Spatial–Temporal Evolution of Land Desertification Sensitivity in Mu Us Desert Ecological Function Reserve ^†

Abstract

Land desertification management in the Mu Us Desert has received widespread attention. Assessing land desertification sensitivity is crucial for desertification monitoring and management. This study constructed a comprehensive evaluation index system using four factors: dryness index, the number of windy and sandy days in the winter and spring, soil texture, and vegetation cover. Land sand sensitivity was divided into five grades, and multi-source data from the Ecological Functional Reserve of the Mu Us Desert from 2002 to 2022 were used to study spatial distribution and dynamic changes. The results show the following: (1) the overall land desertification sensitivity in the Mu Us Desert Ecological Functional Reserve decreased from 2002 to 2022, with the proportion of highly sensitive land decreasing from 92.39% to 82.75%, and the proportion of medium-, medium–low-, and low-sensitivity areas increasing from 0.63% to 1.70%. (2) Low-sensitivity areas were concentrated in Jingbian County, Hengshan District, and southern Uxin Banner. Southeast Otog Banner and northern Jingbian County saw the most significant decreases in land desertification sensitivity since 2002. (3) The four selected factors interacted, with increased vegetation cover being the most crucial factor. This study provides a reference for future ecological restoration in the Mu Us Desert area.

1. Introduction

The Mu Us Desert, located at the junction of Inner Mongolia, Shaanxi and Ningxia, is one of the important sandy areas in China. The region has a dry climate and fragile ecology. Influenced by national policies and human activities, over the past 20 years it has transformed from a fragile ecology to an effective sand control model, achieving significant results in soil and water conservation and in wind and sand fixation projects. Remote sensing provides large-scale, multi-temporal, long-term vegetation coverage data, enabling analysis of the spatial and temporal dynamics of windbreak and sand fixation, aiding desertification management. Common monitoring methods include a differential threshold, supervised classification, thematic index inversion, and a classification decision tree.

Land desertification sensitivity refers to the likelihood of desertification due to human activities and natural changes. The land desertification sensitivity index measures regional ecological quality. Evaluating windbreak and sand fixation and regional ecological construction using these data is crucial. An ecological function protection zone is vital for water conservation, soil conservation, flood regulation, windbreak and sand fixation, and biodiversity maintenance and is designated for key protection and restricted development [1]. Constructing such zones is significant for harmonious development between humans and nature and for managing primary function areas that limit development.

Current research on the Mu Us Desert’s ecological function protection area primarily addresses land use changes [2], desertification causes [3], sand fixation and protection measures [4], landscape pattern changes [5], impacts on the surrounding desertified land distribution [6], and driving force analysis [7]. These studies generally cover short time spans. Additionally, there are few international studies on the spatial and temporal evolution characteristics of land desertification sensitivity in this region. Most international research focuses on desertification causes [8], vegetation cover changes [9], and the effects of climate change and human activities on desertification reversal [10]. There are also studies using trend analysis and mutation point detection methods to examine the spatial and temporal evolution characteristics of vegetation coverage in the Mu Us Desert [11].

This paper analyzed the change in land desertification sensitivity grade in the Mu Us Desert Ecological Function Reserve from 2002 to 2022 using multi-source data. Four factors were selected: aridity index, the number of windy sand days with wind speeds greater than 6 m/s in the winter and spring, soil texture, and vegetation coverage. Sensitivity grade was divided into five levels for evaluation, and the spatial and temporal patterns of windbreak and sand fixation functions were analyzed. Given that many factors affect land desertification and studies have shown that increasing vegetation cover is the most effective way to resist land desertification [12], this paper focuses on vegetation cover. This provides a scientific basis for the ecological protection and restoration of the Mu Us Desert.

2. Overview of the Study Area

Figure 1 shows that the Mu Us Desert Ecological Nature Reserve is located in northwestern China, encompassing the central Loess Plateau, northern Yulin City, southern Ordos City, and northeastern Yanchi County (37.45–39.38° N, 107.67–110.5° E). It lies in an arid and semi-arid transition zone, as well as in a transition zone between desert, grassland, steppe, and forest, making it a typical combined agricultural and pastoral area. The northwest region primarily consists of sand and grassland, while the southeast mainly consists of cultivated land and grassland [13]. The study area has a temperate continental climate, with low and uneven precipitation ranging from 250 to 440 mm annually, increasing from the west to the southeast. This ecologically fragile region in northwestern China is severely affected by land desertification and serves as an important area for windbreak and sand fixation [14,15].

