Jérôme E. Tondoh

Mapping Soil Properties of Africa at 250 m Resolution: Random Forests Significantly Improve Current Predictions

80% of arable land in Africa has low soil fertility and suffers from physical soil problems. Additionally, significant amounts of nutrients are lost every year due to unsustainable soil management practices. This is partially the result of insufficient use of soil management knowledge. To help bridge the soil information gap in Africa, the Africa Soil Information Service (AfSIS) project was established in 2008. Over the period 2008–2014, the AfSIS project compiled two point data sets: the Africa Soil Profiles (legacy) database and the AfSIS Sentinel Site database. These data sets contain over 28 thousand sampling locations and represent the most comprehensive soil sample data sets of the African continent to date. Utilizing these point data sets in combination with a large number of covariates, we have generated a series of spatial predictions of soil properties relevant to the agricultural management—organic carbon, pH, sand, silt and clay fractions, bulk density, cation-exchange capacity, total nitrogen, exchangeable acidity, Al content and exchangeable bases (Ca, K, Mg, Na). We specifically investigate differences between two predictive approaches: random forests and linear regression. Results of 5-fold cross-validation demonstrate that the random forests algorithm consistently outperforms the linear regression algorithm, with average decreases of 15–75% in Root Mean Squared Error (RMSE) across soil properties and depths. Fitting and running random forests models takes an order of magnitude more time and the modelling success is sensitive to artifacts in the input data, but as long as quality-controlled point data are provided, an increase in soil mapping accuracy can be expected. Results also indicate that globally predicted soil classes (USDA Soil Taxonomy, especially Alfisols and Mollisols) help improve continental scale soil property mapping, and are among the most important predictors. This indicates a promising potential for transferring pedological knowledge from data rich countries to countries with limited soil data.

Functions of farmers\' preferred tree species and their potential carbon stocks in southern Burkina Faso: implications for biocarbon initiatives

The success of terrestrial carbon sequestration projects for rural development in sub-Saharan Africa lies in the (i) involvement of local populations in the selection of woody species, which represent the biological assets they use to meet their daily needs, and (ii) information about the potential of these species to store carbon. Although the latter is a key prerequisite, there is very little information available. To help fill this gap, the present study was undertaken in four pilot villages (Kou, Dao, Vrassan and Cassou) in Ziro Province, south-central Burkina Faso. The objective was to determine carbon storage potential for top-priority woody species preferred by local smallholders. We used (i) participatory rural appraisal consisting of group discussions and key informant interviews to identify priority species and functions, and (ii) landscape assessment of carbon stocks in the preferred woody species. Results revealed over 79 priority tree and shrub species grouped into six functions, of which medicine, food and income emerge as the most important ones for the communities. For these functions, smallholders overwhelmingly listed Vitellaria paradoxa, Parkia biglobosa, Afzelia africana, Adansonia digitata, Detarium microcarpum, and Lannea microcarpa among the most important tree species. Among the preferred woody species in Cassou and Kou, the highest quantity of carbon was stored by V. paradoxa (1,460.6 ±271.0 kg C ha−1 to 2,798.1±521.0 kg C ha−1) and the lowest by Grewia bicolor (1.6±1.3 kg C ha−1). The potential carbon stored by the preferred tree communities was estimated at 5,766.2 Mg C ha−1 (95% CI: 5,258.2; 6,274.2 Mg C ha−1) in Kou and 6,664.0 Mg C ha−1 (95% CI: 5,810.2; 7,517.8 Mg C ha−1) in Cassou. The findings of this study will help design data-based development of biocarbon projects, which are rare in the West African Sahel despite being considered as one of the most impactful climate change resilient strategies.

Intensifying Maize Production Under Climate Change Scenarios in Central West Burkina Faso

Combination of poor soil fertility and climate change and variability is the biggest obstacle to agricultural productivity in Sub-Saharan Africa. While each of these factors requires different promising adaptive and climate-resilient options, it is important to be able to disaggregate their effects. This can be accomplished with ordinary agronomic trials for soil fertility and climate year-to-year variability, but not for long-term climate change effects. In turn, by using climate historical records and scenario outputs from climate models to run dynamic models for crop growth and yield, it is possible to test the performance of crop management options in the past but also anticipate their performance under future climate change or variability. Nowadays, the overwhelming importance given to the use of crop models is motivated by the need of predicting crop production under future climate change, and outputs from running crop models may serve for devising climate risk adaptation strategies. In this study we predicted yield of one maize variety named Massongo for the time periods 1980–2010 (historical) and 2021–2050 (2030s, near future) across agronomic practices including the fertilizer input rates recommended by the national extension services (28 kg N, 20 kg P, and 13 kg K ha−1). The performance of the crop model DSSAT 4.6 for maize was first evaluated using on-farm experimental data that encompassed two seasons in the Sudano-Sahelian zone in six contrasting sites of Central West Burkina Faso. The efficiency of the crop model was evidenced by reliable simulations of total aboveground biomass and yields after calibration and validation. The root-mean-square error (RMSE) of the entire dataset for grain yield was 643 kg ha−1 and 2010 kg ha−1 for total aboveground biomass. Three regional climate change projections for Central West Burkina Faso indicate a decrease in rainfall during the growing period of maize. All the three scenarios project that the decrease in rainfall is to the tune of 3–9% in the 2030s under RCP4.5 in contrast to climate scenarios produced by the regional climate model GCM ICHEC-EC-Earth which predicted an increase of rainfall of 25% under RCP8.5. Simulations using the CERES-DSSAT model reveal that maize yields without fertilizer show the same trend as with fertilizer in response to climate change projections across RCPs. Under RCP4.5 with output from the climate model ICHEC-EC-Earth, yield can slightly increase compared to the historical baseline on average by less than 5%. In contrast, under RCP8.5, yield is increased by 13–22% with the two other climate models in fertilized and non-fertilized plots, respectively. Nevertheless, the average maize yield will stay below 2000 kg ha−1 under non-fertilized plots in RCP4.5 and with recommended mineral fertilizer rates regardless of the RCP scenarios produced by ICHEC-EC-Earth. Giving the fact that soil fertility improvement alone cannot compensate for the adverse impact of future climate on agricultural production particularly in case of high rainfall predicted by ICHEC-EC-Earth, it is recommended to combine various agricultural techniques and practices to improve uptake of nitrogen and to reduce nitrogen leaching such as the splitting of fertilizer applications, low-release nitrogen fertilizers, agroforestry, and any other soil and water conservation practices.