To quantify the current status of soil erosion in the Min-Tuo River Basin and predict future soil erosion conditions, the RUSLE model and GIS technology were employed to assess soil erosion in the Min-Tuo River Basin from 2000 to 2020. Based on the CMIP6 model, precipitation in the Min-Tuo River from 2021 to 2050 was predicted, and the CA-Markov model was used to forecast land use types from 2030 to 2050, thereby predicting the spatial distribution trends of soil erosion in 2030, 2040, and 2050. The results indicate that: (1) The overall soil erosion intensity in the study area increased from 2000 to 2020; (2) Future precipitation in the basin, predicted using the CMIP6 model, showed a fluctuating upward trend compared to historical levels. The SSP5-8.5 scenario had the largest fluctuation; (3) In terms of future land use types, an increase in the area of converted cropland may increase the risk of soil erosion; (4) The area affected by soil erosion in the future will increase, with the highest annual average soil erosion modulus and total amount occurring under the SSP5-8.5 high emission scenario; (5) The dynamic evolution of future soil erosion levels indicates that the stability of soil erosion levels under the SSP2-4.5 scenario is higher than under the other scenarios. In the future high CO2 emission scenario, the risk of soil erosion in the Min-Tuo River Basin will increase, necessitating the active implementation of various prevention and control measures to prevent further exacerbation.
Research Article
High greenhouse gas emission scenarios increase soil erosion risk: A case study of the Min-Tuo River Basin, China
https://doi.org/10.21203/rs.3.rs-4952297/v1
This work is licensed under a CC BY 4.0 License
You are reading this latest preprint version
Soil erosion
CMIP6
RUSLE model
CA-Markov model
Min-Tuo River Basin
Soil erosion is a global environmental threat (Keno & Suryabhagavan, 2015). It is the main cause of soil degradation and is related to soil compaction, soil structure and organic matter loss, and soil acidity (Hessel et al., 2011). Almost no form of land on our planet is free from soil erosion (Bekele, 2019). Approximately 30% of the global land area and 80% of ag-ricultural land experience soil erosion, with around 3 billion people living on degraded land (Mirzabaev et al., 2016). Soil erosion is particularly severe in developing regions with high population density and ecological vulnerability (Renard et al., 2011). The characteristics of the Min-Tuo River Basin align with these conditions. Additionally, severe forest resource de-struction, the 2008 Wenchuan earthquake (Guo et al., 2012), and extreme rainfall events (Piao et al., 2019) have further worsened the region's ability to retain soil and resulted in local soil desertification and lithification, leading to serious soil and water loss in the area. Therefore, quantifying soil loss and predicting future soil erosion is of great significance as it can provide scientific basis for watershed soil and water loss management and macro-scale decision-making for soil and water conservation, thus promoting ecological environment protection and restoration in the watershed.
The issue of soil erosion has become a subject of study for numerous scholars. Among these, the use of soil erosion models has proven to be the most widely effective method for researching soil erosion. Several scholars have successively developed models such as the USLE (Wischmeier & Smith, 1958), WEPP (Woodward, 1999), and LISEM (De Roo, 1996), with the USLE model being the most widely utilized. Subsequently, scholars have made supplementary revisions to the USLE model. In 1997, the United States Department of Agriculture proposed the RUSLE model (Renard & Ferreira, 1993), which comprehensively reflects the influencing factors of soil erosion and has been widely used (Rymszewicz et al., 2015; Ma et al., 2023; Wang & Su, 2020).
With the development of GIS and RS technologies, their combined use allows for the comprehensive exploration of spatial information and patterns related to soil erosion, thereby providing a technological possibility for studying the spatiotemporal variations in soil erosion. Over the past century, global climate has been trending significantly warmer (Suresh, 2023; Hairsine & Rose, 1992), leading to increased regional differences in precipitation. Rainfall is one of the important factors influencing soil erosion, and global climate change can cause changes in regional precipitation, thereby affecting soil erosion conditions.Soil erosion is a crucial aspect of sustainable development. Analyzing soil erosion factors and processes using soil erosion models can provide a theoretical basis for formulating regional soil and water conservation measures (Nearing et al., 1999). Therefore, this study is based on the RUSLE model to investigate the current status of soil erosion in the Min-Tuo River Basin and predict future soil erosion conditions based on the CMIP6 model.
In the past, many related studies have studied the current situation of soil erosion in the Min-Tuo River Basin. For instance, Zhong et al. (2022) used the RUSLE model to assess soil erosion in the Tuojiang River from 2000 to 2018 and explored the spatiotemporal dynamic evolution patterns. Deng & Shi (2022) quantitatively evaluated soil erosion and the spatial distribution of soil and water conservation ecosystem services in the Min-Tuo River Basin using the USLE model. Zhang et al. (2020) generated a soil erosion modulus map for the Min-Tuo River Basin using RUSLE based on supervised classification of the current land use map, indicating that the basin is mainly characterized by mild and moderate erosion areas. Liu et al. (2019) investigated the erosion status and characteristics of the upper reaches of the Min-Tuo River Basin and explored erosion risk in the region using a GIS weighted overlay analysis method. However, there is currently very little research on the overall soil and water loss situation in the entire Min-Tuo River Basin after 2010.
Given the background mentioned above, this study focuses on the Min-Tuo River Basin as the main area. Through the use of remote sensing images, precipitation records, DEM, soil types, vegetation types, land use, and other data, the study employed the RUSLE model to analyze soil erosion in the Min-Tuo River Basin from 2000 to 2020. Furthermore, based on the CMIP6 model, six models and three different emission scenarios were selected. After bias correction and downscaling using the multi-model ensemble (MME) approach, the study predicted the precipitation in the basin from 2021 to 2050. Additionally, with the assistance of the CA-Markov model, the study forecasted land use types for the years 2030, 2040, and 2050, and subsequently predicted the soil erosion situation in the Min-Tuo River Basin from 2030 to 2050. This research can provide theoretical references for soil erosion control and soil conservation measures in the study area under future climate change scenarios. It can also scientifically predict the soil erosion in large river basins in China at a spatial scale, aiming to provide a scientific and rational research basis for the ecological environment and comprehensive management projects in the Min-Tuo River Basin.
