Spatial and Temporal Variability of Soil Erosion in Northeast China from 2000 to 2020

Abstract

Northeast China is a prominent base for commercial grain production nationwide. Soil erosion, a primary cause of land degradation and grain yield decrease, has become an imminent issue and is still not well documented in Northeast China. Thus, a comprehensive assessment of soil erosion in Northeast China is essential for deepening our understanding of various geological and agricultural issues, such as control of regional water and soil losses, anti-degeneration attempts on black soil, preservation of land fertility, and safeguarding of national food security. Based on the Revised Universal Soil Loss Equation (RUSLE) and the Revised Wind Erosion Equation (RWEQ) models, this paper comprehensively assessed the water and wind erosion in Northeast China from 2000 to 2020 and analyzed the current situation, as well as the spatial and temporal evolution of soil erosion. The results suggest the following: (1) The degree of soil erosion in Northeast China was mainly slight, and water erosion was more severe than wind erosion. Water and wind erosion in bare land and grassland were more intensive than in cropland and forests. The Liao River Plain (LRP) has undergone relatively intensive water erosion, while the wind erosion in the Greater Kinggan Mountains Region (GKMR) was more intensive than in other sub-regions. (2) A slight intensifying trend of water erosion could be observed in Northeast China from 2000 to 2020, where the area of slight water erosion decreased and that of light and intensive water erosion increased. The water erosion in the Changbai Mountain Region (CBMR), the Sanjiang Plain (SJP), and the Songnen Plain (SNP) intensified, while the LRP has undergone slower water erosion than before. The water erosion in bare land and cropland intensified, while the water erosion in grassland and forests slowed down. Compared to the first decade (2000–2010), the second decade (2010–2020) in the timespan of study had a reversed trend of water erosion from intensifying to moderating, which means water erosion was alleviated. (3) A moderating trend in wind erosion could be found in Northeast China from 2000 to 2020, where the area of slight wind erosion increased and that of light, moderate, and intensive wind erosion decreased. The wind erosion in the LRP showed a pronounced decrease, and the wind erosion in bare land and cropland also considerably decreased. Compared to the first decade, the amount of wind erosion in the second decade decreased by 18.2%, but the rate in the second decade decreased slowly or even increased. These two facts indicate that wind erosion in Northeast China has alleviated, but this trend is gradually slowing down. Soil erosion is caused by multiple factors, such as climate, topography, soil, and human activities. This study provides important implications for our understanding of soil erosion control and management in Northeast China. In sub-regions with severe erosion, such as the LRP and the GKMR, we can adopt methods such as zero tillage, cross ridge tillage, and straw mulching according to the local characteristics of soil erosion to slow down the process.

1. Introduction

Soil erosion is one of the essential processes in physical geography in which soil and its parent material undergo destruction, denudation, transportation, and deposition from natural or anthropogenic erosive forces produced by rain, runoff, wind, gravity, tillage, and other sources [1,2,3]. Soil erosion is a severe global environmental issue that leads to land degradation, soil nutrient loss, non-point source pollution, and increased risk of floods and droughts [4,5,6,7,8]. In particular, soil erosion degrades soil fertility and eventually renders the land unproductive, thus presenting a grave challenge to the sustainable development of agriculture [9,10,11].

Because of its flat terrain and rich black soil, Northeast China has grown into a significant production base for commodity grain and a prominent area for crop production nationwide, contributing to the annual food demand of more than 216 million urban citizens [12,13]. However, soil erosion in the region has been deteriorating the land in recent years, and the disconnection and devouring of farmland caused by the process will eventually threaten food security [14,15,16,17]. For the black soil region in Northeast China, high-intensity continuous cropping has impaired the soil structure, which has then caused multiple issues such as reduced black soil thickness and organic matter content, expansion of erosion area, decreased productivity, and a worsened ecological environment [11]. Severe soil erosion has occurred since the commencement of large-scale cultivation in the 1950s and the average black soil thickness has decreased by 40 cm since then [18,19,20,21]. Previous studies have proposed that the average rate of erosion in this region has reached approximately 15 Mg·ha⁻¹·yr⁻¹ or more [22,23] and that the erosion will consume the black soil within 113 years [24]. Therefore, it is essential to assess water and wind erosion in Northeast China to ensure the sustainable development of agriculture there.

In recent years, the severe soil erosion in Northeast China has caused increasing concern. Several studies have been carried out, including the characterization of hillslope erosion [20,25,26] and gully erosion [12,15,19,27,28,29,30], the impact of soil erosion on SOC, soil productivity and crop production [11,31,32,33,34,35,36], the identification of sediment sources [9,14,22,37,38], factors influencing erosion rates [15,24], gully erosion mapping and monitoring [39,40,41], and the controlling strategies against erosion [13,42,43,44,45]. Most existing studies focus on assessing water erosion in local areas or small watersheds and try to build relationships between the soil erosion rates and various influencing factors. However, there are few studies on water and wind erosion and the spatial and temporal evolution of the process on a large scale in Northeast China.

