-
Rocky desertification is an extreme form of land degradation, which leads to water shortages, reduced and barren soils, and the direct consequence of loss of land resources. Rocky desertification, together with desertification and water/soil loss, are three known types of land ecological disasters (Phillips, 2016; Veress 2020). Karst rocky desertification mainly occurs in the European Mediterranean basin (Lavee, 1998), Dinaric Karst (Gams and Gabrovec, 1999) and southern and western (Jiang et al., 2014). It is also distributed in other countries and regions in the world, such as Belize, Guatemala, Mexico, Israel, Ryukyu Islands (Ford and Williams, 2007), Indonesia (Sunkar, 2008), Caribbean island countries (Jiang et al., 2014) and Haiti (Williams, 2011). Karst landforms in China correspond to one-third of the total national land area and are contiguously exposed across 5.4 × 105 km2 in 465 counties (cities and districts) in Guizhou, Yunnan, Guangxi, Hunan, Hubei, and Sichuan and Chongqing (Wu et al., 1998). Compared with other typical ecologically vulnerable regions such as the Loess Plateau and Tibetan Plateau, the karst region is constrained by its special geological background of easily dissolved carbonates. As a result, within the region, pedogenesis occurs extremely slowly, and the formed layer of soil is generally thin and discontinuous, resulting in rapid impacts of the soil layer to hydrological processes and severely damaged ground vegetation due to human activities. Severe erosion of soils results in the exposure of large areas of the bedrock, eventually driving the region to become mainly rocky and desertified and an ecologically vulnerable region in China (Qin et al., 2006).
The problem of rocky desertification severely limits the development of the local economy. Thus, it is important to study its spatiotemporal characteristics and its evolutionary pattern (Luo et al., 2021), which can assist with implementing targeted controls from a scientific perspective (Ma et al., 2015). Thus far, there have been a series of achievements in studying the spatiotemporal variations in karst rocky desertification in southern and western. By deploying a drone-based remote sensing technology, Wen and Li (2020) investigated both spatiotemporal distribution patterns and evolutionary patterns in the Guizhou rocky desertification region from 2004 to 2016. Based on the comprehensive index method, they analysed the rocky desertification distribution patterns in different ecological protection areas with karst landforms in southern and western, utilizing net primary productivity (NPP), the normalized difference vegetation index (NDVI), and slope indices. By utilizing MODIS (moderate resolution imaging spectroradiometer)-NDVI data from 2003 to 2016, Tian et al. (2017) analysed vegetation variations in rocky desertification areas of different grades in Guizhou in the most recent 14 yr based on the approach of linear trend analysis and interpolation. In addition, by adopting the comprehensive index method, Shi and Shu (2017) created a distribution map for rocky desertification areas with extreme, strong, medium, slight, potential, and no rocky desertification grades. Finally, by utilizing the NDVI and meteorological data, Wang et al. (2021) focused on discussing the NDVI spatiotemporal variation characteristics in rocky desertification areas in Chongqing based on trend analysis, coefficient of variation analysis, and partial correlation analysis.
The mechanism driving the evolution of rocky desertification is very important for rocky desertification control. It is commonly considered that the formation and development of rocky desertification in the karst region in southtern and western China are induced by both natural processes and human activities. Zhang (ang (2018) studied the relationship between vegetation coverage and rocky desertification evolution. Shi and Shu (2017) also investigated the spatiotemporal variation characteristics of rocky desertification in Guizhou and the associated key driving factors. Xu et al. (2019) analysed the contributions of human activities and climate change to the recovery from rocky desertification, which diversified the regional scale of related research. A number of previous studies have come to the conclusion that rocky desertification is the combined result of topography, formation lithology, geological structure, hydrology, meteorology, soil, and vegetation distribution (Wang et al. 2003; Cao et al. 2004; Li et al. 2005; Lv et al. 2007).
Geodynamics sculptured the steep and broken karst landscape, providing a dynamic potential energy for the formation of rocky desertification (Weng, 1995). Zhou and Huang (2003) found that limestone regions possess the largest area of rocky desertification with the highest rocky desertification degree, while marlstone regions rank the lowest in their rocky desertification area. Li et al. (2003) observed that rocky desertification is significantly correlated with lithology. Some scholars have proposed that regions with steeper slopes are more likely to experience water and soil loss, which intensifies rocky desertification (Ji, 2013). When the temperature is higher than a certain threshold value, the karstification in karst regions intensifies with increasing temperature. In addition, precipitation mainly occurs at relatively high temperatures from April to September. Thus, the dual effect of temperature and precipitation induces more severe soil loss in the karst region, where the original soil layer is already thin, thus accelerating the process of rocky desertification (Yuan, 1994; Su, 2002; Su et al., 2006). Zhang et al. (2012) concluded that regional variation in rocky desertification is mainly controlled by the type of material in the underlying surface (i.e., the number of ‘rocky mountains’). The early-to-middle period of the Qing dynasty is considered a critical transitional period for the effect of human activities on rocky desertification. During this period, natural factors started to play a less significant role in affecting rocky desertification compared with human activities, such that human economic activities became dominant in controlling rocky desertification ( Han et al., 2006; Li et al., 2007). The ecological system in the karst region of southern and western is extremely vulnerable and sensitive to disturbances caused by human activities (Su, 2002). Since the founding of the People’s Republic of China, the karst mountainous region has been trapped in a vicious circle of ‘rapid population rise-excessive land reclamation-erosive soil degradation-rocky desertification spread’ (Su and Zhou, 1995; Deng et al., 2009). Zhou and Huang (2003), Zhang et al. (2008) also reported a positive correlation between the occurrence of rocky desertification and population density in karst regions. Finally, deforestation exposes large areas of land, which also accelerates the process of rocky desertification (Huang et al., 2006).
The main natural factors that influence the evolution of rocky desertification include topography, formation lithology, hydrology, meteorology, and soil and vegetation distribution. In addition, human factors mainly include population density, land-use type, cultivation, grazing, and mining (Zhang et al., 2015; Tong et al., 2017). The dominant factor in controlling rocky desertification has gradually changed to human factors with socioeconomic development (Yan, 2018). However, previous studies on the characteristics of changes in karst rocky desertification in southern and western and driving mechanisms mainly focus on certain province, city or county, and there is no investigation on the whole area of southern and western. The time frame of previous studies is either narrow or long from now (Peng et al., 2013). In this study, we investigate the pattern of distribution of rocky desertification areas in the whole karst region of southern and western from 2000 to 2015 and determine the key factors, based on which we present the spatiotemporal variation characteristics of rocky desertification in the southern and western and the associated driving mechanisms.
-
The karst region in southern and western China includes 465 counties (cities and districts) in Guizhou Province, Yunnan Province, Guangxi Zhuang Autonomous Region, Hunan Province, Hubei Province, Sichuan Province, Guangdong Province and Chongqing City, as shown in Fig. 1. The karst region starts from the southern foot of the Qinling Mountains and reaches the Guangxi Basin in the south, the Hengduan Mountains in the west, and the west side of the Luoxiao Mountains in the east, with a total area of approximately 5.4 × 105 km2 (Wu et al., 1998). The southtern and western karst region possesses abundant water resources, including the Yangtze, Pearl, and Lancang Rivers. Within this region, topography exhibits large vertical variations and diversity with rolling mountains and widely distributed carbonates. Topography is high in the northwest and low in the southeast with a large vertical gradient: the highest altitude above sea level is 7143 m, and the lowest altitude is −142 m. The karst region spans the south temperate zone, north subtropical zone, central subtropical zone, south subtropical zone, and plateau climatic zone, with the central subtropical monsoon and humid climate being prominent within the region. The population of this region contains a cluster of minority nationalities in China, including the Buyi, Miao, Gelao, and Yi. From National Bureau of Statistics of China (
https://data.stats.gov.cn/ ), by the end of 2015, the gross domestic product of the eight southtern and western regions had reached 21.8 × 1013 yuan (RMB), 75.03% higher compared with 2010. Specifically, the gross domestic product of the primary industry was 0.774 ×1013 yuan, an increase of 58.04%, the gross domestic product of the secondary industry was 3.58 × 1013 yuan, an increase of 59.05%, and the gross domestic product of the tertiary industry was 4.99 × 1013 yuan, an increase of 98.73%. Among the three industries, the tertiary industry has shown the fastest growth.
