Comparing Four Types Methods for Karst NDVI Prediction Based on Machine Learning

Yuju Ma,Lulu Liu,Qiang Liu,Liyuan Zuo,Jiangbo Gao

doi:10.3390/atmos12101341

Abstract

As a link for energy transfer between the land and atmosphere in the terrestrial ecosystem, karst vegetation plays an important role. Karst vegetation is not only affected by environmental factors but also by intense human activities. The nonlinear characteristics of vegetation growth are induced by the interaction mechanism of these factors. Previous studies of this relationship were not comprehensive, and it is necessary to further explore it using a suitable method. In this study, we selected climate, human activities, topography, and soil texture as the response factors; a nonlinear relationship model between the karst normalized difference vegetation index (NDVI) and these factors was established by applying a back propagation neural network (BPNN), a radial basis function neural network (RBFNN), the random forest (RF) algorithm, and support vector regression (SVR); and then, the karst NDVI was predicted. The coefficient of determination (R2), mean square error (MSE), root mean square error (RMSE), and mean absolute percentage error (MAPE) of the obtained results were calculated, and the mean R2 values of the BPNN, RBFNN, RF, and SVR models were determined to be 0.77, 0.86, 0.89, and 0.91, respectively. Compared with the BPNN, RBFNN, and RF models, the SVR model had the lowest errors, with mean MSE, RMSE, and MAPE values of 0.001, 0.02, and 2.77, respectively. The results show that the BPNN, RBFNN, RF, and SVR models are within acceptable ranges for karst NDVI prediction, but the overall performance of the SVR model is the best, and it is more suitable for karst vegetation prediction.

Highlights

The karst landscape in southwestern China developed on soluble carbonate bedrock, and it is one of the largest contiguous karst regions in the world [1,2,3]
2.4604–5.6112, respectively, with averages of 0.0017, 0.0384, and 4.4570, respectively, which indicate that the back propagation neural network (BPNN) model is acceptable for karst normalized difference vegetation index (NDVI) prediction
Different from the above research, our research found the support vector regression (SVR) model has the lowest errors and the highest accuracy compared with the BPNN, radial basis function neural network (RBFNN), and random forest (RF) models, which was indicating that the SVR model was the best for karst vegetation prediction

Summary

Introduction

The karst landscape in southwestern China developed on soluble carbonate bedrock, and it is one of the largest contiguous karst regions in the world [1,2,3]. The vegetation cover is relatively low because of the slow formation rate and shallow depth of the soil in the karst region [4]. Karst vegetation is the most fundamental component of the terrestrial ecosystem, and it provides a great carbon sink function and a series of ecological services [5,6,7]. It is necessary to monitor and predict the dynamic changes in karst vegetation. Karst vegetation is affected by environmental factors (climate, soil, topography) and by human activities. Karst vegetation is highly sensitive to human activities and climate change [8]. The effects of human activities on karst vegetation have positive and negative impacts. With the rapid development of the economy, most farmers have reduced their dependence on cropland and reduced the impact of unreasonable human activities on the ecological environment; in addition, the government

Methods

Results

Conclusion