2.1 Study area
The The Min-Tuo River Basin (Fig. 1) is mainly composed of four parts: the Minjiang, Dadu River, Qingyijiang, and Tuojiang (Hu et al., 2020). Particularly, the upper reaches of the Minjiang and the arid valleys of the Dadu River have extremely fragile ecosystems. Under the dual pressure of population and economic development, the soil erosion in the Min-Tuo River Basin is becoming more and more serious. With the acceleration of urbanization, industrialization and agricultural modernization, the expansion of economic scale in the future will have an important impact on soil erosion in the Min-Tuo River Basin. At the same time, with the implementation of national strategies such as the Yangtze River Economic Belt and the Chengdu-Chongqing Economic Circle, higher requirements have been put forward for soil and water conservation in river basins.
2.2 RULSE model
The In recent years, with the maturity of "3S" technology, the difficulty of obtaining geographical information by traditional research methods has been overcome, the efficiency of soil erosion assessment has been improved, and the RUSLE model has been better applied. The RUSLE model expression is as follows:
\(\:A=R\times\:K\times\:L\times\:S\times\:C\times\:\) P (1)
Where \(\:\text{A}\) is soil and water conservation quantity (t/hm2·a);\(\:\:\text{R}\) is rainfall erosion factor (MJ·mm/hm2·h·a); \(\:\text{K}\)is soil erodibility factor [t·h/(MJ·mm)];\(\:\:\text{L}\text{S}\) is topographic factor, dimensionless;\(\:\:\text{C}\) is vegetation cover factor, dimensionless; P is soil and water conservation measure factor, dimensionless.
The rainfall data partly comes from the National Science and Technology Infrastructure of China, using the 1km resolution annual precipitation data from 2001 to 2020 in China. This data is derived by synthesizing the yearly cumulative sum based on the monthly precipitation data set at a 1km resolution by Professor Peng Shouzhang from 1901 to 2020, as referenced in Peng et al. (2017) and Su et al. (2022). Another part is sourced from the Resource and Environment Data Center of the Chinese Academy of Sciences, which is the spatially interpolated data set of annual mean temperature and annual precipitation since 1980 in China. This data set is based on daily observational data from over 2400 meteorological stations in China, processed through collection, statistics, and spatial interpolation calculations.
The NDVI data is sourced from the National Ecological Science Data Center of China, as referenced in Yang et al. (2019). The land use/cover data is sourced from the Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences. The soil data is sourced from the World Soil Database, where the HWSD (Harmonized World Soil Database) can be queried for soil texture data, including sand, silt, clay, and organic carbon content. The terrain data uses a 30m resolution Digital Elevation Model (DEM) data sourced from the CGIAR-CSI SRTM Elevation Database.
For the calculation of soil and water conservation, it is necessary to ensure consistency in the resolution and projection coordinate system of each factor when using them collectively. Please refer to Table 1 for the sources and usages of the relevant data.
| Data classification | Processing | Resolution | Data sources |
|---|---|---|---|
| Rainfall data | Estimated R-value | 1km | Data Center for Resources and Environment, CAS/National Earth System Science Data Center |
| Soil data | Estimated K value | 1km | World Soil Database |
| Terrain data | Estimated LS value | 30m | CGIAR-CSI SRTM Elevation Database |
| NDVI data | Estimated C value | 30m | National Ecological Science Data Center |
| Land use/cover data | Estimated P value | 30m | Institute of Geographical Sciences and Natural Resources, CAS |
2.2.1. Rainfall erosivity factor(\(\:R\))
Rainfall erosivity factor is one of the potential forces causing soil erosion, which can be judged by annual rainfall data.Guidelines for Calculation of Soil Loss Amount of Production and Construction Projects (SL773-2018).The formula for calculating \(\:\text{R}\) is as follows:
$$\:R=0.067{P}_{d}^{1.627}$$
2
Where \(\:R\) is the erosive factor of annual average rainfall, and the unit is MJ·mm/(hm²·h); \(\:{P}_{d}\) is annual average rainfall, unit is mm.
2.2.2. Soil erodibility factor(\(\:K\))
Soil erodibility factor is one of the essential parameters of RUSLE model, which is used to reflect soil sensitivity to erosion. \(\:\text{K}\) can be determined by soil texture data by Rao et al.(2020)、Williams (1990) and Fan et al.(2012). An algorithm for estimating soil erodibility factor in EPIC model is proposed to calculate \(\:\text{K}\). The specific formula is as follows:
Where \(\:\text{K}\) is soil erodibility factor, the unit is t·h/(MJ·mm);\(\:\:\text{S}\text{A}\text{N}\:\)is sand content (%); \(\:\text{S}\text{I}\text{L}\) is silt content (%); \(\:\text{C}\text{L}\text{A}\) is clay content (%); \(\:\text{C}\) is organic carbon content (%).
2.2.3. Topographic factors (\(\:LS\))
Topographic factor is also one of the main causes of soil erosion. Slope length factor (\(\:\text{L}\)) and slope gradient factor (\(\:\text{S}\)) are selected in RUSLE model. \(\:\text{L}\text{S}\) reflects the influence of topography fluctuation on soil erosion. The greater the \(\:\text{L}\text{S}\), the faster the soil erosion rate Although it is difficult to measure \(\:\text{L}\text{S}\) in large-scale research, the influence of topographic factors on soil erosion can still be expressed by topographic fluctuation rate, that is, the maximum height change in a certain height area of the ground (Wischmeier & Smith, 1978). The specific formula is as follows:
$$\:H={H}_{max}-{H}_{min}$$
5
Where \(\:H\) topographic relief, \(\:{H}_{max}\) is the highest value of elevation, \(\:{H}_{min}\) is the lowest value of elevation.
2.2.4. Vegetation cover factor(\(\:C\))
Vegetation cover factor plays an important role in soil and water conservation, \(\:\text{C}\:\)ranges from Between 0 and 1(Yang & Huang, 2021). When \(\:\text{C}\)=1, the ground is completely bare without vegetation; when \(\:\text{C}\)=0 the ground vegetation coverage is good. It is a common way to obtain vegetation coverage from remote sensing images and calculate vegetation coverage factors according to vegetation coverage. Cai et al.(2000)The proposed algorithm is used to estimate the vegetation cover factor through artificial and natural rainfall experiments. The specific formula is as follows:
fvc \(\:=\frac{NDVI-{NDVI}_{min}}{{NDVI}_{max}-{NDVI}_{min}}\) (6)
$$\:C=\left\{\begin{array}{c}1,\:fvc\le\:0.095\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\\\:0.6508-0.3436Inc,0.095\:\:<\:fvc<0.783\:\:\\\:0,\:fvc\ge\:0.783\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\:\end{array}\right.\:\:$$
7
Where \(\:\text{N}\text{D}\text{V}\text{I}\) is normalized vegetation index(Wei et al., 2022); \(\:{\text{N}\text{D}\text{V}\text{I}}_{\text{m}\text{i}\text{n}}\) and \(\:{\text{N}\text{D}\text{V}\text{I}}_{\text{m}\text{a}\text{x}}\) are the minimum and maximum values of \(\:\text{N}\text{D}\text{V}\text{I}\) respectively; \(\:\text{f}\text{v}\text{c}\) is vegetation cover index (%);\(\:\:\text{C}\) is vegetation cover factor.