Figure 1. The karst region in southtern and western China
-
According to previous research results (Lou, 2016), slope, vegetation coverage, and lithology are the three important factors that affect rocky desertification. This article evaluates the comprehensive characteristics of the three factors and divides rocky desertification into five levels, namely none rocky desertification, slight, medium, strong, and extreme intensity. The degree of rocky desertification in the three years of 2000, 2010 and 2015 will be evaluated separately (Table 1).
Table 1. Classification of rocky desertification degree
Rocky desertification degree Numerical value Slope / (°) Vegetation coverage / % Lithology None 0 < 5 > 80 Buried carbonatite Slight 1 5−15 60−80 Sub-impure carbonate Medium 2 15−25 40− 60 Impure carbonate Strong 3 25−35 20−40 Dolomite Extreme 4 > 35 < 20 Limestone and dolomite The main data of rocky desertification grade assessment has the following sources. The first is the slope data, which is based on the digital elevation model (DEM) data with a resolution of 90 m downloaded by National Aeronautics and Space Administration of USA (NASA) (
https://www.nasa.gov/ ), and the slope map is generated through the calculation of ArcGIS. The second is the vegetation coverage, which comes from the Institute of Aerospace Information Innovation of the Chinese Academy of Sciences and is retrieved from Moderate-resolution Imaging Spectrora diometer (MODIS), including three years of 2000, 2010 and 2015. The third is lithology data, which comes from the national geological map (https://daac.ornl.gov/cgi-bin/dsviewer.pl?ds_id=1566 ). -
The land-use transfer matrix can be adopted to represent a dynamic process during which the areas with different land-use types are interchangeable at the beginning and end of a certain time period in a certain region. The matrix includes areas of different land-use types at a certain time point and the input and output areas of different land-use types at the beginning and end of a time period. Using ArcGIS, we utilized the land-use transfer matrix to calculate the transfer of rocky desertification areas in 2000, 2010, and 2015. In addition, we conducted a statistical analysis of areas with varied rocky desertification grades on both regional and provincial scales. The transfer matrix is shown as follows:
$${s_{ij}} = \left[ {\begin{array}{*{20}{c}} {{{\rm{s}}_{00}}}&{{{\rm{s}}_{01}}}&{...}&{{{\rm{s}}_{0n}}}\\ {{{\rm{s}}_{10}}}&{{{\rm{s}}_{11}}}&{...}&{{{\rm{s}}_{1n}}}\\ {...}&{...}&{...}&{...}\\ {{{\rm{s}}_{n0}}}&{{{\rm{s}}_{n1}}}&{...}&{{{\rm{s}}_{nn}}} \end{array}} \right]$$ (1) where s is the area, n is the rocky desertification grade before and after the transfer, i and j (i, j = 0, 1, …, n) represent the rocky desertification types before and after the transfer, and Sij is the area of i rocky desertification type that transferred to j rocky desertification type. In the matrix, elements in each row represent the flow information about the transfer of i type rocky desertification, and elements in each column represent the sources of the transferred j type; i = j corresponds to an unchanged i type rocky desertification area.
-
We adopted methods based on value assignment and difference to assess the spatial variation in rocky desertification between two periods. We divided rocky desertification into five degrees (Table 1), namely, none, slight, medium, strong and extreme, with numerical values of 0−4 respectively assigned to these degrees. After the assignment of numerical values to rocky desertification grids in ArcGIS, we conducted difference calculations to characterize changes in rocky desertification and grade classification based on the before-and-after rocky desertification changes. An improvement corresponds to the transfer from a high rocky desertification degree to a low degree, and vice versa. The absolute value of the before-and-after difference of 1−2 denotes a slight change (slight degradation and alleviation), 3−4 denotes a strong change (Strong degradation and alleviation), and 0 denotes no change in the rocky desertification degree (Table 2).
Table 2. Rocky desertification change grade division
Value difference [−4, −3] [−2, −1] 0 [1, 2] [3, 4] Difference-based rating Strong degradation Slight degradation No change Slight alleviation Strong alleviation -
In this study, we adopted redundancy analysis (RDA) to investigate the rocky desertification process between 2000 and 2015. The RDA is a method that extracts and sums changes in a set of response variables, which can provide an explanation
on the basis of a group of explanatory variables. First, we conducted detrended correspondence analysis (DCA) to compare the first axes of areas of different rocky desertification grades. If the axis length of the first axis was greater than 4.0, we adopted a nonlinear single-peak ordination method to conduct canonical correspondence analysis (CCA); if the value was smaller than 3.0, we then adopted a linear ordination method (e.g., principal components analysis (PCA) or RDA); and if the value was between 3.0 and 4.0, either RDA or CCA was adopted. When there were missing or abnormal data in the environmental gradients for any of the influencing factors, we adopted the DCA method to eliminate the data. Based on the DCA ordination method, the data arc effect was removed after the second axis. RDA is a constraint ordination method based on the main component analysis, whose goal is to find a new variable as the best predicting indicator to predict the distribution of response variables. We assume that the new variable X (assuming it is the first axis) has a corresponding value in each sample and that the new variable has a value of Xi in the ith sample; we can predict the k type in the ith sample using the following equation:
$${Y_{ik}} = {b_{0k}} + {b_{1k}}{X_i} + {e_{ik}}$$ (2) where, Xi is the coordinate of the sample in the first axis, b1k is the regression coefficient of each rocky desertification type and denotes the coordinate of the rocky desertification type in the first axis, and the other parameter b0k represents the intercept of the regression line. eik is the random effects of species k in sample i.
The coordinate value Xi of the RDA sample can be obtained based on constraints and is a linear combination of environmental factors. Assuming that there are two measured environmental variables Zi1 and Zi2, the value of the new variable Xi can be expressed using the linear combination of the environmental variables Zi1 and Zi2 using the following equation:
$${X_i} = {c_1}{Z_{i1}} + {c_2}{Z_{i2}}$$ (3) where, c1 and c2 are the correlation coefficients that represent the significance of the correlations between the environmental factors and the corresponding ordination axes.
The two steps above convert the RDA result to a multivariate, multielement regression equation set:
$$ {Y_{ik}} = {b_{0k}} + {b_{1k}}{c_1}{Z_{i1}} + {b_{2k}}{c_2}{Z_{i2}} + {e_{ik}}$$ (4) where, bikcj denotes the regression coefficient in the multivariate, multielement regression model, which describes the abundance of k-grade rocky desertification based on the degree of j environmental factor.
The influencing factor value can randomly assign variations to the rocky desertification grade without causing other effects. Based on data permutation, we obtained statistical test analytical results. The significance level test equation is shown below:
$$ p = \frac{{{n_x} + 1}}{{N + 1}} $$ (5) where, p represents the significance test level, nx represents the number of permutations that never lies below the number of random permutation analyses, and N represents the total number of permutations.
Partial ordination analysis assesses the contribution of each influencing factor variable to the change in rocky desertification by performing a partial Monte Carlo permutation test. Each influencing factor candidate is considered the only variable for analysis and the chosen factor(s) as covariate(s). Finally, based on the ordination model, a Monte Carlo permutation test is conducted to determine the main influencing factor.
The explanation of each influencing factor to response variables can be decomposed to a conditional explanatory variable and a marginal explanatory variable. The difference between the two explanatory variables can be used to determine the interaction between multiple influencing factors. The variance decomposition method can decompose a variable explanation to an unaffected factor explanatory part, an independent explanatory part, and a dependent explanatory part, whose values can be calculated based on the partial constraint analysis method.