According to the dominant exogenic process during erosion, researchers divide soil erosion into two types: water erosion and wind erosion [46]. An essay in 2010 suggested that the area affected by severe soil erosion in Northeast China was 275,900 km², which comprised 27.09% of the total territory. Among the eroded lands, the areas affected by water erosion and wind erosion were 180,000 and 33,600 km², respectively [47]. The study indicated that water erosion was the main erosion form in the region and has caused the most soil loss there. Previous studies have revealed that the area affected by water erosion is the largest in Northeast China, and severe erosion at intensive or higher levels is often associated with wind erosion [48]. Research on wind erosion nationwide started only recently and currently, authors mainly discuss the issue in arid and semi-arid regions. The wind erosion conditions in Northeast China draw relatively less attention. Northeast China is susceptible to frequent wind hazards, and the loose structure of black soil aggravates wind erosion. Therefore, it is also imperative to study the influencing factors and quantitative relationship behind the wind erosion on the cropland there [49].

Researchers have applied various methods and instruments to estimate soil erosion, such as laboratory simulations [20,26,50,51,52], ¹³⁷Cs tracers [14,24,37,53], remote sensing interpretation [17,29,39], and Universal Soil Loss Equation/Revised Universal Soil Loss Equation (USLE/RUSLE) models [4,54,55]. The inherent shortcomings of methods such as ¹³⁷Cs tracers and remote sensing interpretation, such as uncertainty and subjectivity, may affect assessment accuracy. The RUSLE integrates various natural and artificial factors, including rainfall erosivity, soil erodibility, topography, vegetation coverage and management, soil conservation protection, and fits for assessing water erosion at a regional scale [4,56,57]. Various models have also been used in simulating wind erosion, with the Revised Wind Erosion Equation (RWEQ) model being one of the most commonly used models [1,58,59]. The RWEQ model reduces the complexity of wind erosion, but it still covers a wide range of factors such as the effects of wind speed, precipitation, temperature, soil texture, topography, and vegetation cover on wind erosion [60,61,62]. Many scholars have confirmed the accuracy and applicability of the RWEQ model in simulating wind erosion on different terrains, such as plains, valleys, farmland, and grasslands [63]. The RWEQ model has also yielded results in some areas in China at large scales, including the Beijing–Tianjin Sandstorm Source Region [64], the watershed of the Ningxia–Inner Mongolia reaches of the Yellow River [65], the Three-River Headwaters Region [66], Hunshandake Desert [67], and Qinghai Province [68].

In summary, what is the overall macro-scale degree of change in soil erosion in Northeast China over the past 20 years? Which areas and ecosystems are experiencing severer soil erosion? These are urgent questions that we need to address. Along these lines, we try to investigate the changes in water and wind erosion based on the RUSLE and the RWEQ method. These models were selected because they share many common advantages, such as simple modeling processes, fewer input parameters, are suitable for large spatio-temporal scales, and have easy integration with the GIS database [3]. For the models above, we localized all parameters to improve the accuracy of results. The specific objectives of this study are: (1) to quantitatively evaluate the water and wind erosion in Northeast China from 2000 to 2020; (2) to analyze the current situation and the spatial and temporal evolutionary characteristics of water and wind erosion, and reveal the changing patterns of soil erosion among different areas and types of ecosystems in Northeast China from 2000 to 2020; (3) to briefly analyze the factors behind soil erosion and preventive measures against the process, intending to provide a scientific basis for soil erosion control and management in Northeast China.