2.2.5. Soil and water conservation measures(\(\:P\))
Soil and water conservation measures can regulate the amount of soil erosion in a region. The factor of soil and water conservation measures is difficult to calculate directly. Usually, the value of \(\:\text{P}\) factor of a region is determined according to different land use types. The range of \(\:\text{P}\) is between 0 and 1. The larger \(\:\text{P}\) is, the smaller the amount of soil erosion is. When \(\:\text{P}\)=0, it means that no erosion occurs. When \(\:\text{P}\)=1, it means that no soil and water conservation measures are taken (Humphrey et al., 2021). According to the research conclusions of relevant scholars (Peng et al., 2007). The following values were assigned to the study areas with different land use properties (Table 2). Using ArcGIS software to reclassify the assigned data, you can obtain \(\:\text{P}\) values and their distribution maps for different land use types within the basin.
| land use types | cultivated land | forest land | grass | waters | construction land | unused land |
|---|---|---|---|---|---|---|
| soil and water conservation factor | 0.25 | 1 | 1 | 0 | 0 | 0 |
2.3. CMIP6 climate model data
According to the applicability of CMIP6 in the study area (Lu et al., 2022; Wu et al., 2023), six atmospheric circulation models with better performance in CMIP6 (Table 3) were selected. Due to the differences among the selected six climate models, the bilinear interpolation method was used to unify the spatial resolution of all models. After the interpolation, Delta downscaling (Chen et al., 2011) was performed on the interpolated model data at each grid. Finally, the daily meteorological data for the Min-Tuo River Basin under three shared socioeconomic pathways (Table 4) were obtained for the future period.
| Number | Schema name | Countries | Spatial resolution (°) |
|---|---|---|---|
| 1 | ACCESS-ESM1-5(ACC) | Australia | 1.2°× 1.8° |
| 2 | BCC-CSM2-MR(BCC) | China | 2.8°× 2.8° |
| 3 | CanESM5(CAN) | Canadian | 2.8°× 2.8° |
| 4 | IPSL-CM6A-LR(IPSL | France | 1.3°× 2.5° |
| 5 | MRI-ESM2-0(MRI) | Japan | 1.1°× 1.1° |
| 6 | MIROC6(MIR) | Japan | 1.4°× 1.4° |
| Number | Path name | Forced category | Shared socio-economic pathways | 2100 year radiative forcing/(W/m²) |
|---|---|---|---|---|
| 1 | SSP 1-2.6 | low forcing | sustainable development path | 2.6 |
| 2 | SSP 2-4.5 | low forcing | intermediate path | 4.5 |
| 3 | SSP 5-8.5 | high forcing | Traditional fossil fuel-based pathways | 8.5 |
2.2.4. CMIP6 data processing
The Delta method (Chen et al, 2011) is used to downscale and bias correct the CMIP6 data of the selected 6 models. This method compares the differences between GCMs output and observational data, and downscales the climate projection from large scale to smaller spatial scales. It has the advantages of strong interpretability, high accuracy, and high computational efficiency. When describing the changes in climate elements, precipitation is represented using the rate of change, while temperature is represented using the amount of change. The calculation formulas for future precipitation and temperature scenarios at meteorological stations are as follows:
$$\:{P}_{f}={P}_{o}\bullet\:\frac{{P}_{Gf}}{{P}_{Go}}$$
8
$$\:{T}_{f}={T}_{o}+\left({T}_{Gf}-{T}_{Go}\right)$$
9
In the equations, \(\:{\text{P}}_{\text{f}}\)、\(\:{\text{T}}_{\text{f}}\)represent the reconstructed future precipitation and temperature sequences; \(\:{\text{P}}_{\text{G}\text{f}}\)、\(\:{\text{T}}_{\text{G}\text{f}}\) represent the future precipitation and temperature sequences estimated by the climate model; \(\:{\text{P}}_{\text{G}\text{o}}\)、\(\:{\text{T}}_{\text{G}\text{o}}\) represent the multi-year average precipitation and temperature simulated by the climate model during the reference period; \(\:{\text{P}}_{\text{o}}\)、\(\:{\text{T}}_{\text{o}}\) represent the multi-year average precipitation and temperature of the observed field during the reference period.
The standardized Taylor diagram (Taylor, 2001) is used to assess the model performance of different climate models. It effectively and intuitively illustrates the differences in performance among multiple models and the magnitude of the errors between simulated and actual values, and has been widely applied in climate model evaluation research. Single-model to multi-model ensemble prediction is one of the effective ways to improve model accuracy, and the calculation formula for using MME is as follows:
$$\:{F}_{MME}=\frac{1}{n}\sum\:_{i=1}^{n}{F}_{i}$$
10
In the equation, \(\:{\text{F}}_{\text{M}\text{M}\text{E}\:}\)represents the average result of the multi-model ensemble, and\(\:\:{\text{F}}_{\text{i}\:}\:\)represents the simulation results of each individual model.
Quantitative evaluation was conducted on the future precipitation results of the Min-Tuo River Basin estimated by the six climate models and the multi-model ensemble average, which intuitively shows the relationship between the predicted precipitation and observed precipitation under the six models and the ensemble average(Fig. 2) Therefore, overall, the future precipitation simulation values after the multi-model ensemble average are closer to the observed precipitation values than those of a single model, and the reliability of the simulation results is higher. Therefore, in this study, the multi-model ensemble average was used as the future precipitation data, and the future soil erosion was predicted based on this.
2.2.5. CA-Markov Model
In order to estimate the soil erosion modulus of the Min-Tuo River Basin under future climate conditions, it is necessary to predict the land use types in order to obtain the P factor values. Because the Cellular Automata (CA) model and Markov model complement each other well, the CA-Markov model is widely used for future land prediction (Abijith & Saravanan, 2021; Aqil et al., 2022; Wang & Wang, 2022). Therefore, the CA-Markov model is used to predict future land use in the study area.
