（R²）of NPK models were 62%,58%,and 58%,respectively.Compared to single models,ensemble machine learning improved NPK model accuracy up to 5%.We conclude that increasing soil nutrient inputs without reducing soil nutrient loss due to soil degradation will not decrease soil nutrient depletion in Rwanda and ensemble machine learning outperforms single models in predicting soil nutrient balance.The solution to reduce high soil nutrient depletion in all agro ecological zones of Rwanda would be to prioritize soil and water conservation measures and increase soil nutrient inputs.2.The topsoil will become increasingly acidic because of the dramatic increase in N fertiliser in in SSA cropland in future.In the last 42 years,mean annual soil pH decline in SSA cropland was 0.014 pH unit with less than 15 kg N ha^-1 yr^-1.Under BAU scenario,soil pH decline is expected to be 0.024 pH unit from 2022-2050(with 37.5 kg N ha^-1 yr^-1)and 0.048pH unit by 2050 under Eq D scenario(with 143.40 kg N ha^-1 yr^-1)in 2050.From 1980-2022,croplands with soil pH<6.5 have declined significantly,and soil acidification is predicted to become severe by 2050 in the BAU and Eq D scenarios.This was indicated by 3%increase in croplands under risk of aluminium toxicity（soil pH<5.5）from 66×10⁶ ha in 2022 to 78.5×10⁶ ha in 2050.Root mean squared errors（RMSEs）of soil pH model from 1980 to 2022,2022to 2050（BAU）and（Eq D）models were 0.53 pH units,0.54 pH units,and 0.56 pH units respectively,with coefficients of determination（R²）of 0.63,0.64,and 0.66.The findings of this study can be used for the establishment of management strategies for increasing inorganic nitrogen fertilizer use in acidic soils.3.Spatial variation of soil acidification under optimized fertilizer use in 16 countries revealed that soil acidification（H⁺）ranged from 0 to 16 keq H⁺ha^-1 yr^-1.The most protons（H⁺）were produced by cassava,banana,and Irish potatoes systems with 12.0,9.8,and 8.9 keq H⁺ha^-1 yr^-1 respectively.Net basic cation loss was identified as the main driver of soil acidification under optimized fertilizer use in 16 countries studies.This net basic cation loss was caused by high caused by high basic cation removal,high basic cation leached due to high precipitation in tropical climate.The results of the 10-fold cross validation for the soil acidification model were a coefficient of determination（R ²）of 0.60,a root mean square error（RMSE）of 2.1,and a mean absolute error（MAE）of 1.4.Relative importance analysis revealed that climate variables were identified as main drivers of soil acidification than soil properties,anthropogenic activities,terrain features and parent material covariates in 16 countries studied.Digital soil mapping can produce soil acidification maps for sustainable land use and management plans.4.Lime requirement（LR）prediction revealed that mean of LR Mg Ca CO₃（1 Mg=10⁶g）ha^-1 for cereal crops were 6.34,6.35,and 4.41 for maize,sorghum,and upland rice,respectively.Mean of LR(Mg ha^-1)for pulses were 6.28,5.19,and 4.90 for common beans,soybeans,and cowpeas,respectively.Mean of LR Mg Ca CO₃（1 Mg=10⁶ g）ha^-1 for cash crops were 3.41 and 6.29 for coffee and cotton,respectively.Spatial variation showed that LR in croplands was higher in tropical humid regions than in semi-arid and arid regions and ranged from 0 to 8.8 Mg ha^-1.The results of 10-fold cross validation for high model performance of LR for tested crops were coefficient of determination（R²）of 0.61,a root mean square error（RMSE）of 0.5,and a mean absolute error（MAE）of 0.31,maize LR with RMSE=0.9,MAE=0.24,and R²=0.51,and cotton LR with RMSE=0.5,MAE=0.31,and R²=0.60.We recommend predicting lime requirement in acidic soils of Sub-Saharan Africa by adjusting Al saturation up to the tolerance of the grown crop,updating soil surveys in Sab Saharan Africa,and using digital soil mapping to monitor soil acidity and lime requirement.5.The results of pot experiment showed that the pH of the NPK+Biochar treatment（7.8）was higher in moderately acidic soils than the control（6.1）,but the pH of the soil in the NPK treatments did not vary significantly.This observation was also observed in strong acidic soils where there was an increase of 1.8 pH unit due to the combination of NPK+B compared to soil pH in control treatment（4.7）.NP,NPK and NPK+B treatments had higher total dry biomass than control in both soils.Combined NPK+B had effect on total dry biomass only in strong acidic soils.NPK+B treatment increased total nitrogen compared to all other treatments in both soils.N uptake was also increased significantly in NPK+B in strong acidic soils but in moderately soil N uptake was higher in NPK treatments compared to other treatments.In moderately acidic soil,P treatment had higher NUE with 69%while NPK and NPK+B decreased NUE with 49%and 48%respectively.However in strong acidic soils,P treatment had higher NUE（46%）followed NPK+B treatment with 44%.We conclude that combined NPK+B in strong acidic soils is recommended in improving soil pH,total N,biomass and NUE while Biochar had a negative effect in soil with moderately to high soil pH（pH>5.5）.