2. Materials and Methods

2.1. Study Area

Northeast China refers to the region that is located between 115°32′ E to 135°09′ E and 38°42′ N to 53°35′ N. The region encompasses the Heilongjiang (HLJ), Jilin (JL), and Liaoning (LN) provinces, as well as the eastern portion of the Inner Mongolia Autonomous Region (IM) (Figure 1). The land area of Northeast China is approximately 1.25 × 10⁶ km², accounting for 13% of China’s total area. Most of the study area is characterized by a temperate continental monsoon climate. Over the past 30 years, the average annual air temperature in Northeast China has been 5.68 °C and the average annual precipitation has been 615 mm, with significant variation across the region [48]. This study divided Northeast China into six geographic sub-regions based on the terrain, vegetation, and climate, namely the Greater Kinggan Mountains Region (GKMR), the Lesser Kinggan Mountains Region (LKMR), the Changbai Mountain Region (CBMR), the Sanjiang Plain (SJP), the Songnen Plain (SNP), and the Liao River Plain (LRP). The GKMR and the LKMR are located in the western and the northern parts of the study area, while the CBMR is in the southeastern part of it. The three plains (SJP, SNP, and LRP) form the central part of the study area from north to south [4,69]. Northeast China is a prominent base for commercial grain production nationwide because of its fertile black soil, but high-intensity reclamation has caused severe erosion and degradation of the black soil, as well as severe water and soil losses on slope cropland. The area affected by water and soil losses in the region accounts for about one-fourth of the total land, and the annual yield decrease caused by soil erosion alone can reach 14.7% [70]. Meanwhile, the region also boasts the most extensive natural forests in China, which are indispensable in water conservation, soil conservation, and sand fixation. The forests in Northeast China mainly grow in the GKMR, LKMR, and CBMR. Their forests account for 37% of China’s total forest area, with a timber volume of 3.2 billion cubic meters, equaling one-third of the total timber volume nationwide. Zhang et al. [71] found that although the multi-year soil erosion in Northeast China has generally been moderating, the gully erosion has been increasing due to human activities. Furthermore, soil erosion in this area may worsen with future climate change.

2.2. Data

The data adopted in the study include land use/cover change (LUCC) data, normalized difference vegetation index (NDVI) data, meteorological observation data, digital elevation model (DEM) data, and soil data.

LUCC datasets in 2000, 2010, and 2020 were obtained from the Resource and Environment Science and Data Center (https://www.resdc.cn (accessed on 1 August 2022)) and were generated from Landsat TM/ETM+ remote sensing imagery by manual visual interpretation with a spatial resolution of 30 m. The spatial distribution data of the types of ecosystems were obtained by reclassifying the data in combination with the LUCC datasets, with comprehensive classification accuracies of more than 94%. The ecosystems can be divided into six types, namely cropland, forest, grassland, wetland, urban, and bare land [72]. As soil erosion occurs mainly on cropland, forest, grassland, and bare land, we focused our study on the four abovementioned types of ecosystems in this paper.

The MODIS NDVI data (MOD13Q1), with a spatial resolution of 1 km and a temporal resolution of 16 days from 2000 to 2020, were applied to calculate vegetation coverage. The NDVI datasets were processed via format conversion, calibration, projection transformation, view geometry, and radiation and image mosaics with the MODIS Reprojection Tool (MRT) and the Savitzky–Golay filter. According to the theory of pixel dichotomy modeling, a pixel’s NDVI value can be considered a combination of information contributed by pure vegetation and bare land. The 16-day maximum synthesis vegetation coverage datasets for the timespan 2000–2020 were obtained by the following equation:

$$f=\frac{NDVI-NDV{I}_{soil}}{NDV{I}_{veg}-NDV{I}_{soil}}$$

(1)

where $f$ is vegetation coverage, $NDV{I}_{veg}$ is the NDVI value of a pure vegetation pixel, and $NDV{I}_{soil}$ is the NDVI value of lands without vegetation (bare). The pixels were identified based on their types of ecosystems.

The meteorological observation data from national stations, including precipitation, temperature, relative humidity, sunshine hours, and wind speed, were downloaded from the China Meteorological Data Service Center (CMDSC) (http://data.cma.cn (accessed on 1 August 2022)). Then, the data underwent interpolation via the ANUSPLIN method for the timespan 2000–2020. The interpolation of precipitation values was controlled and calibrated using 0.25° daily gridded precipitation data from the National Meteorological Information Centre.

The DEM data we applied were the 90 m SRTM (SRTM3 V4.1) (Shuttle Radar Topography Mission) images (http://srtm.csi.cgiar.org (accessed on 1 August 2022)), which we introduced to calculate slope steepness and slope length factors.

The soil data in the study were from the 1:1 million Chinese Soil Database of the Resource and Environment Science and Data Center (https://www.resdc.cn (accessed on 1 August 2022)). They are a spatial vector data including soil type, soil particle content, organic matter content, and other attributes.