The \(\:\text{K}\text{a}\text{p}\text{p}\text{a}\) coefficient can be used to test the consistency between the simulation results and the actual data, and is commonly used to assess the accuracy of predicting land use changes. The formula for calculating the \(\:\text{K}\text{a}\text{p}\text{p}\text{a}\) coefficient is as follows:
$$\:Kappa=\frac{{p}_{a}-{p}_{c}}{{p}_{p}-{p}_{c}}$$
11
The formula where \(\:{\text{p}}_{\text{a}}\)represents the probability of correct simulation, \(\:{\text{p}}_{\text{c}}\) represents the probability of simulated prediction, and \(\:{\text{p}}_{\text{p}}\) represents the probability of ideal simulation. The \(\:\text{K}\text{a}\text{p}\text{p}\text{a}\) coefficient ranges between [-1,1], and when 0.6≤\(\:\text{K}\text{a}\text{p}\text{p}\text{a}\)≤0.8, it indicates significant consistency; when 0.8<\(\:\text{K}\text{a}\text{p}\text{p}\text{a}\)≤1, it indicates very good consistency.
3.1 Soil Erosion Historical Analysis
Based on the calculated soil erosion impact factors, the soil erosion modulus for the years 2000, 2010, and 2020 was obtained, and the intensity of soil erosion was analyzed. According to the "Classification and Grading Standards for Soil Erosion in China" (SL190-2007) (2008), the soil erosion intensity in the study area was divided into six erosion levels: slight erosion, mild erosion, moderate erosion, severe erosion, very severe erosion, and extreme erosion (Table 5).
| level | micrometric | mild | moderate | strong | extremely strong | violently |
|---|---|---|---|---|---|---|
| Erosion modulus [t/(km²·a)] | <500 | 500 ~ 2500 | 2500 ~ 5000 | 5000 ~ 8000 | 8000 ~ 15000 | >15000 |
As seen from Fig. 3, the erosion in Ya'an, Liangshan Prefecture, Ganzi Prefecture, and the southern part of Aba Prefecture is showing an increasing trend.
According to Table 6, it is known that the soil erosion modulus reached its highest value in 2020, at 34939.8t/(km²·a). The overall trend over the 20 years from 2000 to 2020 first decreased and then increased. The increase from 2010 to 2020 was greater than the decrease from 2000 to 2010, at 7.1% and 2.6% respectively.
| time | Minimum t/(km²·a) | Maximum value [t/(km²·a)] | Range of change (%) |
|---|---|---|---|
| 2000 | 0 | 33493.6 | |
| 2010 | 0 | 32621.4 | -2.6 |
| 2020 | 0 | 34939.8 | 7.1 |
From Table 7, it can be observed that: (1) In 2000, slight erosion and moderate erosion were the main types, covering areas of 112688.41km² and 20007.15km², accounting for 69.72% and 12.38% of the total erosion area, respectively; (2) In 2010, slight erosion and moderate erosion were predominant, covering areas of 97840.53km² and 25076.86km², accounting for 60.54% and 15.52% of the total erosion area, respectively; (3) In 2020, slight erosion and moderate erosion were predominant, covering areas of 54167.59km² and 48477.77km², accounting for 33.52% and 29.99% of the total erosion area, respectively; (4) From 2000 to 2020, the area of slight erosion showed a decreasing trend, reaching its minimum in 2020 at 54167.59km², accounting for 33.52% of the total erosion area. The areas of mild, moderate, and severe erosion showed an increasing trend over the 20 years, reaching their maximum values in 2020 at 28669.73km², 48477.77km², and 24358.86km², accounting for 17.74%, 29.99%, and 15.07% of the total erosion area, respectively. The areas of very severe and extremely severe erosion showed an increasing trend over the 20 years, but with a smaller magnitude, reaching their maximum values in 2020 at 5862.79km² and 84.56km², accounting for 3.63% and 0.05% of the total erosion area, respectively.
| Year | 2000 | 2010 | 2020 | ||||
|---|---|---|---|---|---|---|---|
| Area (proportion) | Area/km² | Ratio/% | Area/km² | Ratio/% | Area/km² | Ratio/% | |
| type | Micro erosion | 112688.41 | 69.72 | 97840.53 | 60.54 | 54167.59 | 33.52 |
| Mild erosion | 17808.95 | 11.02 | 17830.72 | 11.03 | 28669.73 | 17.74 | |
| Moderate erosion | 20007.15 | 12.38 | 25076.86 | 15.52 | 48477.77 | 29.99 | |
| Intense erosion | 8677.3 | 5.37 | 16063.14 | 9.94 | 24358.86 | 15.07 | |
| Extreme erosion | 2403.79 | 1.49 | 4743.96 | 2.94 | 5862.79 | 3.63 | |
| Severe erosion | 35.7 | 0.02 | 66.1 | 0.04 | 84.56 | 0.05 | |
The Sankey diagram (Fig. 4) provides a more intuitive view of the erosion grade transitions.
3.2. Soil Erosion Future Forecast and Analysis
3.2.1. Precipitation Forecast
The precipitation forecast based on the multi-model ensemble average is shown in Fig. 5.
In spatial terms, Wang et al. (2021) used China as the study area, and the results indicated that the spatial estimates of precipitation and extreme precipitation indices from CMIP6 were higher than the actual values. In this study, based on the MME simulations, the future high precipitation area in the Min-Tuo River basin is located in the southwest of the study area. This may be due to the enhanced disturbance of climate in the high-altitude areas of the Min-Tuo River basin under the selected future climate change scenarios, resulting in abnormally high precipitation in these high-altitude regions. The low precipitation area in the future is located in the northern part of the study area, which aligns with the actual spatial distribution of annual precipitation in the study area at present.
3.2.2. Land Use Forecast
With the help of IDRISI software, the Kappa index for simulated land use and actual land use in the study area in 2010 and 2020 was calculated to be 0.87 and 0.86, respectively, indicating good consistency. Considering that the forecast period of this study is from 2030 to 2050, we selected representative years 2030, 2040, and 2050 with a 10 year interval to visually display the evolution of future land use types in the Min-Tuo River basin (Fig. 6). Under future climate change scenarios, the area of cropland conversion increases. However, if agricultural activities are not managed properly, this could potentially increase the risk of soil erosion in the Min-Tuo River basin.
