Spatial modelling of potential suitable areas for cultivation of tomatoes (Solanum Lycopersicum) in Limpopo Province under climate change scenarios

Abstract

Tomatoes (Solanum Lycopersicum) are cultivated globally by commercial and subsistence farmers due to their major role in eradicating poverty and sustaining food security. Tomato production, however, is limited to a specific set of environmental and topo-edaphic factors while highly sensitive to unstable and unpredicted fluctuations in climatic conditions. It is, therefore, imperative to identify areas characterised by optimum biophysical conditions to grow tomato crops with maximum yield potential and minimum risk of crop failure. In light of this, the study aimed to model potentially suitable areas for the cultivation of tomatoes in Limpopo Province under current climatic conditions and future climate change scenarios. 19 bio-climatic and topo-edaphic variables were used to predict current suitable areas using the MaxEnt and Random Forest models, while climate change scenarios, SSP2-4.5 and SSP5-8.5, were used for future prediction. The results reveal that the lowveld region, which is characterised by arid conditions and summer rainfall, has high to very high potential for tomato cultivation. Meanwhile, the middleveld and highveld were consistently classified as having low potential, with moderate, high, and very high in smaller patches when compared to the lowveld. Additionally, there is an increase in low potential under predicted future conditions that worsen under a high emission scenario. Permutation of importance analysis was used to evaluate variable importance, while Area under the curve (AUC) values from the receiver operating characteristics (ROC) were used to evaluate the models’ performance for 1970-2000, 2020-2040, and 2040-2060. The results revealed the dominant role played by climate factors, while topo-edaphic factors proved to be important but not similar to climatic factors. Meanwhile, the models showed good to excellent discrimination performance for MaxEnt with an AUC of 0.637- 0.909 and RF with an AUC of 0.778 - 0.904. The generated suitability maps will play a critical role in informing decision-making, promoting sustainable farming by guiding cultivation toward ecologically appropriate zones, reducing land degradation, overuse of inputs, and deforestation.