Exploring Spatial Influence of Remotely Sensed PM2.5 Concentration Using a Developed Deep Convolutional Neural Network Model.

Junming Li,Meijun Jin,Honglin Li

doi:10.3390/ijerph16030454

Abstract

Currently, more and more remotely sensed data are being accumulated, and the spatial analysis methods for remotely sensed data, especially big data, are desiderating innovation. A deep convolutional network (CNN) model is proposed in this paper for exploiting the spatial influence feature in remotely sensed data. The method was applied in investigating the magnitude of the spatial influence of four factors—population, gross domestic product (GDP), terrain, land-use and land-cover (LULC)—on remotely sensed concentration over China. Satisfactory results were produced by the method. It demonstrates that the deep CNN model can be well applied in the field of spatial analysing remotely sensed big data. And the accuracy of the deep CNN is much higher than of geographically weighted regression (GWR) based on comparation. The results showed that population spatial density, GDP spatial density, terrain, and LULC could together determine the spatial distribution of annual concentrations with an overall spatial influencing magnitude of 97.85%. Population, GDP, terrain, and LULC have individual spatial influencing magnitudes of 47.12% and 36.13%, 50.07% and 40.91% on annual concentrations respectively. Terrain and LULC are the dominating spatial influencing factors, and only these two factors together may approximately determine the spatial pattern of annual concentration over China with a high spatial influencing magnitude of 96.65%.

Highlights

Remote sensing technology has developed rapidly since the 1960s [1], and an abundance of remote sensing data has been accumulated in this 50-year period
This study proposed a spatial analysis method that exploits the spatial influencing feature of remotely sensed data based on the deep convolutional network (CNN)
This study aimed to present a deep CNN model exploiting the magnitude of spatial influence of four factors—population, gross domestic product (GDP), terrain, and land-use and land-cover (LULC)—to remotely sense the annual mean concentration of PM2.5 over China

Summary

Introduction

Remote sensing technology has developed rapidly since the 1960s [1], and an abundance of remote sensing data has been accumulated in this 50-year period. The mainstream classical spatial analysis models, e.g., spatial lag model [10,11], spatial error model [10,11], and Bayesian spatial regression model [12], can only evaluate the overall or average linear correlation feature over a whole study region, neglecting the details of local area. These methods ignore the consequences of spatial heterogeneity [13].

Objectives

Methods

Discussion

Conclusion