2.3. Methods

2.3.1. Water Erosion Assessment

The calculation of water erosion intensity was based on the RUSLE model. The corresponding formulas are listed as follows:

$$SE=R\times K\times L\times S\times C\times P$$

(2)

$$K=\left[2.1\times {10}^{-4}\left(12-OM\right)M^{1.14}+3.25\left(St-2\right)+2.5\left(P-3\right)\right]/100\times 0.1317$$

(3)

$$L={\left(\frac{\lambda }{22.13}\right)}^{\frac{\beta }{1+\beta }}$$

(4)

$$\beta =\left(\frac{sin\theta }{0.0896}\right)/\left(3.0sin{\theta }^{0.8}+0.56\right)$$

(5)

$$S=\{\begin{array}{cc}10.8\sin \left(s\right)+0.03& \theta <9\%\\ 16.8\sin \left(s\right)-0.5& 9\%\le \theta \le 18\%\\ 21.91\sin \left(s\right)-0.96& \theta >18\%\end{array}$$

(6)

$$C=\{\begin{array}{l}1\hfill \\ 0.6508-0.3436lgf\hfill \\ 0\hfill \end{array}\begin{array}{r}\hfill f=0\\ \hfill 0<f\le 78.3\%\\ \hfill f>78.3\%\end{array}$$

(7)

where $SE$ is the annual soil erosion modulus (t·hm⁻²). $R$ is the rainfall erodibility factor (MJ·mm/(hm²·h·a)), which was calculated using a daily rainfall fitting model [73]. $K$ is the soil erodibility factor (t·h/(MJ·mm)), which was estimated by the Nomo model [74] based on the 1:1 million soil data. $L$ (m) is the slope length factor and $S$ (%) is the slope steepness, which were calculated by the core algorithm of McCool et al. [75] and Liu et al. [76] and improved by considering runoff barriers of land surface parameters. $C$ is the vegetation cover factor calculated based on $f$ and time intervals for both rainfall erodibility and fractional coverage factors were determined as 16 days to avoid time asynchronism and to reduce errors [77]. $P$ is the erosion control practice factor with a value range between 0 and 1, determined primarily from the relevant literature and expert consultations. p values in grassland, forest, and bare land were set at 1, paddy fields at 0.15, dry land at 0.352, settlements at 0.01, and wetland/water bodies at 0 [78,79].

2.3.2. Wind Erosion Assessment

The RWEQ model was used to assess the wind erosion in Northeast China from 2000 to 2020, considering multiple factors such as climate, soil erodibility, soil crust, surface roughness, and vegetation. Wind erosion intensity was thus calculated using the RWEQ model. The relevant equations are as follows:

$$SL=Q_x/x$$

(8)

$$Q_x=Q_{max}\left[1-e^{{\left(\frac{x}{s}\right)}^2}\right]$$

(9)

$$Q_{max}=109.8\left(WF\times EF\times SCF\times K\prime \times COG\right)$$

(10)

$$s=105.71{\left(WF\times EF\times SCF\times K\prime \times COG\right)}^{-0.3711}$$

(11)

$$WF=\frac{{{\displaystyle \sum }}_{i=1}^{N}W{S}_2{\left(W{S}_2-W{S}_t\right)}^2\times N_d\rho }{N\times g}\times SW\times SD$$

(12)

$$EF=\frac{29.09+0.31S_a+0.17S_i+0.33\frac{S_a}{C_l}-2.59OM-0.95CaC{O}_3}{100}$$

(13)

$$SCF=\frac{1}{1+0.0066{\left(C_l\right)}^2+0.021{\left(OM\right)}^2}$$

(14)

where $SL$ is the annual wind erosion modulus (t·hm⁻²), $x$ is the field length (m), $Q_x$ is the sand flux at the length $x$ of the field (kg/m). $Q_{max}$ is the maximum sand transport capacity under the wind force (kg/m), $s$ is the critical field length (m). $WF$ is the weather factor (kg/m), which reflects the combined effect of the wind factor, the air density, the acceleration due to gravity, the soil wetness factor, and the snow cover factor on wind erosion. The wind and soil wetness factors used the station’s daily data from the CMDSC, and the snow cover factor came from the long-term snow depth dataset. $EF$ is the soil erodibility factor, and $SCF$ is the soil crust factor. $K\prime$ is the soil roughness factor measured using a roller chain method. Here, the measured roughness factors on sand and grassland were 0.96 and 0.69. $COG$ is the combined vegetation factor that includes three sub-factors (flat residues, standing residues, and growing vegetation). It is calculated by the fractional vegetation cover under the effect of either withered or growing vegetation on wind erosion via the combination of field observation and visual estimation, as well as traditional photographic recording [62,78,80].

2.3.3. Classification of Soil Erosion

In this paper, we determine the types and severity of soil erosion according to the Standards for Classification and Gradation of Soil Erosion (SL 190-2007) [81]. The original standards for classification and gradation of soil erosion were six grades, including slight, light, moderate, intensive, very intensive, and severe. In this paper, intensive, very intensive, and severe are combined into intensive to better show the changing characteristics of soil erosion in Northeast China.

Table 1. Classes of soil erosion intensity.