-
To accurately analyse the rocky desertification pattern in southern and western region, we collected data relevant to economic activities from the statistical yearbooks (https://data.stats.gov.cn/) of the eight regions in southern and western China (gross domestic product (GDP), industrial product, population data, and agricultural population data). Economic activity, population, and human health data used in this study for exploring the mechanisms driving rocky desertification (ion (Table 3) were obtained on the basis of changes between the two periods. We present the data changes in percentage as our dependent variables.
Table 3. Selected socioeconomic driving factors of rocky desertification changes
Social and economic factors Abbreviation Unit Gross domestic product GDP 104 Yuan Gross domestic product of primary industry GDP1 104 Yuan Gross domestic product of secondary industry GDP2 104 Yuan Gross domestic product of tertiary industry GDP3 104 Yuan Fixed-asset investment FixInv 104 Yuan Year-end population PopYe 104 -
Previous study on the mechanisms driving rocky desertification has pointed to a significant role of topographic condition (Qin et al., 2006). Thus, in ArcGIS, we chose each county as our research unit and extracted their mean altitude, terrain slope, terrain orientation, and mean topographic fractal dimension (Table 4). For analysing the causes of rocky desertification, we adopted changes (in percentage) in the county’s topographic condition between two years as dependent variables. The DEM data in 30 m for provinces in China are from the Resources and Environmental Science Data Centre.
Table 4. Selected terrain driving factors of rocky desertification changes
No. Terrain factors Abbreviation Unit 1 Elevation ELEV m 2 Slope SLO ° 3 Topographic fractal dimension index TPI To calculate changes in percentage in the county’s topographic condition between two years, we adopted the rocky desertification areas of each county in 2000 and 2015 to obtain the means of the topographic factors in these rocky desertification areas, based on which we calculated the changes in percentage for each county. Among topographic factors, the topographic fractal dimension index is a critical factor that affects the spatial difference in the surface and is a comprehensive reflection of topography and elevation. During our analyses of topographic differences, we introduced the topographic fractal dimension index when a single elevation or terrain slope could not explain the phenomenon. In this study, we utilized the geographic information modelling method that combines elevation and terrain slope to extract the topographic fractal dimension index, which was able to better explain rocky terrain changes in the southtern and western region. The equation describing this yields:
$$ T{\rm{ = }}\log \left[ {\left( {\frac{E}{{\overline E }} + 1} \right) \times \left( {\frac{S}{{\overline S }} + 1} \right)} \right] $$ (6) where T is the topographic fractal dimension index, E and S represent the elevation and slope of any point in space, respectively, and
$ \overline E$ and$ \overline S$ represent the mean elevation and mean slope of a certain region, respectively. The higher the elevation is, the steeper the slope and the larger the topographic fractal dimension index is, and vice versa. For cases with a high elevation but a shallow slope or with a low elevation but a steep slope, the topographic fractal dimension index has medium values. -
Different ecological systems exhibit differences in soil erosion intensity, thus leading to differences in the distribution of ecological systems of different rocky desertification degrees. In addition, ecological system type plays a certain role in affecting the evolution of rocky desertification. For many years, China has conducted multiple ecological protection projects targeted ecological environmental problems, including the Returning Farmland to Forest project, the Natural Forest Protection project, and the Karst Rocky Desertification Control project, which are closely related to the control of rocky desertification in southern and western China. Thus, to investigate changes in rocky desertification, we chose county as our research unit and calculated changes in percentage in areas between 2000 and 2015 for different ecological systems as the influencing factors. There is no rocky desertification process in wetland, desert and bareland, as a result, wetland, desert and bareland were not considered in driving mechanisms analysis. The Resources and Environment Science and Data Center in China is a world-class research platform for land surface system science. The data used in this study are the remote sensing data for land use in China in 2000, 2010, and 2015 from The Resources and Environment Science and Data Center in China (
https://www.resdc.cn/ ), which were classified based on the ecological system division scheme. The selected ecological system factors are summarized in Table 5.Table 5. Selected ecosystem driving factors of rocky desertification changes
No. Ecosystem factors Abbreviation Unit 1 Forest FOE km2 2 Shrubland SHE km2 3 Grassland GRE km2 4 Farmland FAE km2 5 City/town TOE km2 -
From 2000 to 2015, ecosystem area in the rocky desertification region in southern and western was significantly enhanced in towns, which increased by 48.42% and reached 1.810 × 104 km2. On the other hand, the farmland area significantly decreased by approximately2.220 × 104 km2, corresponding to a 4.13% decrease (Fig. 2). In addition, areas of forest, grassland, wetland, and desert were enlarged by 0.210 × 104 km2, 0.010 × 104 km2, 0.190 × 104 km2, and 0.100 × 104 km2, respectively, corresponding to increases of 0.26%, 0.05%, 3.80%, and 6.13%, respectively, in 2000. In addition to the reduced farmland area, shrubland and bare land also decreased by 0.30% and 0.11%, respectively.
Figure 2. Percentage changes in areas for different ecological systems from 2000 to 2015 in southtern and western China
Except for bare lands, a total area of 5.570 × 104 km2 of ecosystems experienced transformation in the southwestern region from 2000 to 2015. Specifically, the area of farmland that transformed to other ecosystems was the highest, that is followed by forest and shrubland. In addition, the transferred farmland area was 2.916 × 104 km2, with contributions from town, forest, and shrubland areas of 1.401 × 104 km2, 0.704 × 104 km2, and 0.343 × 104 km2, respectively. The transformed forest area was 1.100 × 104 km2, and the transformations of for-est to farmland, shrubland and town were 0.331 × 104 km2, 0.292 × 104 km2, and 0.256× 104 km2, respectively. The transferred shrubland area was 0.785 × 104 km2, and the transformations of shrubland to forest, farmland, and city/town were 0.400 × 104 km2, 0.162 × 104 km2, and 0.126 × 104 km2, respectively. Ecosystems with relatively high input transformation areas were town, forest, and shrubland. The transformed town area was 1.973 × 104 km2, and the transformed forest and shrubland areas were also very large: 1.307 × 104 km2 and 0.703 × 104 km2, respectively. (Table 6).
Table 6. Ecosystem transfer matrix of southtern and western China from 2000 to 2015 / 104 km2
Year Ecosystem 2015 Forest Shrubland Grassland Wetland Farmland City/town Desert Out 2000 Forest 79.883 0.292 0.065 0.038 0.331 0.256 0.118 1.100 Shrubland 0.400 26.917 0.030 0.047 0.162 0.126 0.020 0.785 Grassland 0.125 0.044 20.591 0.024 0.038 0.065 0.005 0.301 Wetland 0.012 0.007 0.013 4.876 0.070 0.112 0.006 0.220 Farmland 0.704 0.343 0.200 0.230 50.673 1.401 0.036 2.916 City/town 0.020 0.012 0.001 0.036 0.087 3.581 0.005 0.162 Desert 0.043 0.005 0.001 0.014 0.012 0.013 1.557 0.089 In 1.307 0.703 0.311 0.389 0.700 1.973 0.190 -
From 2000 to 2015, the rocky desertification area in southtern and western China continuously shrank, indicating an alleviated rocky densification condition. The rocky desertification area decreased from 13.200 × 104 km2 in 2000 to 10.950 × 104 km2 in 2010 and to 9.570 × 104 km2 in 2015. Compared with 2000, the rocky desertification area decreased by 17.08% in 2010 and by 27.49% in 2015.