3.2.3. Soil Erosion Time Distribution Simulation
Using the MME method, the future soil erosion values for the Min-Tuo River basin were calculated for three scenarios. Considering the length of the time series, representative years were selected at 10-year intervals to calculate the average soil erosion modulus and total erosion for the Min-Tuo River basin (Table 8). Over time, there are differences in soil erosion in the Min-Tuo River basin under different scenarios. Comparatively, in the SSP5-8.5 scenario, the multi-year average soil erosion modulus and total amount in the Minjiang River basin are ranked as follows: SSP5-8.5 > SSP1-2.6 > SSP2-4.5.
| SSPs | Year | Average soil erosion modulus [t/(km²·a)] | Total erosion(10⁸t) |
|---|---|---|---|
| SSP1-2.6 | 2030 | 442.88 | 7.22 |
| 2040 | 442.86 | 7.22 | |
| 2050 | 445.46 | 7.26 | |
| SSP2-4.5 | 2030 | 437.39 | 7.13 |
| 2040 | 437.99 | 7.14 | |
| 2050 | 440.64 | 7.18 | |
| SSP5-8.5 | 2030 | 446.58 | 7.28 |
| 2040 | 443.36 | 7.23 | |
| 2050 | 447.74 | 7.3 |
In the SSP5-8.5 scenario, the future changes in soil erosion in the Min-Tuo River basin exhibit a "V" shape, meaning that the total soil erosion decreases before 2030, but gradually increases after 2030. In terms of the year when the maximum erosion occurs, in all three scenarios, the maximum soil erosion occurs in 2050. This may be due to the fact that the maximum rainfall also occurs in the 2040s in all three scenarios. Over the future time scale, the increase in heavy rainfall or precipitation variability leads to soil erosion and substantial soil loss.
Figure 7 presents the average soil erosion intensity level in the Min-Tuo River basin during the forecast period under the three different scenarios. Therefore, it is necessary to establish an early warning system for soil erosion forecasting to address the future changes in soil erosion.
3.2.4. Prediction of spatial evolution of soil erosion
Furthermore, the spatial distribution of soil erosion in Min-Tuo River in the future is discussed in three scenarios on the spatial scale (Fig. 8). Under SSP1-2.6 scenario, the annual average soil erosion intensity grade and intensity change of Min-Tuo River in the future are shown in Fig. 8(a). Under SSP1-2.6 scenario, the area of soil erosion in Min-Tuo River reaches 5.35×10⁴km² (above mild erosion), accounting for 32.81% of the total area of the basin. The areas with increased soil erosion intensity are scattered in Ya'an city, Liangsha, Ganzi and the south of Aba in the southwest of the basin. Therefore, under the low emission forcing scenario of SSP1-2.6 in the future, it is necessary to focus on preventing further increase of soil erosion intensity in the corresponding areas mentioned above. As far as the soil erosion grade change under SSP1-2.6 scenario is concerned, the soil erosion grade of 91.76% area of Min-Tuo River will remain stable in the future, 7.13% area will be aggravated, and 1.11% area will be alleviated.
Under SSP2-4.5 scenario, the spatial characteristics of soil erosion intensity in Min-Tuo River are shown in Fig. 8(b). SSP2-4.5 medium emission scenario, the soil erosion grade distribution tends to be consistent with SSP1-2.6 scenario, but in the future, the soil erosion intensity grade is different from SSP1-2.6 scenario, specifically reflected in: 91.89% of the Min-Tuo River soil erosion grade will be stable, 7.00% of the area soil erosion degree will be aggravated, erosion degree will be slightly reduced.
Similarly, under SSP5-8.5 scenario, the spatial characteristics of soil erosion intensity of Min-Tuo River are similar to those of SSP1-2.6 and SSP2-4.5 scenarios, as shown in Fig. 8(c), except that the change of soil erosion grade is different from the other two models. The differences are reflected in: In future SSP5-8.5 high emission scenario, the area of soil erosion grade increase in Min-Tuo River will further increase on the basis of SSP2-4.5 scenario, reaching 12.50%, in which the area of soil erosion intensity increase sharply is the same as that in SSP2-4.5 scenario, and the soil erosion grade in 91.39% area of Min-Tuo River will remain stable in the future, and the soil erosion degree will increase by 0.52% compared with SSP2-4.5 scenario. Therefore, it is necessary to identify potential areas with abrupt increase in soil erosion intensity rate under future high emission scenarios, and to prevent slight soil erosion areas from transforming into severe soil erosion areas.
This study demonstrates the dynamic changes in soil erosion patterns in the Min-Tuo River basin over a 10-year period and forecasts future soil erosion using the recently developed CMIP6 model for approximately 30 years. The selected CMIP6 model is part of the World Climate Research Programme, and the approved climate change simulation and estimation data ensure the scientific credibility of the predictions.
4.1. Soil erosion aggravation in Min-Tuo River in the future
The severity of soil erosion in the Min-Tuo River basin is expected to intensify in the future, leading to increasingly serious issues of soil and water loss. The relationship between future climate and soil erosion is closely intertwined, with climate factors directly or indirectly influencing soil erosion. Among these factors, precipitation has the most significant impact on soil erosion and forms its foundation. This study considers future precipitation conditions, which are projected to increase under three scenarios compared to historical periods, especially with intensified and prolonged precipitation events, including heavy rainfall, which will exacerbate soil and water loss (Yao et al., 2023). Additionally, as future temperatures are expected to rise continuously (Song & Yan, 2022), leading to high-temperature droughts and extreme heatwaves, soil cracking will occur, resulting in serious consequences such as debris flows when rainfall infiltrates. Addressing the increasingly severe soil erosion caused by future high-temperature heavy rainfall is a matter of significant concern and requires attention and resolution.
4.2. Strengthening soil and water conservation measures in farmland
Under future climate scenarios, the increased magnitude of precipitation in the basin will exacerbate soil erosion, and changes in land use may further deteriorate soil erosion in certain areas. Considering the compounded effects of land use change, a significant conversion of grassland to cropland may potentially worsen soil erosion. Therefore, in the process of soil and water erosion control in the future Min-Tuo River basin, attention should be paid to maintaining forest and grassland coverage in the watershed and ensuring stable growth of vegetation cover to mitigate or alleviate the current soil and water loss situation. In the context of increased rainfall in the future, it is advisable to increase the proportion of grassland appropriately, leveraging the role of maintaining surface cover to mitigate soil and water loss while reducing ecological water consumption in the region to safeguard water resources.
Furthermore, to prevent the worsening of soil and water loss, the conversion of land from forests, grasslands, and wetlands to cropland should be avoided (Fang, 2021). It is recommended to consolidate the achievements of returning farmland to forest and grassland projects, continue to maintain stable growth of vegetation cover to mitigate future soil erosion caused by surface cover factors, and strengthen comprehensive management of small watersheds with slope farmland treatment and accompanying measures for slope surface water systems to enhance watershed water conservation capacity in the upstream areas of the basin.