Erosion Degree	Average Amount of Erosion [t/(hm−2·a−1)]
1: Slight	<5
2: Light	5–25
3: Moderate	25–50
4: Intensive	>50

lists the different degrees corresponding to the average amount of erosion. Because the definition of slight erosion refers to no apparent soil loss, we mainly discuss soil erosion at Light and higher levels below [6,48].

2.3.4. Analyses of Change Trends

The linear slopes of analytical factors over time can reflect changing trends and geographical differences more logically and precisely [62,82]. We used a least-squares linear regression model to detect variability in soil erosion during the research period (2000–2020) and to fit soil erosion variables as a function of time (year). The slope of the linear time trend was calculated for a certain timespan and for each pixel by ArcGIS. The formula is as follows:

$$Slope=\frac{n\times {{\displaystyle \sum }}_{i=1}^{n}i\times X_i-\left({{\displaystyle \sum }}_{i=1}^{n}i\right)\left({{\displaystyle \sum }}_{i=1}^n{X}_i\right)}{n\times {{\displaystyle \sum }}_{i=1}^n{i}^2-{\left({{\displaystyle \sum }}_{i=1}^{n}i\right)}^2}$$

(15)

where $i$ is the sequence indicator for each year, starting from 1 (for the Year 2000); $n$ is the total number of years included, e.g., 21; $X_i$ is the value of analytical factors (the annual value of soil erosion modulus) for the Year $i$. Negative slope values represent a decreasing trend, and positive values indicate an increasing trend.

2.3.5. Spatial Statistics Analysis

Based on the abovementioned assessment of soil erosion and comprehensive consideration of geological sub-regions and ecosystems, we calculated the erosion modulus in different timespans (2000–2010 and 2010–2020), the area proportion for different erosion degrees, and the interannual variation rate of soil erosion via spatial analysis in ArcGIS. The results reveal the features of the spatial difference and the temporal evolution of soil erosion in Northeast China.

3. Results

3.1. Current Situation of Soil Erosion

From 2000 to 2020, the multi-year mean water erosion modulus in Northeast China was 1.52 t·hm⁻² and the multi-year mean water erosion amount was 1.88 × 10⁸ t (

Table 2. Water and wind erosion in Northeast China from 2000 to 2020, displayed in different timespans.

Types of Soil Erosion			2000–2010			2010–2020			2000–2020
			Mean Erosion Modulus	Amount of Erosion	Interannual Variation	Mean Erosion Modulus	Amount of Erosion	Interannual Variation	Mean Erosion Modulus	Amount of Erosion	Interannual Variation
			(t·hm−2)	(108 t)	(t·hm−2·a−1)	(t·hm−2)	(108 t)	(t·hm−2·a−1)	(t·hm−2)	(108 t)	(t·hm−2·a−1)
Water erosion	Geographic sub-regions	SNP	0.469	0.105	0.025	0.587	0.131	0.000	0.517	0.116	0.012
		SJP	0.619	0.066	0.046	0.796	0.085	−0.016	0.692	0.074	0.017
		LRP	3.718	0.689	0.021	2.858	0.530	−0.055	3.244	0.601	−0.066
		LKMR	0.384	0.034	0.020	0.392	0.035	−0.017	0.373	0.033	0.001
		GKMR	1.253	0.565	0.016	1.345	0.606	−0.032	1.280	0.577	0.004
		CBMR	2.919	0.520	0.442	2.700	0.481	−0.066	2.661	0.474	0.038
	Types of ecosystems	Cropland	1.178	0.437	0.049	1.146	0.425	−0.018	1.140	0.423	0.002
		Forest	1.843	0.919	0.165	1.584	0.789	−0.047	1.649	0.822	−0.002
		Grassland	1.859	0.432	−0.006	1.874	0.435	−0.052	1.832	0.426	−0.007
		Bare land	1.769	0.049	−0.036	2.021	0.056	−0.018	1.890	0.052	0.010
	Northeast China		1.604	1.983	0.083	1.514	1.872	−0.032	1.520	1.878	0.001
Wind erosion	Geographic sub-regions	SNP	1.286	0.288	−0.090	1.077	0.241	0.038	1.154	0.259	−0.024
		SJP	0.700	0.075	−0.016	0.539	0.058	0.009	0.603	0.065	−0.012
		LRP	2.113	0.392	−0.147	1.197	0.222	−0.045	1.591	0.295	−0.089
		LKMR	0.262	0.023	−0.024	0.231	0.021	0.006	0.239	0.021	−0.005
		GKMR	1.572	0.709	−0.067	1.512	0.682	0.067	1.513	0.682	−0.006
		CBMR	0.437	0.078	−0.006	0.315	0.056	−0.007	0.367	0.066	−0.009
	Types of ecosystems	Cropland	1.146	0.425	−0.070	0.843	0.312	0.011	0.964	0.357	−0.030
		Forest	0.431	0.215	−0.020	0.325	0.162	−0.002	0.369	0.184	−0.010
		Grassland	2.446	0.568	−0.108	2.251	0.523	0.091	2.302	0.535	−0.019
		Bare land	4.808	0.132	−0.338	3.448	0.095	0.070	3.985	0.110	−0.133
	Northeast China		1.270	1.571	−0.067	1.039	1.285	0.025	1.126	1.393	−0.023

