Development of a multiscale discretization method for the geographical detector model

ABSTRACT

The geographical detector model (GDM) is based on the spatial variance analysis of geographical strata of variables to assess the association between the independent variables ($X$) and dependent variables ($Y$). The independent variables of the GDM must be discretized into classes. However, current discretization methods employ univariate analysis, which may lead to inaccurate results. The aim of this study was to develop a novel bivariate optimal discretization approach, known as the multiscale discretization (MSD) method. The objective of the MSD method is to determine an appropriate set of thresholds for $X$, thereby minimizing the variance of $Y$ within the spatial partitions determined by the discrete $X$. We successfully applied the MSD method to assess the relationship between the precipitation and enhanced vegetation index on the African continent, as well as the habitat range of pandas in Ya’an County, Sichuan Province, China. The results demonstrate that the MSD is a feasible, robust, and rapid method for converting continuous data into discrete data, with globally optimal discretization results. Furthermore, the MSD method can evaluate the degree of association between $X$ and $Y$ more accurately, and can optimize the results of the GDM.

KEYWORDS:

• Geographical detector model
• discretization
• upscaling
• downscaling
• multiscale discretization

Acknowledgments

We thank the anonymous reviewers for their constructive comments on the manuscript.

Data and codes availability statement

The codes of MSD method are available at https://doi.org/10.6084/m9.figshare.13643318. For related software and reference information of the GDM, readers can refer to http://geodetector.cn/. The enhanced vegetation index (EVI) data is available from https://lpdaac.usgs.gov/. The Tropical Rainfall Measuring Mission (TRMM) data is available from https://trmm.gsfc.nasa.gov/. The MODIS Land Cover Type data (MCD12Q1) is available from https://lpdaac.usgs.gov/products/mcd12q1v006/. The TerraClimate data is available from https://climate.northwestknowledge.net/TERRACLIMATE/. The Shuttle Radar Topography Mission (SRTM) digital elevation data is available from http://srtm.csi.cgiar.org/. The statistical functions website of SciPy is https://docs.scipy.org/doc/scipy/reference/stats.html.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was funded by the Strategic Priority Research Program of the Chinese Academy of Sciences (No. XDA20030102) and the National Natural Science Foundation of China (No. 41861144020).

Notes on contributors

Xiaoyu Meng

Xiaoyu Meng is a PhD candidate at the Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences and the University of Chinese Academy of Sciences. His research interests include spatial analysis, ecological remote sensing, machine learning and geographic information science.

Xin Gao

Xin Gao is a Professor at the Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences. His research interests include aeolian physics, spatial analysis,  prevention and control of desertification and ecological restoration.

Jiaqiang Lei

Jiaqiang Lei is a Professor at the Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences. His research interests include prevention and control of desertification, soil erosion, and ecological restoration in arid land.

Shengyu Li

Shengyu Li is a Professor at the Xinjiang Institute of Ecology and Geography, Chinese Academy of Sciences. His research interests include prevention and control of desertification, and vegetation physiology and ecology in arid land.