4.3. Uncertainty of simulation results
The simulation results exhibit a certain degree of uncertainty, stemming from both the uncertainties in climate simulation and soil erosion simulation. Firstly, the uncertainty in climate simulation arises from the inability to access high-resolution climate simulation data, leading to reliance on downscaled climate simulation data for the study area. Enhancing the resolution of climate simulation in the future could improve the accuracy of simulations.
Secondly, the uncertainty in soil erosion simulation is intrinsic to the soil erosion models themselves, affecting the precision of future soil erosion simulations. In this study, the classic RUSLE model was employed to calculate the soil erosion modulus in the Min-Tuo River basin. Compared to other methods for assessing soil loss, the RUSLE model tends to overestimate erosion due to its omission of the sedimentation process in soil erosion calculations, focusing solely on soil erosion quantity. Therefore, it is necessary to consider how to adjust the model to better suit the soil erosion conditions in the Min-Tuo River basin.
In estimating the soil erosion modulus, it is essential to select more reasonable and applicable methods for calculating the factors of the RUSLE model based on actual conditions in the study area. Specifically, the factors are determined based on land use type, but the allocation values are somewhat subjective due to the lack of consideration for different soil and water conservation measures for different land use categories. Implementing appropriate land management interventions to reverse the trends in land use/land cover change and soil erosion in the study watershed is crucial. Additionally, precision in the calibration and calculation of soil erosion factors needs further improvement. For instance, in calculating the rainfall erosivity factor, using daily or even sub-daily rainfall data when available can enhance accuracy. Moreover, for the calculation of soil erodibility values, field sampling to determine basic soil properties and subsequent verification can enhance precision. Furthermore, while this study only considered land use type as a human factor, factors such as population density and GDP could also be included to further investigate the impact of socio-economic development on soil erosion in the Min-Tuo River basin, enriching the research content.
Regarding the prediction of future soil erosion in the Min-Tuo River basin, this study only calculated the soil erosion modulus and severity at the annual temporal scale, without elucidating the seasonal or monthly variations in future soil erosion conditions. Thus, the seasonal and monthly variations in soil erosion severity under future scenarios remain unclear. Future research could explore the spatiotemporal distribution characteristics of soil erosion at seasonal and monthly scales.
Based on the prediction of future precipitation and land use types of Min-Tuo River, this study predicts the soil erosion modulus in the future period (2021 ~ 2050) by using the CMIP6 and scenario data published by the World Climate Research Program, and expounds its temporal and spatial evolution characteristics as follows:
The analysis of soil erosion in the study area shows that the soil erosion types from 2000 to 2020 are mainly slight erosion, light erosion and moderate erosion, and the area of light erosion decreases especially from 69.72% in 2000 to 33.52% in 2020. The area of soil erosion reduction is less than that of soil erosion intensification in the study area from 2000 to 2010 and 2010 to 2020. It indicates that soil erosion will be aggravated gradually from 2000 to 2020;
Six CMIP6 models and three different emission scenarios were selected to predict the future precipitation of Min-Tuo River. The prediction results showed that the future precipitation of Min-Tuo River fluctuated and increased under three scenarios, and the fluctuation amplitude and increment of future precipitation of Min-Tuo River were the largest under SSP5-8.5 scenario;
As far as the future land use types are concerned, the top three categories are: cultivated land to forest land, forest land to cultivated land and grassland to cultivated land, with the conversion areas of 5339km², 40874km² and 27393km² respectively. Under the future climate change scenario, the conversion area of cultivated land will increase, but if the agricultural management activities are improper, it is likely to increase the risk of soil erosion in Min-Tuo River;
Compared with the present stage, the area of soil erosion in Min-Tuo River will increase under the future three climate scenarios. Comparatively speaking, the annual average soil erosion modulus and total amount of Min-Tuo River will be the largest under SSP5-8.5 low emission scenario, with the variation between 446.58 ~ 447.74 t/(km²·a) and 7.28 ~ 2.30×10⁸t respectively, followed by SSP1-2.6 scenario, and the soil erosion degree of Min-Tuo River will be the lightest under SSP2-4.5 scenari.
Spatially, high-intensity soil erosion will occur in Ya'an, Liangshan, Ganzi and the south of Aba in the southwest of the basin under three different scenarios in the future The stability of soil erosion grade in SSP2-4.5 scenario is higher than that in SSP1-2.6 scenario and SSP5-8.5 scenario; Comparing the changes of soil erosion grade under three scenarios, we can see that the risk of soil erosion degree will become larger and larger with the emission grade of greenhouse gases in the future.
Credit authorship contribution statement
Nan Jiang: Data curation, Formal analysis, Investigation, Methodology, Resources, Software, Supervision, Validation, Visualization, Writing–original draft. Fuquan Ni: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing–review & editing. Yu Deng: Data curation, Formal analysis, Investigation, Methodology, Resources. Mingyan Wu: Formal analysis, Investigation, Methodology, Resources, Software. Mengyu Zhu: Formal analysis, Investigation, Methodology, Resources, Software. Yuxuan Wang: Data curation, Formal analysis, Investigation, Methodology, Software. Huazhun Ren: Formal analysis, Investigation, Methodology, Resources, Software. Ziying Yue: Data curation, Formal analysis, Investigation, Methodology, Software.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:
Data availability
The data used in this study is as follows: the parameters for the RULSE equation are sourced from Table 1, where the rainfall data is obtained from National Geographic Resource Science SubCenter, National Earth System Science Data Center, National Science & Technology Infrastructure of China, it can be found by the DOI:10.12041/geodata.269728669112188.ver1.db of the data used; Soil data is sourced from the World Soil Database, with the website address http://www.fao.org/soils-portal/so, after entering the URL, select Soil Maps and Databases in the Date Hub to download the relevant data; Terrain data is derived from the global 30 m resolution DEM data set, which is derived from the CGIAR-CSI SRTM Elevation Database(http://srtm.csi.cgiar.org/download); The NDVI utilizes NASA's MODIS satellite data, visit NASA's MODIS data download website (https://modis.gsfc.nasa.gov/data/dataprod/mod13.php) and select the NDVI data required; Land use / cover data can be accessed through http://www.globeland30.org, and data download requires registration and application submission, this data is prepared for estimating the P factor, and the value of the P factor is ultimately determined by empirical methods based on relevant literature of the study area(Peng et al. 2017). The data for CMIP6 is obtained from https://esgf-node.llnl.gov/search/cmip6/, the models and scenarios used are listed in Table 2 and Table 3, respectively.