3. Research Data and Methods

3.1. Data Source

The remote sensing data in this paper consist of different datasets of the study area from 2002,2008, 2015, and 2022, from the sources shown in

Table 1. Data types and data sources.

Data Type	Dataset Name	Data Source
Surface evaporation data	Terra Net Evapotranspiration 8-Day Global 500 m	MOD16A2.006
Rainfall data	Climate Hazards Group InfraRed Precipitation With Station Data (Version 2.0 Final)	CHIRPS Daily
Soil texture data	OpenLandMap Soil Texture Class	USDA System
Sand days data	Global Land Data Assimilation System	GLDAS-2.1
Vegetation coverage data	Terra Vegetation Indices 16-Day Global 500 m	MOD13Q1.006

.

3.2. Land Desertification Sensitivity Evaluation Factor and Calculation Method

The data show that under different vegetation types and topographic conditions, soil is affected by climatic factors such as wind and precipitation, resulting in significant differences in the degree of desertification in different regions [16].

The desiccation index measures the interaction between heat and moisture and reflects precipitation’s role in soil wind erosion. A higher desiccation index indicates a drier climate and greater potential for wind erosion. High winds are the primary cause and external driving force of soil wind erosion, with wind intensity being crucial for soil particle transport [17]. Since the study area mostly has sandy soil, days with wind speeds over 6 m/s were selected. More high-wind days and longer durations increase wind erosion effects on surface soils [18]. Soil texture affects land susceptibility to desertification, with different grain sizes offering varying resistance to wind uplift. Vegetation cover is another crucial factor, negatively correlated with soil wind erosion under typical conditions [19].

Combined with the above factors, according to the ‘Technical Guide for Resource and Environmental Carrying Capacity and Land Space Development Suitability Evaluation’ and further research results, it can be seen that there are many factors affecting the sensitivity of land to desertification. This paper selects the dryness index (I), the number of wind-blown sand days (W), soil texture (K), and vegetation coverage (C) as factors, for days with wind speeds greater than 6 m/s in the winter and spring, to evaluate the sensitivity of land to desertification in the target area [20].

(1)

Dryness index (I)

The aridity index (I) is the ratio of potential evapotranspiration to precipitation. The formula is as follows:

$$I={\displaystyle \frac{E{T}_0}{P}}$$

(1)

In this formula, I is the dryness index, ET₀ is the potential evapotranspiration in mm, and ET₀ is calculated using the Penman–Monteith model revised by the Food and Agriculture Organization of the United Nations. P is the annual precipitation in mm.

(2)

Speed data (W)

Using wind speed data from the meteorological stations in the study area, the maximum daily wind speed was taken as the value for each day, and the number of days with wind speeds greater than 6 m/s was counted year by year.

(3)

Soil texture data (K)

Soil texture was measured at depths of 0, 10, 30, 60, 100, and 200 cm. This study used the 0 cm soil data, assigning values based on soil texture grading, as shown in

Table 2. Land desertification sensitivity evaluation index and assigned grades.

Evaluation Factor	High Grade	Medium–High Grade	Medium Grade	Medium–Low Grade	Low Grade
Dryness index	≥16.0	4.0–16.0	1.5–4.0	1.0–1.5	<1.0
Days with wind-blown sand (days)	≥30	20–30	10–20	5–10	<5
Soil texture	Sandy	Loamy	Gravelly	Clayey	Rocky
Vegetation coverage (%)	<0.2	0.2–0.4	0.4–0.6	0.6–0.8	≥0.8
Hierarchical assignment	9	7	5	3	1

. Since soil textures at the same site are unaffected by seasonal changes, K values for the assessment were directly derived from the annual datasets.

(4)

Vegetation cover (C)

The normalized vegetation index, NDVI, is one of the important parameters reflecting crop growth and nutritional information, and can better reflect vegetation cover and its inter-annual spatial distribution differences in the study area.

$$NDVI={\displaystyle \frac{(NIR-R)}{(NIR+R)}}$$

(2)

Vegetation cover based on the NDVI was derived using the image element binary model.

$$C={\displaystyle \frac{NDVI-NDV{I}_{soil}}{NDV{I}_{\mathit{veg}}-NDV{I}_{\mathit{soil}}}}$$

(3)

where $NDV{I}_{soil}$ is the NDVI value of the area covered entirely by bare soil; $NDV{I}_{veg}$ is the NDVI value of the area covered entirely by vegetation; and C is the degree of vegetation cover.