). The degree of water erosion in Northeast China was mainly slight and its area was about 1,030,166.8 km², accounting for 83.3% of the total study area. Light erosion was the second most prevalent degree of water erosion, affecting an area of 200,228.4 km², or about 16.2% of the total area. The moderate and intensive water erosion areas were 4574.8 km² and 983.2 km², respectively, accounting for 0.4% and 0.1% (

Table 3. Area percentage (%) of different degrees of water and wind erosion in Northeast China, displayed in different timespans.

Types of Erosion			2000–2010				2010–2020				2000–2020
			Slight	Light	Moderate	Intensive	Slight	Light	Moderate	Intensive	Slight	Light	Moderate	Intensive
Water erosion	Geographic sub-regions	SNP	97.004	2.978	0.013	0.005	95.357	4.596	0.035	0.011	96.371	3.599	0.023	0.007
		SJP	93.212	6.760	0.018	0.010	90.728	9.203	0.050	0.019	92.229	7.726	0.031	0.014
		LRP	60.993	37.441	1.337	0.228	65.390	33.728	0.693	0.189	62.893	36.023	0.892	0.192
		LKMR	98.100	1.880	0.013	0.007	97.998	1.976	0.017	0.009	98.282	1.695	0.015	0.008
		GKMR	86.255	13.350	0.330	0.065	85.285	14.240	0.381	0.094	85.949	13.638	0.338	0.076
		CBMR	68.825	29.888	1.108	0.179	66.907	32.241	0.706	0.145	68.981	30.152	0.728	0.139
	Types of ecosystems	Cropland	87.312	12.382	0.266	0.040	86.914	12.871	0.184	0.031	87.284	12.480	0.205	0.031
		Forest	80.363	19.062	0.524	0.051	80.350	19.320	0.288	0.042	80.832	18.812	0.318	0.038
		Grassland	80.883	18.227	0.759	0.131	79.995	19.185	0.674	0.145	80.550	18.655	0.669	0.126
		Bare land	84.624	14.381	0.729	0.267	82.160	16.669	0.793	0.378	83.381	15.552	0.745	0.323
	Northeast China		83.342	16.087	0.486	0.086	82.847	16.709	0.357	0.088	83.350	16.200	0.370	0.080
Wind erosion	Geographic sub-regions	SNP	83.992	15.971	0.033	0.003	89.304	10.678	0.017	0.002	87.186	12.797	0.015	0.001
		SJP	97.268	2.732	0.000	0.000	97.569	2.431	0.000	0.000	97.664	2.336	0.000	0.000
		LRP	67.673	32.113	0.204	0.010	87.628	12.311	0.060	0.001	80.207	19.646	0.142	0.005
		LKMR	99.614	0.386	0.000	0.000	99.540	0.460	0.000	0.000	99.596	0.404	0.000	0.000
		GKMR	76.944	22.975	0.074	0.006	81.044	18.877	0.070	0.009	79.491	20.443	0.059	0.007
		CBMR	98.239	1.759	0.001	0.000	98.836	1.163	0.000	0.000	98.571	1.428	0.001	0.000
	Types of ecosystems	Cropland	86.723	13.271	0.006	0.001	95.525	4.472	0.003	0.000	92.660	7.336	0.003	0.000
		Forest	97.266	2.726	0.007	0.001	98.543	1.444	0.012	0.001	98.104	1.886	0.008	0.001
		Grassland	59.166	40.811	0.020	0.002	67.029	32.937	0.031	0.003	63.738	36.240	0.020	0.002
		Bare land	37.319	61.834	0.798	0.049	54.558	44.925	0.435	0.082	46.596	52.870	0.469	0.065
	Northeast China		83.247	16.684	0.064	0.004	88.807	11.151	0.038	0.004	86.715	13.235	0.046	0.004

). Those areas could be found in Chifeng City (IM) and Chaoyang and Huludao in western LN (Figure 2).

