Desertification Information Extraction Based on Feature Space Combinations on the Mongolian Plateau

Abstract

The Mongolian plateau is a hotspot of global desertification because it is heavily affected by climate change, and has a large diversity of vegetation cover across various regions and seasons. Within this arid region, it is difficult to distinguish desertified land from other land cover types using low-quality vegetation information. To address this, we analyze both the effects and the applicability of different feature space models for the extraction of desertification information with the goal of finding appropriate approaches to extract desertification data on the Mongolian plateau. First, we used Landsat 8 remote sensing images to invert NDVI (normalized difference vegetation index), MSAVI (modified soil adjusted vegetation index), TGSI (topsoil grain size index), and albedo (land surface albedo) data. Then, we constructed the feature space models of Albedo-NDVI, Albedo-MSAVI, and Albedo-TGSI, and compared their extraction accuracies. Our results show that the overall classification accuracies of the three models were 84.53%, 85.60%, and 88.27%, respectively, indicating that the three feature space models are feasible for extracting information relating to desertification on the Mongolian plateau. Further analysis indicates that the Albedo-NDVI model is suitable for areas with a high vegetation cover or a high forest ratio, whilst the Albedo-MSAVI model is suitable for areas with relatively low vegetation cover, and the Albedo-TGSI model is suitable for areas with extremely low vegetation cover, including the widely distributed Gobi Desert and other barren areas. This study provides a technical selection reference for the investigation of desertification of different zones on the Mongolian plateau.

Desertification detection model in Naiman Banner based on the albedo-modified soil adjusted vegetation index feature space using the Landsat8 OLI images

“…In this study, the desertification index with respect to Naiman Banner was calculated using two categories of feature space models. The desertification indexes of the two categories of the feature space models were divided into five categories (Table 1, Wei et al 2018) for obtaining the spatial distributions of different levels of desertification (Figure 8) using the natural breaks method of ArcGIS 10.2, which can completely consider the image histogram distribution o and the spatial clustering characteristics of the features. 245 sites from regions exhibiting different types of landscape were selected by utilizing Google Earth and field observations to validate the reversion and classification results for the two categories of the feature space models (Tables 2 and 3).…”

“…The two aforementioned sensitive desertification indices were calculated using the reflectance data of blue (B blue ), red (B red ), near-infrared (B nir ), and short-wave infrared band (B swir1 and B swir2 ) obtained from Landsat8 OLI (August 29, 2017; path/row, 121/30). The formulas for calculating albedo and MSAVI are as follows (Wei et al 2018):…”

“…Further, Ma et al (2011) determined the quantitative relation between albedo and NDVI in case of different desertification levels using statistical regression analysis. Wei et al (2018) analyzed the effects and applicability of different feature space models with respect to the extraction of the desertification information of the Mongolian plateau based on four sensitive parameters, i.e., NDVI, topsoil grain size index, and land-surface albedo. Wu, Lei, et al (2019) proposed a desertification monitoring index based on the albedo-modified soiladjusted vegetation index (MSAVI) feature space for semi-arid steppe regions and obtained improved results.…”

Special issue «environmental change on the Mongolian plateau: atmosphere, forests, soils and water»

“…Land cover changes in the region are related to both climate change Miao et al 2015) as well as anthropogenic impacts related to urbanization (Allington et al 2017;Fan et al 2016), livestock herding (Allington et al 2017, Sternberg 2012, mining (Batbayar et al 2019;Jarsjö et al 2017) and logging (Batkhuu et al 2011;Tsogtbaatar 2004). In vast parts of the Mongolian Plateau, forest degradation and losses (Gradel et al 2017;Juřička et al 2019a) as well as desertification of grasslands (Khodolmor et al 2013;Wei et al 2019) have modified a natural land cover. The key drivers of these processes are mining, agriculture / urbanization and deforestation, which are at the same time major water users and polluters (Batbayar et al 2019;Jarsjö et al 2017;Karthe et al 2017).…”

“…The key drivers of these processes are mining, agriculture / urbanization and deforestation, which are at the same time major water users and polluters (Batbayar et al 2019;Jarsjö et al 2017;Karthe et al 2017). The above mentioned processes have also had strong effects on the soils, which are affected by anthropogenic impacts in three major ways: nutrient depletion due to intensive agriculture (Hofmann et al 2016); soil erosion due to land use change, mining and increasing livestock densities (Sasaki et al 2018;Sternberg 2012;Wei et al 2019); and soil pollution (mostly with heavy metals) due to mining, industry and coal combustion in urban areas (Jarsjö et al 2017;Kosheleva et al 2019).…”