Funding
The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
- Abijith, D., & Saravanan, S. (2021). Assessment of land use and land cover change detection and prediction using remote sensing and CA Markov in the northern coastal districts of Tamil Nadu, India. Environmental Science and Pollution Research, 29(57), 86055-86067. https://doi.org/10.1007/s11356-021-15782-6
- Aqil, T., Jianguo, Y., & Faisal, M. (2022). Land change modeler and CA-Markov chain analysis for land use land cover change using satellite data of Peshawar, Pakistan. Physics and Chemistry of the Earth 128, 103286. https://doi.org/10.1016/j.pce.2022.103286
- Bekele, T. (2019). Effect of Land Use and Land Cover Changes on Soil Erosion in Ethiopia. Peertechz Publications Private Limited, 20(1), 79-91. https://doi.org/10.1080/15715124.2020.1767636
- Cai, C. F., Ding, S. W., Shi, Z. H., Huang, L., & Zhang, G. Y. (2000). Study of applying USLE and geographical information system IDRISI to predict soil erosion in small watershed. Journal of Soil and Water Conservation, 14, 19–24. doi: 10.13870/j.cnki.stbcxb.2000.02.005
- Chen, J., Brissette, F. O. P., & Leconte, R. (2011). Uncertainty of downscaling method in quantifying the impact of climate change on hydrology. Journal of Hydrology 401(3-4), 190-202, https://doi.org/10.1016/j.jhydrol.2011.02.020
- Deng, B.,& Shi, L. (2022). Remote Sensing Quantitative Research on Soil Erosion in the Upper Reaches of the Minjiang River. Frontiers in Earth Science 10, 930535, https://doi.org/10.3389/feart.2022.930535
- Woodward, D. E. (1999). Method to predict cropland ephemeral gully erosion - ScienceDirect. Catena 37(3-4), 393-399. https://doi.org/10.1016/S0341-8162(99)00028-4
- Fan, L., Lu, C., & Chen Z. (2012). A Review of EPIC Model and Its Applications. Progress in Geography, 31(05), 584-592. doi: 10.11820/dlkxjz.2012.05.006
- Suresh, B. S. (2023). Adaptation in the Scientific Method: An Outline for Mutual Learning and Knowledge Co-production in Climate Science. Science, Technology and Society 28(4), 621-638. https://doi.org/10.1177/09717218231197448
- Guo, B., Tao, H., Liu, B., & Jiang, L. (2012). Characteristics and analysis of soil erosion in Li country after Wenchuan earthquake based on GIS and USLE. Transactions of the Chinese Society of Agricultural Engineering, 28(14), 118-126. doi: 10.3969/j.issn.1002-6819.2012.14.01
- Hairsine, P. B.,& Rose, C. W. (1992). Modeling water erosion due to overland flow using physical principles: 2. Rill flow. Water Resources Research 28(1), 245–250. https://doi.org/10.1029/91WR02381
- Fang, H. (2021). Changes in Cultivated Land Area and Associated Soil and SOC Losses in Northeastern China: The Role of Land Use Policies. International Journal of Environmental Research and Public Health 18(21), 11314. https://doi.org/10.3390/ijerph182111314
- Hessel, R., Jetten, V. G., Liu, B., & Qiu, Y. (2011). Evaluating effects of soil and water management and land use change on the Loess Plateau of China using LISEM. Handbook of Erosion Modelling. Chichester, West Sussex: Wiley-Blackwell, 223-248. https://doi.org/10.1002/9781444328455.ch12
- Humphrey, O. S., Osano, O., Aura, C. M., Marriott, A. L., Dowell, S. M., Blake, W. H., et al. (2022). Evaluating spatio-temporal soil erosion dynamics in the Winam Gulf catchment, Kenya for enhanced decision making in the land-lake interface. Science of the Total Environment, 815, 151975. https://doi.org/10.1016/j.scitotenv.2021.151975
- Hu, J., Ma, J., Nie, C., Xue, L., Zhang, Y., Ni, F., et al. (2020). Attribution Analysis of Runoff change in Min-tuo River Basin based on SWAT model simulations, china. Scientific reports, 10(1), 2900. doi: 10.1038/s41598-020-59659-z
- Zhang, J., Li, H., Zhang, X., Yu, H., Liang, F., Qin, Z., et al. (2020). Sensitivity evaluation of soil erosion based on land use types: A case study of Minjiang River Basin. Journal of Intelligent & Fuzzy Systems, 38(5), 5697-5705. https://doi.org/10.3233/JIFS-179658
- Keno, B., & Suryabhagavan, K. V. (2014). Multitemporal remote sensing of landscape dynamics and pattern change in Dire district, southern Ethiopia. Journal of Geomatics, 8(2), 189-194. doi: 10.4172/2157-7617.1000226
- Liu, J., Ni, F., Deng, Y., Li, L., Wang, Z., Zhou, D., et al. (2019). Risk assessment of soil erosion in the upper reaches of Minjiang Rive. South-to-North Water Transfers and Water Science & Technology, 17(01), 105-112. doi: 10.13476/j.cnki.nsbdqk.2019.0015.
- Lu, K., Arshad, M., Ma, X., Ullah, I., Wang, J., & Shao, W. (2022). Evaluating observed and future spatiotemporal changes in precipitation and temperature across China based on CMIP6‐GCMs. International Journal of Climatology, 42(15), 7703-7729. https://doi.org/10.1002/joc.7673
- Ministry of Water Resources of the People's Republic of China. (2008). SL190-2007, classification and grading standards for soil erosion. Beijing: China Water & Power Press.
- Ministry of Water Resources of the People's Republic of China. (2018). SL773-2018, guidance on estimation of soil loss in production and construction projects. Beijing: China Water & Power Press.