However, changes in rocky desertification areas of different degrees show inconsistent trends with respect to the overall rocky desertification trend in southtern and western China. Specifically, the area of slight rocky desertification first decreased by 33.15% from 2000 to 2010 and then increased by 0.370 × 104 km2 in the following five years. The area of medium rocky desertification first increased by 10.48% from 2000 to 2010 and then decreased. As a result, its medium rocky desertification area in 2015 remained similar to that in 2000. On the other hand, strong and extreme rocky desertification areas show a trend consistent with the overall trend of rocky desertification in southtern and western China, which continuously decreased in area at an accelerated rate. The strong rocky desertification area decreased from 1.820 × 104 km2 in 2000 to 1.630 × 104 km2 in 2010 and then decreased by another 73.23% from 2010 to 2015. As a result, only 0.440 × 104 km2 of strong rocky desertification area was remained in 2015. In 2000, the extreme rocky desertification area was 0.290 × 104 km2, which decreased to 0.14 0× 104 km2 in 2010, shrinking by 52.64% and further declining by 60.48% from 2010 to 2015, after which only a 0.050 × 104 km2 area was remained (Fig. 3).
Figure 3. Rocky desertification area in southern and western China in 2000, 2010 and 2015
Table 7 shows the complexity and diversity in the process of rocky desertification in southern and western from 2000 to 2015 with a total of 24 transformation forms. The newly established rocky desertification area was 0.700 × 104 km2, which was dominated by slight rocky desertification areas (83.61%). In addition, the newly established medium rocky desertification area corresponded to 14.85%, and the newly established strong and extreme rocky desertification areas were rare, only 1.54%. From 2000 to 2015, rocky desertification disappeared in a total area of 4.330 × 104 km2 with varying rocky desertification degrees in the region. The area transformed from slight rocky desertification to no rocky desertification, which was the highest at 93.90%, while the areas transformed from medium, strong, and extreme rocky desertification to no rocky desertification were 4.83%, 1.54%, and 1.54%, respectively. The area transformed between different rocky desertification degrees (except the no rocky desertification degree) was 8.870 × 104 km2 with 57.54%, maintaining their initial rocky desertification degrees. The transformation of slight and medium rocky desertification areas from other degrees were 20.49% and 19.59%, respectively. In comparison, the transformed strong and extreme rocky desertification area was only 0.18%.
Table 7. Process of changes in rocky desertification areas in southern and western China from 2000 to 2015 / 104 km2
Year Degree 2015 None Slight Medium Strong Extreme 2000 None 0 58.530 10.400 0.990 0.090 Slight 40.660 15.440 12.060 0.750 0.040 Medium 20.910 66.430 123.570 4.790 0.190 Strong 4.760 9.990 54.890 13.360 0.450 Extreme 0.690 0.830 6.900 2.710 2.140 As shown in Table 8, the transformation of no rocky desertification to rocky desertification mostly occurred in Guizhou and Yunnan. Except of Guangdong, the other regions were dominated by transformations from no rocky desertification to slight rocky desertification. In comparison, Guangdong mostly experienced transformation from no rocky desertification to medium rocky desertification. From 2000 to 2015, among regions with disappearing rocky desertification areas, Guizhou, Sichuan and Hubei ranked the highest in disappearing areas. Among regions with unchanged rocky desertification areas, Guangdong and Guangxi and Chongqing had the highest percentages of medium rocky desertification areas, corresponding to 57.50%, 61.10%, and 55.24%, respectively. Among all types of rocky desertification transformations, the transformations to slight and medium degrees were dominating. Specifically, transformed slight rocky desertification was dominantly from medium rocky desertification with Hunan ranking the highest (54.72%). The transferred medium rocky desertification, on the other hand, was mostly from strong rocky desertification with Guangxi exhibiting the highest percentage (61.92%). Regarding the transformation of strong rocky desertification, Sichuan was the only region that mainly experienced degradation to extreme rocky desertification, while all other regions were dominated by transformations to medium rocky desertification. Finally, extreme rocky desertification was dominantly transformed to the strong degree.
Table 8. Transformation of rocky desertification in different degrees in southtern and western China from 2000 to 2015
Type Grades transformation Guangdong Guangxi Guizhou Hubei Hunan Sichuan Yunnan Chongqing New established Area / 104 km2 0.005 0.050 0.282 0.014 0.027 0.023 0.282 0.015 0→1 /% 39.04 51.19 86.78 73.64 77.91 87.90 88.63 66.46 0→2 /% 48.60 42.62 12.15 25.55 16.99 10.17 10.79 30.80 0→3 /% 12.16 5.81 0.93 0.75 4.58 1.91 0.54 2.46 0→4 /% 0.20 0.38 0.14 0.06 0.52 0.02 0.04 0.29 Disappear Area / 104 km2 0.013 0.251 1.216 0.471 0.396 0.821 0.900 0.258 1→0 /% 65.18 85.73 93.61 97.87 93.26 94.59 93.97 96.00 2→0 /% 17.65 10.42 5.28 1.76 5.59 4.12 4.90 3.16 3→0 /% 13.74 3.41 0.99 0.34 1.03 1.12 0.93 0.75 4→0 /% 3.43 0.44 0.12 0.02 0.12 0.17 0.21 0.09 Unchanged Area / 104 km2 0.051 1.104 1.402 0.255 0.505 0.252 1.199 0.339 1→1 /% 34.75 33.18 64.63 51.17 55.17 54.30 58.60 43.79 2→2 /% 57.50 61.10 33.27 47.85 37.45 32.53 33.38 55.24 3→3 /% 7.56 5.59 2.03 0.93 7.03 4.00 7.26 0.90 4→4 /% 0.20 0.14 0.07 0.05 0.35 9.17 0.75 0.06 Conversion Area / 104 km2 0.028 0.699 1.000 0.138 0.469 0.228 1.022 0.183 2→1 /% 28.69 25.63 46.39 47.10 54.72 42.35 41.84 36.45 3→1 /% 1.94 1.41 8.48 7.47 4.36 12.66 6.43 8.07 4→1 /% 0.34 0.10 0.54 0.44 0.25 1.74 0.61 0.67 1→2 /% 3.84 3.47 9.83 3.85 4.22 3.28 11.84 3.51 3→2 /% 53.20 61.92 26.27 33.95 29.22 24.79 25.83 42.21 4→2 /% 3.93 2.81 4.61 4.86 2.82 6.96 4.57 6.99 1→3 /% 0.26 0.16 0.49 0.27 0.10 0.34 0.98 0.22 2→3 /% 5.34 3.37 2.51 1.02 2.15 1.07 4.55 1.14 4→3 /% 2.05 0.88 0.45 0.94 1.95 6.42 2.56 0.64 1→4 /% 0.02 0.01 0.02 0.01 0.00 0.14 0.06 0.01 2→4 /% 0.14 0.06 0.19 0.04 0.03 0.06 0.16 0.04 3→4 /% 0.27 0.16 0.22 0.05 0.17 0.18 0.58 0.07 Note: 0, 1, 2, 3 ,4 are the rockey deserrification degree shown in Table 1 As shown in Table 9, from 2000 to 2015, 11.25% of the rocky desertification area was further degraded in southtern and western China, corresponding to a total area of 1.130 × 104 km2. Yunnan, Guizhou and Guangxi ranked the highest in their degraded areas: 0.467 × 104 km2, 0.415 × 104 km2, and 0.101 × 104 km2, respectively. The degraded rocky desertification area was the lowest in Guangdong, with 0.008 × 104 km2. A total of 5.106 ×104 km2 did not experience any changes in rocky desertification grade, including 1.402 × 104 km2 in Guizhou, which ranked the highest, followed by Yunnan and Guangxi with the areas of 1.199 × 104 km2 and 1.104 × 104 km2, respectively. From 2000 to 2015, a total area of 7.664 × 104 km2 experienced alleviated roc-ky desertification in the region, including 2.084 × 104 km2 in Guizhou, which ranked the highest; and 1.737 × 104 km2 in Yunnan and 1.037 × 104 km2 in Sichuan, which ranked second and third, respectively. Rocky desertification changes varied among the different regions. Specifically, rocky desertification conditions remained unchanged in Guangdong and Guangxi, while rocky desertification conditions were mostly alleviated in Sichuan, Hubei, Hunan, Guizhou, and Yunnan and in Chongqing. Among the regions with alleviated rocky desertification conditions, Sichuan ranked first in its rocky desertification area percentage at 78.35%, followed by Hubei at 68.50%.
