Browsing by Author "Slater, B. K."

Now showing 1 - 2 of 2

Mapping numerically classified soil taxa in Kilombero Valley, Tanzania using machine learning
(Elsevier, 2016-11-24) Massawe, B. H. J.; Subburayalu, S. K.; Kaaya, A. K.; Winowiecki, L.; Slater, B. K.
Inadequacy of spatial soil information is one of the limiting factors to making evidence-based decisions to improve food security and land management in the developing countries. Various digital soilmapping (DSM) techniques have been applied inmany parts of theworld to improve availability and usability of soil data, but less has been done in Africa, particularly in Tanzania and at the scale necessary tomake farmmanagement decisions. The Kilombero Valley has been identified for intensified rice production. However the valley lacks detailed and up-todate soil information for decision-making. The overall objective of this study was to develop a predictive soilmap of a portion of Kilombero Valley using DSM techniques. Two widely used decision tree algorithms and three sources of Digital ElevationModels (DEMs) were evaluated for their predictive ability. Firstly, a numerical classification was performed on the collected soil profile data to arrive at soil taxa. Secondly, the derived taxawere spatially predicted and mapped following SCORPAN framework using Random Forest (RF) and J48 machine learning algorithms. Datasets to train the model were derived from legacy soil map, RapidEye satellite image and three DEMs: 1 arc SRTM, 30 m ASTER, and 12 m WorldDEM. Separate predictive models were built using each DEM source. Mapping showed that RF was less sensitive to the training set sampling intensity. Results also showed that predictions of soil taxa using 1 arc SRTM and 12mWordDEMwere identical.We suggest the use of RF algorithmand the freely available SRTMDEMcombination formapping the soils for thewhole Kilombero Valley. This combination can be tested and applied in other areas which have relatively flat terrain like the Kilombero Valley
Multi-criteria land evaluation for rice production using GIS and analytic hierarchy process in Kilombero Valley, Tanzania
(Tanzania Journal of Agricultural Sciences, 2019) Massawe, B .H. J.; Kaaya, A. K.; Winowiecki, L.; Slater, B. K.
A GIS-based multi-criteria land evaluation (MCE) was performed in Kilombero Valley, Tanzania to avail decision makers and farmers with evidence based decision support tool for improved and sustainable rice production. Kilombero valley has been identified by the government and investors for rice production intensification. Five most important criteria for rice production in the area were identified through literature search and discussion with local agronomists and lead farmers. The identified criteria were 1) soil properties, 2) surface water resources, 3) accessibility, 4) distance to markets, and 5) topography. Surveys, on-screen digitizations, reclassifications and overlays in GIS software were used to create spatial layers of the identified criteria. Analytic hierarchy process (AHP) method was used to score the criteria using local extension staff and lead farmers as domain experts on a scale of 0.0 – 1.0. Surface water resource scored the highest weight (0.462) followed by soil chemical properties (0.234). Other criteria and their weight in paranthesis are soil physical properties (0.19), topography (0.052), accessibility (0.036), and distance to market (0.025). The MCE results showed that about 8% of the study area was classified as having low suitability for rice production while only 2% was highly suitable. The majority of the area (about 89%) was classified as having medium suitability for rice production. Since the suitability decision was dominated by the surface water resource criterion, the rice suitability in the study area can be greatly improved by improving the water resources management.