Ultimately, the following formula was used for the assessment of the land desertification sensitivity classes:

$$[\mathit{\text{Land desertification sensitivity}}]=\sqrt[4]{I\times W\times K\times C}$$

(4)

The assignment of these four influencing factors into grades is based on annual data, allowing this formula to represent the land desertification sensitivity grade for the study area each year. The final desertification sensitivity evaluation factors were categorized into five levels—very high (>7.0), high (6.1–7.0), medium (5.1–6.0), low (3.1–5.0), and very low (1.0–3.0)—as shown in the following Table 2.

4. Analysis of Results

4.1. Temporal Variation Trend of Land Desertification Sensitivity in the Study Area

Figure 2 shows the trend of land desertification sensitivity area proportion in the Mu Us Desert Ecological Functional Reserve during the period of 2002–2022. Since 2002, the proportion of highly sensitive areas in the study area has decreased from 92.39% to 82.75%, showing a fluctuating downward trend, with an average annual decrease. In contrast, the proportion of more sensitive areas has increased from 6.98% to 15.55%, showing a fluctuating upward trend. The proportions of medium-sensitivity, less sensitive, and low-sensitivity areas have changed relatively stably, slowly increasing from 0.63% to 1.70%. This reflects the fact that the overall land desertification sensitivity level of the Mu Us Desert Ecological Functional Reserve has shown a decreasing trend, with the high sensitivity level of the study area transitioning to higher, medium, lower, and low sensitivity levels. The process of land desertification has been curbed, and the ecological function of vegetation in preventing wind and fixing sand has been enhanced.

4.2. Changes in the Spatial Patterns of Land Sensitivity to Desertification in the Study Area

As shown in Figure 3 from the spatial distribution pattern of land desertification sensitivity in 2002, 2008, 2015, and 2022 shows that high-sensitivity areas occupy the largest portion of the study area. These are mainly in the western, northern, and central parts of the Mu Us Desert Ecological Functional Reserve, where vegetation coverage is low and the climate is dry. Medium-sensitivity areas are mostly in the eastern and southern parts, as well as in the transition zones. Low-sensitivity areas, which are less prone to desertification and dust, are very small and mainly found in the southern part of the study area where vegetation cover is high and human activities, such as sand control, are more intensive.

According to area statistics, the 2002 evaluation showed that low-sensitivity areas accounted for 0.16% of the total area, lower-sensitivity areas for 0.33%, medium-sensitivity areas for 0.14%, higher-sensitivity areas for 6.80%, and high-sensitivity areas for 92.39%. In 2008, the proportion of low-sensitivity areas was 0.03%, that of lower-sensitivity areas 0.47%, that of medium-sensitivity areas 0.77%, that of higher-sensitivity areas 20.74%, and that of high-sensitivity areas 77.99%. In 2015, the proportion of low-sensitivity areas was 0.04%, that of lower-sensitivity areas 0.45%, that of medium-sensitivity areas 0.23%, that of higher-sensitivity areas 5.69%, and that of high-sensitivity areas 93.59%. In 2022, the proportion of low-sensitivity areas was 0.02%, that of lower-sensitivity areas 0.50%, that of medium-sensitivity areas 1.17%, that of higher-sensitivity areas 15.55%, and that of high-sensitivity areas 82.75%.

Based on a comparison of the results from the four periods, from 2008 to 2022, sand sensitivity increased the most in Otog Front Banner, the junction of Otog Front Banner and Uxin Banner, and the central part of Shenmu City. Conversely, sand sensitivity decreased the most in the southeastern part of Otog Front Banner and the northern part of Jingbian County (at the junction with Uxin Banner and Otog Front Banner).

According to provincial statistics, compared with 2002, the proportion of area with a high land desertification sensitivity grade in the Inner Mongolia and Shaanxi regions to the total land area of the region in 2022 decreased by 4.53% and 18.95%, respectively. The land desertification sensitivity grades in the southern part of Otog Front Banner within the Inner Mongolia Autonomous Region and in the northern part of Jingbian County within Shaanxi Province were significantly reduced. From 2000 to 2022, the area of land in the high-sensitivity class of land desertification in the Kerqin sandy land area in southeastern Inner Mongolia significantly decreased, gradually transitioning to the high-sensitivity and medium–high-sensitivity classes. Most areas in the northwestern part of Shaanxi Province transitioned from the medium–high-sensitivity class in 2002 to the medium-sensitivity class.