Table 9. Changes in rocky desertification in southern and western China from 2000 to 2015
Difference-based rating Statistical Guangdong Guangxi Guizhou Hubei Hunan Sichuan Yunnan Chongqing Total Strong and slight degradation Area /104 km2 0.008 0.101 0.415 0.022 0.058 0.035 0.467 0.024 1.130 Ratio /% 7.89 4.80 10.64 2.47 4.15 2.64 13.74 3.05 No change Area /104 km2 0.051 1.104 1.402 0.255 0.505 0.252 1.199 0.339 5.106 Ratio/% 52.83 52.44 35.94 29.03 36.17 19.02 35.23 42.61 Strong and slight alleviation Area /104 km2 0.038 0.900 2.084 0.602 0.834 1.037 1.737 0.432 7.664 Ratio /% 39.28 42.76 53.42 68.50 59.67 78.35 51.04 54.34 From 2000 to 2015, a majority of rocky desertification areas were alleviated with a relatively small number of degraded areas. Regarding the spatial distribution (Fig. 4), degraded rocky desertification areas were mainly distributed in the southwestern part of Guizhou and south-eastern part of Yunnan; areas that experienced no change in rocky desertification grade were mostly distributed in Guizhou, the central and western part of Guangxi, and eastern Yunnan and at the junction of Chongqing, Hubei, and Guizhou.
Figure 4. Classification of rocky desertification changes in southtern and western China from 2000 to 2015
-
RDA ordination results show that among topographic factors, elevation, terrain slope, and topographic fractal dimension did not pass the Monte Carlo permutation test, indicating their low significance. In addition, socioeconomic factors are strongly correlated with rocky desertification changes, including GDP1 (Gross domestic product of primary industry), GDP (Gross domestic product), FixInv (Fixed-asset investment) and GDP3 (Gross domestic product of tertiary industry) (Table 10). Rocky desertification changes were not significantly correlated with Year-end population and gross domestic product of secondary industry. Regarding correlation with ecosystem factors, only farmland ecosystem and grassland ecosystem passed the significance test.
Table 10. Test results of influencing factors of rocky desertification changes in southern and western China
Factor F value P value Gross domestic product 14.3 0.002** Gross domestic product of primary industry 25.4 0.002** Gross domestic product of secondary industry 2.9 0.054 Gross domestic product of tertiary industry 3.4 0.020* Fixed-asset investment 7.2 0.002** Year-end population 1.2 0.298 Forest 2.3 0.106 Shrubland 1.7 0.170 Grassland 7.1 0.002** Farmland 78.3 0.002** City/town 2.3 0.098 Topography fractal dimension index 0.7 0.526 Elevation 1.6 0.200 Slope 2.4 0.054 Notes: * P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.001 In this study, we adopt the RDA ordination plot to illustrate the correlations of the six ecological and socioeconomic factors with the change in rocky desertification grades obtained by the Monte Carlo permutation test (Fig. 5). Fig. 5 shows positive correlations between the first RDA axis (AX1) and farmland ecosystem factors and GDP1, thus indicating that AX1 mainly reflects changes in agricultural indices. In comparison, the second axis (AX2) is strongly correlated with FixInv and grassland.
Figure 5. Redundancy analysis (RDA) ordination of ecological factors, economic factors and rocky desertification changes
Table 11 lists the eigenvalues and canonical coefficients of topographic and socioeconomic factors for the four ordination axes. The first four RDA ordination axes are able to explain 41.75% of the rocky desertification change. Specifically, the contribution of the first axis is the highest (34.90%), followed by the other three axes in decreasing order: 5.75%, 0.89%, and 0.21%, which were obtained from the cumulative percentage difference between the corresponding axis and the previous axis. These four axes are strongly correlated with environmental, socioeconomic, and ecological factors (Table 11). Among them, the first and second axes passed the significance test (Table 10).
Table 11. Environmental explanation of the first four axes of the redundancy analysis (RDA)
RDA axes 1 2 3 4 Eigenvalue 0.3490 0.0575 0.0089 0.0021 Cumulative percentage of rocky desertification change variables /% 34.90 40.65 41.54 41.75 Relationship between rocky desertification and environmental factors 0.7962 0.4113 0.3366 0.2907 Cumulative percentage of environmental factors / % 83.53 97.28 99.42 99.91 RDA partial ordination analytical results show that ecosystem variables and socioeconomic factors can only explain 8.2% and 12.0% of the rocky desertification change, respectively, while the ecological system variables and socioeconomic factors can synergistically explain another 15.6% of the rocky desertification change. The explanatory ability of the socioeconomic variables is apparently significantly higher than that of the ecosystem variables, revealing that human activities (socioeconomic factor variables) play a more significant role in controlling rocky desertification changes than the ecosystem variables.
The arrow length for each factor in Fig. 5 can reflect the significance of the factor in influencing the rocky desertification change pattern, which reveals that the main factors were farmland ecosystem, GDP1, GDP3, GDP, and FixInv. In addition, the relationships between rocky desertification grade changes and the different factors can be assessed based on the cosine value of the corresponding angle: the smaller the angle is, the more significant the correlation is. When the angle is 90°, the correlation equals zero, and when the angle is greater than 90°, the correlation becomes negative. The different alleviated grades of rocky desertification are driven by different factors. The strongly alleviated grade of rocky desertification is positively correlated with farmland ecosystem factors and GDP1, mainly because the reduced farmland area and improved farmland management increased the agricultural product and furthermore reduced its disturbance on rocky desertification, both of which assisted with the alleviated development of rocky desertification. The slightly alleviated rocky desertification was the dominant form of changes that occurred in the southern and western region, as continuously increasing GDP and GDP3 were favourable for slight alleviation in rocky desertification.
The strong rocky desertification grade is strongly correlated with farmland variables and GDP1, almost uncorrelated with FixInv, and negatively correlated with GDP3 and GDP. The southern and western region is dominated by agricultural population and barren soils, which led to excessive land reclamation, severe water and soil losses, and eventually intensified rocky desertification. In addition, given the rapid development of the tertiary industry in the southern and western region compared with the primary industry, the development of the land-independent economy precluded strong degradation in rocky desertification. Slight rocky desertification is positively correlated with GDP3 and GDP. Economic development induced a certain negative effect on rocky desertification, possibly resulting from the development of tourism in the karst rocky desertification region, because tourism is the leading industry in the tertiary industry and plays an important role in the economic construction in southern and western (Wang, 2019). The negative correlations of the slightly degraded rocky desertification with GDP1 and farmland area indicate that agriculture has no influence on controlling slight rocky desertification degradation.
