A simple model to predict the spatiotemporally vegetation dynamics in terms of precipitation and temperature.

Abstract

Climate factors like precipitation and temperature fulfil a significant role in assessing the status of the vegetation. Vegetation prediction in terms of climatic conditions represents a critical issue because it helps monitor and maintain the ecosystem's well-being by maintaining track of changes in vegetation disturbance. It is challenging to locate large-scale current vegetation data for use in a variety of applications. In this study, temperature and precipitation were applied to develop a simple linear regression model that predict Normalized Difference Vegetation Index (NDVI) over the African continent. For training data, the third-general Normalized Difference Vegetation Index (NDVI3g) generated from Global Inventory Modeling and Mapping Studies (GIMMS) at a 1/12° resolution was the main input for simulation from 1982 to 2005. From 2006 to 2015, the third-general Normalized Difference Vegetation Index was used to validate the model outputs. The model's accuracy in the arid climate zone (poor vegetation areas) is stronger than in the other four climate zones of Africa, with an R2 of 74%, the lowest p-value of 0.001, the lowest root mean square error (RMSE) of 0.027 and the lowest mean absolute error (MAE) of 0.022. Based on future vegetation condition projections, this approach will assist a more effective management of the future food reservations for the battle against hunger. It will also help with the design, management, and execution of ecological restoration initiatives, research data availability as well as understanding the relationship between vegetation variation and climate variability.