Assessing trends and predicting the cover management factor in a tropical island state using Enhanced Vegetation Index

Abstract

Soil erosion is a complex process involving multiple factors that contribute to the amount of soil loss. The amount of vegetation cover is one of the main factors used to estimate soil loss and is an important risk factor in informing land use management and soil conservation policies. Cover/crop management (C-factor) is a dynamic soil loss factor and analysing C-factor trends in the context of both space and time, to build a multi-dimensional data structure, increased the value of the trend analysis. The objectives of the study were to (1) utilise EVI dependent static and dynamic predictive equations to compute the C-factor, and (2) to investigate the spatio-temporal changes of the C-factor and hotspots in a tropical small island developing state (SIDS) from 2010 to 2019. ArcGIS Model Builder was utilised to automate the computation of the C-factor using Moderate Resolution Imaging Spectroradiometer (MODIS) Enhanced Vegetation Index (EVI) data and compute the ordinary least square regression. Spatio-temporal analysis was performed using the novel emerging hot spot analysis in ArcGIS to identify statistically significant hot and cold spot trends of cover management to locate new, intensifying, persistent, or sporadic hot spot patterns at different time-step intervals. The regression output for the C-factor and EVI values indicated a strong r2, explaining on average 90% of the models. There was no statistically significant (P value > 0.05) increase in C-factor trend over a 10 year period (2010–2019, P-value = 0.92), a 5 year period (2015–2019; P-value = 0.59), or a 3 year period (2017–2019; P-Value = 0.31). Our results showed that intensifying hot spots (27%) were concentrated along the north–south corridor of the study area, highlighting areas with a statistically significant increase of the C-factor, thus a reduction in vegetation cover over the study period. Integrating spatio-temporal data and spatial technology provided vital cover management estimates for soil practitioners and farmers to guide conservation strategies.