-
About the evolution characteristics of karst rocky desertification, on the area of the research region, previous studies mainly focused on a certain county (Zhang et al., 2010; Pu et al., 2021), city (Hu et al., 2018; Pu et al., 2021) or province (Huang et al., 2006; Liu et al., 2008). southern and western, as a whole, boasts the largest rocky desertification area regions among the three largest karst regions in the world (Jiang et al., 2014; Jiang et al., 2020). As a result, it is more significant to in investigate the evolution characteristics of karst rocky desertification taking southern and western as a whole. As for the time frame of the study, either the time range of the study is narrow or it is long from now. For example, the time range is 4 years and 18 years far from now (Peng et al., 2013). Since 2000, China has paid more attention to rocky desertification in southern and western and adopted a series of control measures (Jiang et al., 2016). It is more meaningful to study the evolution law in a closer and wider time range for further guiding the control of rocky desertification. As a result, the investigation in this paper, in which the research area covers the whole southern and western region and the time covers 2000−2015, is more helpful for the further control of rocky desertification southern and western. From 2000 to 2015, rocky desertification in southern and western presented a benign development trend, which is consistent with the research conclusions of others (Jiang et al., 2016). In the driving mechanism of karst rocky desertification in southern and western, the change of rocky desertification is mainly influenced by human activities, which is also consistent with other researches (Wang, 2018). The population density of karst rocky desertification area is high, which is more than 1.5 times of the national average (
http://www.forestry.gov.cn/main/3457/20181214/161611806917453.html ). As a result, it is significantly important to control and transfer the population, and to achieve the moderate population goal in harmony with the law of ecological and economic development. Moreover, based on the ecological background of karst rocky desertification area, it is also significantly important to develop ecological agriculture and green industry, and to establish their own industrial clusters. From the driving mechanism of karst rocky desertification, it can be found that the karst tourism has a negative influence on the rocky desertification. The proportion of tertiary industry in the gross national product of southern and western is also gradually increasing, among which tourism is the leading industry of tertiary industry (Wang, 2019). The karst tourism should be programmed scientifically and developed moderately, and a long-term mechanism should be established to realize the value of ecological product. -
In this study, we analysed the changes in the karst ecosystem and changes in the characteristics of rocky desertification in eight regions in southtern and western China from 2000 to 2015. By considering socioeconomic, topographic, and ecological factors, we established a method to assess changes in the rocky desertification area and grade with the goal of revealing the mechanisms for karst rocky desertification in southtern and western China. Based on our analytical results, the following conclusions were drawn:ere drawn:
1) From 2000 to 2015, the areas of forest, grassland, wetland and desert increased in the southtern and western China; the areas of farmland, shrubland and bare land decreased; the city/town area significantly increased; and the area of farmland significantly decreased. In addition, the rocky desertification area continuously decreased during this period, corresponding to a benign development trend in rocky desertification, which is related to the rocky desertification control measures taken since 2000.
2) No strong correlations of rocky desertification change with topographic factors but did observe strong correlations with ecosystem and socioeconomic factors. These findings reveal that rocky desertification change is mainly controlled by human activities and that socioeconomic factors exhibit a stronger capability in explaining the dynamics of rocky desertification than ecosystem variables, which is consistent with other researches.
3) Reduced farmland area and improved farmland management were favourable to the alleviation of rocky desertification, while regional GDP and continuous development of the tertiary industry helped slightly alleviate rocky desertification. Increase in the land-independent economy prevented the occurrence of any strong degradation in rocky desertification grade. However, the development of tourism in southern and western China induced some slight degradation in rocky desertification, because tourism plays an important role in the economy in southtern and western China.hina.
Characteristics of Changes in Karst Rocky Desertification in Southtern and Western China and Driving Mechanisms
-
Abstract: The karst landform is a typical ecologically vulnerable region, and the problem of karst rocky desertification in southtern and western China has led to impoverishment and a degraded local ecological environment, which severely limits local socioeconomic development. An effective and appropriate control of karst rocky desertification in southtern and western China requires knowledge about its characteristics of variation and driving mechanisms. In this study, we chose eight regions in the southtern and western China as research areas and analysed the characteristics of the changes in karst ecosystem patterns and rocky desertification from 2000 to 2015. Based on these characteristics, we present the mechanisms that drive karst rocky desertification in the southtern and western China by utilizing the redundancy analysis (RDA) ordination method. The results show that the total area of rocky desertification in southtern and western China had been continuously decreasing from 2000 to 2015, revealing a positive development trend in rocky desertification. Rocky desertification variations were mainly affected by human activities. The reduction in farmland area improved farmland management and increased regional gross industrial product, which together with continuously rising gross domestic product of the tertiary industry caused a positive rocky desertification development. However, the local karst tourism has a certain effect on inducing slight rocky desertification.
-
Figure 2. Percentage changes in areas for different ecological systems from 2000 to 2015 in southtern and western China
Figure 3. Rocky desertification area in southern and western China in 2000, 2010 and 2015
Figure 4. Classification of rocky desertification changes in southtern and western China from 2000 to 2015
Figure 5. Redundancy analysis (RDA) ordination of ecological factors, economic factors and rocky desertification changes
Table 1. Classification of rocky desertification degree
Rocky desertification degree Numerical value Slope / (°) Vegetation coverage / % Lithology None 0 < 5 > 80 Buried carbonatite Slight 1 5−15 60−80 Sub-impure carbonate Medium 2 15−25 40− 60 Impure carbonate Strong 3 25−35 20−40 Dolomite Extreme 4 > 35 < 20 Limestone and dolomite 
下载: 导出CSV
Table 2. Rocky desertification change grade division
Value difference [−4, −3] [−2, −1] 0 [1, 2] [3, 4] Difference-based rating Strong degradation Slight degradation No change Slight alleviation Strong alleviation
下载: 导出CSV
Table 3. Selected socioeconomic driving factors of rocky desertification changes