The condition of water erosion in each sub-region indicates that the LRP underwent the most severe water erosion, with a multi-year mean erosion modulus of 3.24 t·hm⁻², followed by the CBMR and the GKMR, with a water erosion modulus of 2.66 and 1.28 t·hm⁻², respectively. Different ecosystems undergo different degrees of water erosion, namely bare land, grassland, forests, and cropland, from the most severe to the least affected (sorted by the water erosion modulus from large to small). The water erosion modulus in cropland was only 1.14 t·hm⁻², far lower than that in other ecosystems (Table 2).

Northeast China’s multi-year mean wind erosion modulus was 1.13 t·hm⁻² and the wind erosion amount was 1.4 × 10⁸ t (Table 2). The degree of wind erosion in Northeast China was also slight. Its area was about 1,072,463.6 km², accounting for 86.7% of the total study area. Light erosion was the second most prevalent degree of wind erosion, affecting an area of 163,686.1 km², or about 13.2% of the total area. The moderate and intensive wind erosion mainly occur in IM, and their total area was 615.6 km², accounting for about 0.1% (Table 3).

The condition of wind erosion in each sub-region indicates that the LRP and the GKMR underwent the most severe wind erosion, with a multi-year mean modulus of 1.59 and 1.51 t·hm⁻², respectively. They were followed by the SNP, with a wind erosion modulus of 1.15 t·hm⁻². The wind erosion moduli in the SJP, the LKMR, and the CBMR were small. Regarding the wind erosion condition in different ecosystems, the erosion moduli of bare land and grassland were the largest, which were 3.99 and 2.30 t·hm⁻², respectively. The wind erosion moduli of cropland and forests were smaller, at 0.36 and 0.18 t·hm⁻², respectively (Table 2).

By comparing the assessment results of water and wind erosion in Northeast China, we could find that in the past two decades, the primary type of erosion in Northeast China was water erosion; the percentage of water erosion at light and higher degrees was 3.4% more than wind erosion and the multi-year mean amount of water erosion was 34.3% more than wind erosion. The results indicate that water erosion was more severe than wind erosion in Northeast China.

3.2. Changes in Soil Erosion

From 2000 to 2020, water erosion in Northeast China had an overall intensifying trend (Figure 3). The area of water erosion at light or higher degrees increased by 672.1 km² per year on average. Most regions (except LRP) had a greater area of water erosion at light or higher degrees in the latter decade (2010–2020) than in the first decade (2000–2010) (Figure 4). In the past 20 years, the water erosion modulus in Northeast China increased by 0.001 t·hm⁻² per year on average. Among the sub-regions, the CBMR, the SJP, and the SNP’s water erosion showed an intensifying trend, increasing by 0.038, 0.017, and 0.012 t·hm⁻², respectively. At the same time, the LRP’s water erosion showed a moderating trend and decreased by 0.066 t·hm⁻² (Table 2).

Different types of ecosystems showed different change trends of water erosion. The water erosion in bare land and cropland increased significantly, as seen from their interannual variation rates of water erosion modulus at 0.01 and 0.002 t·hm⁻²·a⁻¹, respectively. On the contrary, grassland and forests’ water erosion showed a decreasing trend at the rates of −0.007 and −0.002 t·hm⁻²·a⁻¹, respectively (Table 2).

Figure 5 shows the year-by-year changes in soil erosion in Northeast China. From 2000 to 2005, the water erosion modulus in Northeast China increased steadily, but after 2005 the fluctuation of the modulus was significant, mainly due to the dramatic fluctuation of water erosion moduli in sub-regions such as the CBMR and the LRP (LN) (Figure 5a). The changing trend in the water erosion modulus in the two sub-regions was still close to the overall modulus trend in Northeast China. If we compare the water erosion condition of the first and the second decades from 2000 to 2020, we find that, except for the LRP, the other sub-regions have all shown a trend where the area of slight water erosion decreased, and that of light water erosion increased. The trend means the slight water erosion in some areas has intensified into light erosion. The mean water erosion modulus of the first decade in Northeast China was 1.60 t·hm⁻² and the water erosion amount was 1.98 × 10⁸ t. In the second decade, the mean water erosion modulus was 1.51 t·hm⁻² and the water erosion amount was 1.87 × 10⁸ t. The water erosion amount in the second decade decreased by 5.6%, indicating that the capacity for soil conservation in Northeast China has improved. In the second decade, the water erosion amount in cropland and forests decreased by 2.7% and 14.1%, respectively, while the water erosion for bare land increased by 14.3%. Regarding the interannual variation rate of water erosion, the erosion modulus in Northeast China increased by 0.083 t·hm⁻² per year in the first decade and decreased by 0.032 t·hm⁻² per year on average in the second decade. The interannual variation rates of the water erosion modulus in the second decade for every type of ecosystem (except for bare land) were all smaller than that in the first decade, which means the water erosion has been alleviated to a certain extent in the second decade (Figure 3 and Table 2).