- Mirzabaev, A., Goedecke, J., Dubovyk, O., Djanibekov, U., Le, Q. B., & Aw-Hassan, A. (2016). Economics of land degradation in Central Asia. Economics of land degradation and improvement–A global assessment for sustainable development, 261-290. doi: 10.1007/978-3-319-19168-3_10
- Nearing, M. A., Simanton, J. R., Norton, L. D., Bulygin, S. J., & Stone, J. (1999). Soil erosion by surface water flow on a stony, semiarid hillslope. Earth Surface Processes and Landforms: The Journal of the British Geomorphological Research Group, 24(8), 677-686. https://doi.org/10.1002/(SICI)1096-9837(199908)24:8<677::AID-ESP981>3.0.CO;2-1
- Peng, J., Li, D. D., & Zhang, Y. Q. (2007). Analysis of spatial characteristics of soil erosion in mountain areas of northwestern Yunnan based on GIS and RUSLE. Journal of Mountain Science, 25(5), 548-556. ISSN:1008-2186/CN:51-1516
- Peng, S., Ding, Y., Wen, Z., Chen, Y., Cao, Y., & Ren, J. (2017). Spatiotemporal change and trend analysis of potential evapotranspiration over the Loess Plateau of China during 2011–2100. Agricultural and forest meteorology, 233, 183-194. https://doi.org/10.1016/j.agrformet.2016.11.129
- Wang, Q., & Wang, H. (2022). An integrated approach of logistic-MCE-CA-Markov to predict the land use structure and their micro-spatial characteristics analysis in Wuhan metropolitan area, Central China. Environmental Science and Pollution Research, 1-24. doi: 10.1007/s11356-021-17750-6
- Rao, L., Xu, Y., Hu, J., Wang, Z., Yao, W., & Shen, Z. (2020). Study on Soil Erodibility Factoe K on Soil Cover Area of Piaha Sandstone Region. Journal of Basic Science and Engineering, 28(04), 763-773. doi: 10.16058/j.issn.1005-0930.2020.04.002
- Renard, K. G., & Ferreira, V. A. (1993). RUSLE model description and database sensitivity. Journal of environmental quality, 22(3), 458-466. https://doi.org/10.2134/jeq1993.00472425002200030009x
- Renard, K. G., Yoder, D. C., Lightle, D. T., & Dabney, S. M. (2011). Univeral soil loss equation and revised universal soil loss equation. Handbook of Erosion Modelling, edited by: Morgan R PC and Nearing, MA, 137-67. https://doi.org/10.1002/9781444328455.ch8
- De Roo, A. P. J. (1996). The LISEM project: an introduction. Hydrological Processes, 10(8), 1021-1025. https://doi.org/10.1002/(SICI)1099-1085(199608)10:8<1021::AID-HYP407>3.0.CO;2-I
- Rymszewicz, A., Mockler, E., O'Sullivan, J., Bruen, M., Turner, J., Conroy, E., et al. (2015). Assessing the applicability of the Revised Universal Soil Loss Equation (RUSLE) to Irish Catchments. Proceedings of the International Association of Hydrological Sciences, 367, 99-105. https://doi.org/10.5194/piahs-367-99-2015
- Piao, S., Zhang, X., Chen, A., Liu, Q., Lian, X., Wang, X., et al. (2019). The impacts of climate extremes on the terrestrial carbon cycle: A review. Science China Earth Sciences, 62, 1551-1563. doi: 10.1007/s11430-018-9363-5
- Ma, S., Wang, L. J., Wang, H. Y., Zhao, Y. G., & Jiang, J. (2023). Impacts of land use/land cover and soil property changes on soil erosion in the black soil region, China. Journal of Environmental Management, 328, 117024. https://doi.org/10.1016/j.jenvman.2022.117024
- Song, S., & Yan, X. (2022). Evaluation of events of extreme temperature change between neighboring days in CMIP6 models over China. Theoretical and Applied Climatology, 150(1-2), 53-72. doi: 10.21203/rs.3.rs-601223/v1
- Su, Y., Wu, X., Guo, C., Chen, X., Liu, L., & Peng, S. (2022). The spatiotemporal pattern and trend of annual precipitation over China from 2001 to 2100. Journal of Lanzhou University (Natural Sciences), 58(05), 641-649. doi: 10.13885/j.issn.0455-2059.2022.05.009
- Wang, Y., Li, H., Wang, H., Sun, B., & Chen, H. (2021). Evaluation of CMIP6 model simulations of extreme precipitation in China and comparison with CMIP5. Acta Meteorologica Sinica, 79(03), 369-386. doi: 10.11676/qxxb2021.031
- Wang, Z., & Su, Y. (2020). Assessment of soil erosion in the Qinba mountains of the Southern Shaanxi Province in China using the RUSLE model. Sustainability, 12(5), 1733. https://doi.org/10.3390/su12051733
- Williams, J. R. (1990). The erosion-productivity impact calculator (EPIC) model: a case history. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences, 329(1255), 421-428. https://doi.org/10.1098/rstb.1990.0184
- Wischmeier, W. H., & Smith, D. D. (1958). Rainfall energy and its relationship to soil loss. Eos, Transactions American Geophysical Union, 39(2), 285-291. doi: 10.1029/TR039I002P00285
- Wischmeier, W. H., & Smith, D. D. (1978). Predicting rainfall erosion losses: a guide to conservation planning (No. 537). Department of Agriculture, Science and Education Administration. http://topsoil.nserl.purdue.edu/usle/AH_537.pdf
- Wu, J., Xia, J., Zeng, S., Liu, X., & Fan, D. (2023). Evaluation of the Performance of CMIP6 Models andFuture Changes Over the Yangtze River Basin. Resources and Environment in the Yangtze Basin, 32(01), 137-150, doi:10.11870/cjlyzyyhj202301013
- Yang, J., & Huang, X. (2021). The 30 m annual land cover dataset and its dynamics in China from 1990 to 2019. Earth System Science Data, 13(8), 3907-3925, https://doi.org/10.5194/essd-13-3907-2021
- Yang J., Dong J., Xiao X., Dai J. Wu C., Xia J., et al. (2019). Divergent shifts in peak photosynthesis timing of temperate and alpine grasslands in China. Remote Sensing of Environment, 233, 111395, https://doi.org/10.1016/j.rse.2019.111395
- Wei, Y., Sun, S., Liang, D., & Jia, Z. (2022). Spatial–temporal variations of NDVI and its response to climate in China from 2001 to 2020. International Journal of Digital Earth, 15(1), 1463-1484, https://doi.org/10.1080/17538947.2022.2116118
- Yao, Y., Li, J., Jiang, Y., & Huang, G. (2023). Evaluating the response and adaptation of urban stormwater systems to changed rainfall with the CMIP6 projections. Journal of Environmental Management, 347, 119135, https://doi.org/10.1016/j.jenvman.2023.119135
- Zhong, X., Zhang, S., Wu, R., Jing, Y., Men, L., & Zhou, T. (2022). Analysis of Dynamic Changes and Driving Forces of Soil Erosion in Tuojiang River Basin. Research of Soil and Water Conservation, 29(02), 43-49+56, doi:10.13869/j.cnki.rswc.2022.02.003