Social and economic factors Abbreviation Unit Gross domestic product GDP 104 Yuan Gross domestic product of primary industry GDP1 104 Yuan Gross domestic product of secondary industry GDP2 104 Yuan Gross domestic product of tertiary industry GDP3 104 Yuan Fixed-asset investment FixInv 104 Yuan Year-end population PopYe 104
下载: 导出CSV
Table 4. Selected terrain driving factors of rocky desertification changes
No. Terrain factors Abbreviation Unit 1 Elevation ELEV m 2 Slope SLO ° 3 Topographic fractal dimension index TPI
下载: 导出CSV
Table 5. Selected ecosystem driving factors of rocky desertification changes
No. Ecosystem factors Abbreviation Unit 1 Forest FOE km2 2 Shrubland SHE km2 3 Grassland GRE km2 4 Farmland FAE km2 5 City/town TOE km2
下载: 导出CSV
Table 6. Ecosystem transfer matrix of southtern and western China from 2000 to 2015 / 104 km2
Year Ecosystem 2015 Forest Shrubland Grassland Wetland Farmland City/town Desert Out 2000 Forest 79.883 0.292 0.065 0.038 0.331 0.256 0.118 1.100 Shrubland 0.400 26.917 0.030 0.047 0.162 0.126 0.020 0.785 Grassland 0.125 0.044 20.591 0.024 0.038 0.065 0.005 0.301 Wetland 0.012 0.007 0.013 4.876 0.070 0.112 0.006 0.220 Farmland 0.704 0.343 0.200 0.230 50.673 1.401 0.036 2.916 City/town 0.020 0.012 0.001 0.036 0.087 3.581 0.005 0.162 Desert 0.043 0.005 0.001 0.014 0.012 0.013 1.557 0.089 In 1.307 0.703 0.311 0.389 0.700 1.973 0.190
下载: 导出CSV
Table 7. Process of changes in rocky desertification areas in southern and western China from 2000 to 2015 / 104 km2
Year Degree 2015 None Slight Medium Strong Extreme 2000 None 0 58.530 10.400 0.990 0.090 Slight 40.660 15.440 12.060 0.750 0.040 Medium 20.910 66.430 123.570 4.790 0.190 Strong 4.760 9.990 54.890 13.360 0.450 Extreme 0.690 0.830 6.900 2.710 2.140
下载: 导出CSV
Table 8. Transformation of rocky desertification in different degrees in southtern and western China from 2000 to 2015
Type Grades transformation Guangdong Guangxi Guizhou Hubei Hunan Sichuan Yunnan Chongqing New established Area / 104 km2 0.005 0.050 0.282 0.014 0.027 0.023 0.282 0.015 0→1 /% 39.04 51.19 86.78 73.64 77.91 87.90 88.63 66.46 0→2 /% 48.60 42.62 12.15 25.55 16.99 10.17 10.79 30.80 0→3 /% 12.16 5.81 0.93 0.75 4.58 1.91 0.54 2.46 0→4 /% 0.20 0.38 0.14 0.06 0.52 0.02 0.04 0.29 Disappear Area / 104 km2 0.013 0.251 1.216 0.471 0.396 0.821 0.900 0.258 1→0 /% 65.18 85.73 93.61 97.87 93.26 94.59 93.97 96.00 2→0 /% 17.65 10.42 5.28 1.76 5.59 4.12 4.90 3.16 3→0 /% 13.74 3.41 0.99 0.34 1.03 1.12 0.93 0.75 4→0 /% 3.43 0.44 0.12 0.02 0.12 0.17 0.21 0.09 Unchanged Area / 104 km2 0.051 1.104 1.402 0.255 0.505 0.252 1.199 0.339 1→1 /% 34.75 33.18 64.63 51.17 55.17 54.30 58.60 43.79 2→2 /% 57.50 61.10 33.27 47.85 37.45 32.53 33.38 55.24 3→3 /% 7.56 5.59 2.03 0.93 7.03 4.00 7.26 0.90 4→4 /% 0.20 0.14 0.07 0.05 0.35 9.17 0.75 0.06 Conversion Area / 104 km2 0.028 0.699 1.000 0.138 0.469 0.228 1.022 0.183 2→1 /% 28.69 25.63 46.39 47.10 54.72 42.35 41.84 36.45 3→1 /% 1.94 1.41 8.48 7.47 4.36 12.66 6.43 8.07 4→1 /% 0.34 0.10 0.54 0.44 0.25 1.74 0.61 0.67 1→2 /% 3.84 3.47 9.83 3.85 4.22 3.28 11.84 3.51 3→2 /% 53.20 61.92 26.27 33.95 29.22 24.79 25.83 42.21 4→2 /% 3.93 2.81 4.61 4.86 2.82 6.96 4.57 6.99 1→3 /% 0.26 0.16 0.49 0.27 0.10 0.34 0.98 0.22 2→3 /% 5.34 3.37 2.51 1.02 2.15 1.07 4.55 1.14 4→3 /% 2.05 0.88 0.45 0.94 1.95 6.42 2.56 0.64 1→4 /% 0.02 0.01 0.02 0.01 0.00 0.14 0.06 0.01 2→4 /% 0.14 0.06 0.19 0.04 0.03 0.06 0.16 0.04 3→4 /% 0.27 0.16 0.22 0.05 0.17 0.18 0.58 0.07 Note: 0, 1, 2, 3 ,4 are the rockey deserrification degree shown in Table 1
下载: 导出CSV
Table 9. Changes in rocky desertification in southern and western China from 2000 to 2015
Difference-based rating Statistical Guangdong Guangxi Guizhou Hubei Hunan Sichuan Yunnan Chongqing Total Strong and slight degradation Area /104 km2 0.008 0.101 0.415 0.022 0.058 0.035 0.467 0.024 1.130 Ratio /% 7.89 4.80 10.64 2.47 4.15 2.64 13.74 3.05 No change Area /104 km2 0.051 1.104 1.402 0.255 0.505 0.252 1.199 0.339 5.106 Ratio/% 52.83 52.44 35.94 29.03 36.17 19.02 35.23 42.61 Strong and slight alleviation Area /104 km2 0.038 0.900 2.084 0.602 0.834 1.037 1.737 0.432 7.664 Ratio /% 39.28 42.76 53.42 68.50 59.67 78.35 51.04 54.34
下载: 导出CSV
Table 10. Test results of influencing factors of rocky desertification changes in southern and western China
Factor F value P value Gross domestic product 14.3 0.002** Gross domestic product of primary industry 25.4 0.002** Gross domestic product of secondary industry 2.9 0.054 Gross domestic product of tertiary industry 3.4 0.020* Fixed-asset investment 7.2 0.002** Year-end population 1.2 0.298 Forest 2.3 0.106 Shrubland 1.7 0.170 Grassland 7.1 0.002** Farmland 78.3 0.002** City/town 2.3 0.098 Topography fractal dimension index 0.7 0.526 Elevation 1.6 0.200 Slope 2.4 0.054 Notes: * P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.001
下载: 导出CSV
Table 11. Environmental explanation of the first four axes of the redundancy analysis (RDA)
RDA axes 1 2 3 4 Eigenvalue 0.3490 0.0575 0.0089 0.0021 Cumulative percentage of rocky desertification change variables /% 34.90 40.65 41.54 41.75 Relationship between rocky desertification and environmental factors 0.7962 0.4113 0.3366 0.2907 Cumulative percentage of environmental factors / % 83.53 97.28 99.42 99.91
下载: 导出CSV
-
[1] Cao Jianhua, Yuan Daoxian, Zhang Cheng et al., 2004. Karst ecosystem constrained by geological conditions in Southwest China. Earth and Environment, 1: 1–8. (in Chinese) [2] Deng Jufen, Cui Geying, Wang Yuedong et al., 2009. Rocky desertification and comprehensive improvement of karst areas in Yunnan. Pratacultural Science, 26(2): 33–38. (in Chinese) [3] Ford D C, Williams P, 2007. Karst Hydrogeology And Geomorphology, New York: John Wiley and Sons Ltd. , 77−102. [4] Gams I, Gabrovec M, 1999. Land use and human impact in the Dinaric karst. International Journal of Speleology, 28(1): 55–70. doi: 10.5038/1827-806X.28.1.4 [5] Han Zhaoqing, 2006. Exploitation of Guizhou Province during the Yongzheng reign period and its effect on the rock-desertification in this area. Fudan Journal (Social Sciences Edition), 2: 120–127+140. (in Chinese) [6] Hu Y F, Han Y Q, Zhang Y Z, 2018. Information extraction and spatial distribution of research hot regions on rocky desertification in China. Applied Sciences, 8(11): 2075. doi: 10.3390/app8112075 [7] Huang Q H, Cai Y L, 2006. Assessment of karst rocky desertification using the radial basis function network model and GIS technique: a case study of Guizhou Province, China. Environmental Geology, 49: 1173–1179. doi: 10.1007/s00254-005-0162-4 [8] Ji Qifang, 2013. Effects of vegetation cover on runoff and sediment reduction on slop land in karst areas of Guizhou Province. Nanjing: Nanjing University. (in Chinese) [9] Jiang M, Lin Y, Chan T O et al., 2020. Geologic factors leadingly drawing the macroecological pattern of rocky desertification in southwest China. Scientific Reports, 10(1): 1440. doi: 10.1038/s41598-020-58550-1 [10] Jiang Z C, Lian Y Q, Qin X Q, 2014. Rocky desertification in Southwest China: impacts, causes, and restoration. Earth-Science Reviews, 132: 1–12. doi: 10.1016/j.earscirev.2014.01.005 [11] Jiang Zhongcheng, Luo Weiqun, Tong Lliqiang et al., 2016. Evolution features of rocky desertification and influence factors in karst areas of southwest China in the 21st century. Carsologica Sinica, 35(5): 461–468. [12] Lavee H, Imeson A C, Sarah P, 1998. The impact of climate change on geomorphology and desertification along a Mediterranean-arid transect. Land Degradation & Development, 9(5): 407–422. doi: 10.1002/(SICI)1099-145X(199809/10)9:5<407::AID-LDR302>3.0.CO;2-6 [13] Li Ruiling, Wang Shijie, Zhou Dequan et al., 2003. The correlation between rock desertification and lithology in karst area of Guizhou. Acta Geographica Sinica, 58(2): 314–320. (in Chinese) [14] Li Sen, Donguxiang, Wang Jinhua et al., 2007. Re-discussion on the concept and classification of rocky desertification. Carsologica Sinica, 4: 279–284. (in Chinese) [15] Li Yangbing, Tan Qiu, Wang Shijie et al., 2005. Current status, problems analysis and basic framework of karst rocky desertification research. Science of Soil and Water Conservation, 3: 27–34. (in Chinese). doi: 10.16843/j.sswc.2005.03.006 [16] Liu Y S, Wang J Y, Deng X Z, 2008. Rocky land desertification and its driving forces in the karst areas of rural Guangxi, Southwest China. Journal of Mountain Science, 5: 350–357. doi: 10.1007/s11629-008-0217-6 [17] Lou Fangji, 2016. Quantitative evaluation and Analysis of Rocky Desertification Remote Sensing in Karst Mountainous Area based on Landscape Unit. Guiyang: Guizhou Normal University. (in Chinese) [18] Luo Xuling, Wang Shijie, Bai Xiaoyong et al., 2021. Analysis on the spatio-temporal evolution process of rocky desertification in Southwest Karst area. Acta Ecologica Sinica, 41(2): 680–693. (in Chinese) [19] Lv Minghui, Wang Hongya, Cai Yunlong, 2007. General review of soil erosion in the karst area of Southwest China. Progress in Geography, 26(2): 87–96. (in Chinese) [20] Ma Lichi, Wang Jinliang, Liu Guangjie, 2015. Spatial variation of rock desertification at typical karst region in southeastern part of Yunnan Province. Bulletin of Soil and Water Conservation, 35(5):327−333. (in Chinese) [21] Peng J, Xu Y Q, Zhang R et al., 2013. Soil erosion monitoring and its implication in a limestone land suffering from rocky desertification in the Huajiang Canyon, Guizhou, Southwest China. Environmental Earth Sciences, 69(3): 831–841. doi: 10.1007/s12665-012-1968-5 [22] Phillips J D, 2016. Biogeomorphology and contingent ecosystem engineering in karst landscapes. Progress in Physical Geography:Earth and Environment, 40(4): 503–526. doi: 10.1177/0309133315624641 [23] Pu J W, Zhao X Q, Dong P L et al., 2021. Extracting information on rocky desertification from satellite images: A comparative study. Remote Sensing, 13(13): 2497. doi: 10.3390/rs13132497 [24] Qin Xiaoqun, Zhu Mingqiu, Jiang Zhongcheng, 2006. A review on recent advances in rocky desertification in south west China karst region. Carsologica Sinica, 25(3): 234–238. (in Chinese) [25] Shi Yingchun, Shu Yingge, 2017. Analysis on karst rocky desertification temporal and spatial variation characteristics and driving factors—a case study of Qinglong County of Guizhou Province. Forest Resources Management, (1): 135–143+152. (in Chinese) [26] Su Weici, 2002. Controlling model for rocky desertification of karst mountainous region and its preventing strategy in Southwest China. Journal of Soil and Water Conservation, 2: 29–32. (in Chinese). doi: 10.13870/j.cnki.stbcxb.2002.02.008 [27] Su Weici, Yang Hua, LI Qing et al., 2006. Rocky land desertification and its controlling measurements in the karst mountainous region Southwest of China. Chinese Journal of Soil Science, 37(3): 447–451. (in Chinese) [28] Su Weici, Zhou Jizuo, 1995. Rocky desertification in Guizhou karst region and its preventive strategy. Resources and Environment in the Yangtze Basin, (2): 177–182. (in Chinese) [29] Sunkar A, 2008. Deforestation and rocky desertification processes in Gunung Sewu karst landscape. Media Konservasi, 13(3): 1–7. doi: 10.29244/medkon.13.3.%25p [30] Tian Pengju Xu Dandan, Ding Liguo et al., 2017. Analysis of spatial-temporal variation characteristic of vegetation in Guizhou during 2005-2014 period based on MODIS -NDVI. Journal of Guizhou Meteorology, 41(2): 8–13. (in Chinese) [31] Tong X W, Wang K L, Yue Y M et al., 2017. Quantifying the effectiveness of ecological restoration projects on long-term vegetation dynamics in the karst regions of Southwest China. International Journal of Applied Earth Observation and Geoinformation, 54: 105–113. doi: 10.1016/j.jag.2016.09.013 [32] Veress M, 2020. Karst types and their karstification. Journal of Earth Science, 31(3): 621–634. doi: 10.1007/s12583-020-1306-x [33] Wang Jialu, Li Weijie, Wang Yong et al., 2021. Spatial-temporal variation of NDVI and its responses to hydrothermal condition in rocky desertification area of Chongqing City from 2005 to 2014. Research of Soil and Water Conservation, 28(2): 217–223. (in Chinese) [34] Wang Shijie, Li Yangbing, Li Ruiling, 2003. Karst rocky desertification: formation background, evolution and comprehensive taming. Quaternary Sciences, 23(6): 657–666. (in Chinese) [35] Wang Xiaofan, 2018. Research on Mutual Feed between Land Rocky Desertification Succession and Socio-economic Activities in Guizhou. Qufu: Qufu Normal University. (in Chinese) [36] Wang Yao, 2019. Research on the Linkage of Tourism Industry in Guizhou Province to the Tertiary Industry. Guiyang: Guizhou University. (in Chinese) [37] Wen Linqin, Li Zhongfei, 2020. Evolution characteristics of rocky desertification during 2004–2016 in Guizhou Province, China. Acta Ecologica Sinica, 40(17): 5928–5939. (in Chinese) [38] Weng Jintao, 1995. The effect of carbonate rocks on global carbon cycle. Advances in Earth Science, 10(2): 154–158. (in Chinese) [39] Williams V, 2011. A case study of the desertification of Haiti. Journal of Sustainable Development, 4(3): 20–31. doi: 10.5539/jsd.v4n3p20 [40] Wu Yingke, Bi Yuyuan, Guo Chunqin, 1998. A summary of basic features, resources, environment, sociality and economy in the Karst areas of South-west China. Carsologica Sinica, 17(2): 141–150. (in Chinese) [41] Xu D Y, Song A L, Li D J et al., 2019. Assessing the relative role of climate change and human activities in desertification of North China from 1981 to 2010. Frontiers of Earth Science, 13(1): 43–54. doi: 10.1007/s11707-018-0706-z [42] Yan Lihui, 2018. Study on RS-informatic Tupu of Rocky Desertification Evolution in Karst Plateau—A Case Study in Huajiang Area. Guiyang: Guizhou Normal University. (in Chinese) [43] Yuan Daoxian, 1994. Chinese Karstology. Beijing: Geological Publishing House. (in Chinese) [44] Zhang Lin, 2018. Dynamic Changes of Karst rocky Desertification in Wuling Mountain Area and Their driving Mechanism—Taking the Pengshui Hmong and Tujia Autonomic County as An Example. Chongqing: Chongqing Normal University. (in Chinese) [45] Zhang M Y, Wang K L, Liu H Y et al., 2015. How ecological restoration alters ecosystem services: an analysis of vegetation carbon sequestration in the karst area of northwest Guangxi, China. Environmental Earth Sciences, 74(6): 5307–5317. doi: 10.1007/s12665-015-4542-0 [46] Zhang Panpan, Hu Yuanman, Xiao Duning et al., 2010. Rocky desertification risk zone delineation in karst plateau area: a case study in Puding County, Guizhou Province. Chinese Geographical Science, 20(1): 84–90. doi: 10.1007/s11769-010-0084-2 [47] Zhang Pingcang, Ding Wenfeng, 2008. Review on rock desertification research in China. Journal of Yangtze River Scientific Research Institute, 25(3): 1–5. (in Chinese) [48] Zhang Xinbao, Wang Shijie, Meng Tianyou, 2012. The management model of rocky desertification slope farmland. Soil and Water Conservation in China, (9): 41–44. (in Chinese) [49] Zhou Zhongfa, Huang Lujia, 2003. An analysis on relation of rock desertification to stratum and lithology in karst region—a case study at Qingzhen City of Guizhou Plateau. Bulletin of Soil and Water Conservation, 1: 19–22. (in Chinese) -
点击查看大图
图(5) / 表ll (11)
计量
- 文章访问数: 248
- HTML全文浏览量: 48
- PDF下载量: 40
- 被引次数: 0