4.3. Relationship Between Influencing Factors and Land Sensitivity to Desertification

Figure 4 illustrates that from 2002 to 2022 the temperature and precipitation in the Mu Us Desert Ecological Function Reserve increased, with the average annual temperature rising by 0.29 °C per decade and precipitation increasing by about 1.04 mm per year.

This rise in heat and precipitation favors vegetation recovery and growth. Experimental results show that, except for slight decreases in vegetation cover in the northern, central, and southeastern parts of Otog Front Banner, the rest of the area has experienced increasing vegetation cover. This trend is particularly notable in the western part of Shenmu City, in Yuyang District, in northwestern Hengshan District, and at the junction of Jingbian County and Uxin Banner, where vegetation cover has significantly increased over the past 21 years. Good vegetation cover effectively protects surface soil, reduces wind-carried sand and dust, and inhibits land desertification.

Improved surface vegetation also stabilizes the soil, reducing its contribution to sand and dust during windy weather, thus curbing land desertification. Additionally, the average annual number of windy days in the Mu Us Desert Ecological Functional Reserve has shown a decreasing trend, with fluctuations since 2006. Fewer windy days have weakened wind erosion on the surface soil, contributing to reduced land desertification susceptibility.

Figure 5 illustrates the changes in vegetation cover across different areas of the Mu Us Desert Ecological Functional Reserve in 2008, 2015, and 2022. In 2008, the high-sensitivity vegetation cover accounted for 8.48% of the total area, decreasing by 1.54% to 6.94% by 2022. The medium-high-sensitivity class covered 45.86% in 2008 but dropped significantly by 20.97% to 24.89% in 2022. In contrast, the medium-sensitivity class increased slightly from 32.37% in 2008 to 33.49% in 2022. The low-sensitivity class showed a notable rise, growing from 10.06% in 2008 to 17.26% in 2022. Similarly, the lowest-sensitivity class expanded from 2.76% in 2008 to 5.03% in 2022.

5. Results and Discussion

5.1. Results

The main conclusions of this paper are as follows:

(1) Over the period from 2002 to 2022, the area of highly sensitive land decreased from 92.39% to 82.75%, while the area of land with medium-, medium–low-, and low-sensitivity grades increased from 0.63% to 1.70%. The evaluation grade of land desertification sensitivity in the Mu Us Desert Ecological Functional Reserve as a whole shows a decreasing trend, gradually transitioning from high-sensitivity to medium–high-, medium-, and medium–low-sensitivity grades.

(2) Spatial observations show that, compared to 2002, the land desertification sensitivity of the Mu Us Desert Ecological Functional Reserve remained stable in 2022. Areas of a low-sensitivity grade were concentrated in Jingbian County, Hengshan District, and the southern part of Uxin Banner. High- and medium–high-sensitivity levels in the southeastern part of Otog Front Banner and the northern part of Jingbian County (at the border of Uxin Banner and Otog Front Banner) were significantly reduced, and the overall land desertification sensitivity improved. Increased vegetation cover and fewer windy days are key factors contributing to these changes.

5.2. Discussion

(1) Most of the Mu Us Desert is in a semi-arid zone. The climate’s dryness, drastic temperature changes, frequent strong winds, and intense sunlight are key factors affecting vegetation recovery. Since 2002, ecological projects have increased vegetation cover, reduced strong winds’ impact, and curbed land desertification. This is especially evident in southern Inner Mongolia and the border areas of Shaanxi and Ningxia, where the proportion of highly sensitive land has significantly decreased, consistent with this study’s findings.

(2) Although ecological projects in the region have led to increased precipitation and temperature and fewer windy days, reducing land desertification sensitivity to some extent, the Mu Us Desert Ecological Functional Reserve remains one of the most severely desertified areas in the country. Studies have shown that increasing vegetation cover is the most effective way to combat land desertification. The Mu Us Desert should actively respond to policies by adhering to the principles of “protection priority, scientific management, and moderate use”, increasing the coverage of drought-tolerant shrub vegetation, and focusing on restoring native vegetation [21]. Additionally, locally suitable measures should be sought to consolidate an ecological barrier centered on drought-tolerant vegetation and improve the ecological carrying capacity of the Mu Us Desert Ecological Functional Reserve [22].

(3) Many factors influence land desertification, and they constrain each other. The selection of the aridity index, the number of windy days in the winter and spring, soil texture, and vegetation cover as the four factors for constructing an evaluation model in this study is effective, but other methods with different focuses also exist. The choice of influencing factors in the study of land desertification sensitivity affects the evaluation results.