From 2000 to 2020, wind erosion in Northeast China had an overall moderating trend (Figure 3). The area of slight wind erosion increased by 5161.3 km² per year on average, and the area of light, moderate, and intensive wind erosion all decreased slightly. The decrease amounts in those degrees were 5143.4 km², 17.0 km², and 0.9 km² per year, respectively (Table 3). The wind erosion modulus decreased by 0.023 t·hm⁻² per year on average, and among the sub-regions, the LRP’s wind erosion modulus decreased by 0.089 t·hm⁻², which was much higher than others.

Different ecosystems showed different degrees of decreasing trends of wind erosion. The wind erosion in bare land and cropland decreased significantly, as seen from their interannual variation rates of wind erosion modulus at −0.13 and −0.03 t·hm⁻²·a⁻¹, respectively. Grasslands’ and forests’ wind erosion showed a slightly decreasing trend at the rates of −0.02 and −0.01 t·hm⁻²·a⁻¹, respectively (Table 2).

Regarding the year-by-year changes in wind erosion modulus in Northeast China (Figure 5), there was a gradually decreasing trend in fluctuations from 2000 to 2014. After 2014, the wind erosion modulus rebounded, mainly because the soil erosion amount in the GKMR (primarily in IM) increased significantly during this period. In terms of ecosystem types, the year-by-year changes in wind erosion in bare land had pronounced fluctuation, followed by grasslands. Wind erosion in forests was relatively stable. If we compare the condition of the first and the second decades from 2000 to 2020, we find that all sub-regions have shown a trend where the area of slight wind erosion increased, and that of light wind erosion decreased. The trend means the light degree of wind erosion has alleviated to a slight one. The mean wind erosion modulus of the first decade in Northeast China was 1.27 t·hm⁻² and the wind erosion amount was 1.57 × 10⁸ t. In the second decade, the mean wind erosion modulus was 1.04 t·hm⁻² and the wind erosion amount was 1.29 × 10⁸ t. The wind erosion amount in the second decade decreased by 18.2%, indicating that the wind erosion in Northeast China has alleviated significantly. In the second decade, the wind erosion amount in bare land, cropland, grassland, and forests decreased by 28.3%, 26.4%, 24.6%, and 8.0%, respectively. Regarding the interannual variation rate of wind erosion, the process in Northeast China showed a trend of decreasing before increasing, but it was overall decreasing. The wind erosion modulus in Northeast China decreased by 0.067 t·hm⁻² per year on average in the first decade and increased by 0.025 t·hm⁻² per year in the second decade. Among the different types of ecosystems, the forests’ wind erosion modulus showed a decreasing trend in the first and second decades. This fact indicates that although the wind erosion modulus of the forests in Northeast China did not decrease significantly, the forests can still perform long-term windbreak and sand fixation. Regarding the difference among the geographical sub-regions, the wind erosion moduli in the LRP and the CBMR showed a decreasing trend in both decades, while other sub-regions decreased before increasing during the decades. In general, wind erosion in Northeast China showed a decreasing trend from 2000 to 2020. The wind erosion rate decreased rapidly in the first decade and then decreased slowly or even increased in the second decade (Figure 3 and Table 2).

4. Discussion

4.1. Assessment Results of Soil Erosion

In this paper, we conducted an in-depth analysis of the spatial and temporal evolution of water and wind erosion in Northeast China from 2000 to 2020. The results indicate that the erosion degree in Northeast China was mainly slight, followed by light erosion. The moderate and intensive erosion areas were few. This result was consistent with that of Wang et al. [48]. In addition, Wang et al. [48] suggested that the proportion of erosion areas with light and higher degrees was slightly higher than that in this paper, which may be due to differences in assessment methods. In this study, we adopted the RUSLE and RWEQ models to evaluate the amount of water and wind erosion quantitatively. Meanwhile, the soil erosion results from [48] come through the interpretation of remote sensing images, comprehensively considering reference indicators, such as land cover type, gully density, vegetation structure and coverage, slope, elevation, soil type, and others, which belong to qualitative analyses. In other existing studies, Wang et al. [83] used the RUSLE model to estimate water erosion in Northeast China from 2000 to 2018 and the results suggested that the CBMR and the LRP underwent severer water erosion, which was consistent with the results here. In addition, the results of this paper show that although water erosion has increased overall in the last 20 years, the degree of water erosion has decreased in the second decade compared with the first decade, which means that the sharp increase in water erosion mainly occurred in the first decade. The research results of Jiang et al. [84] showed that water erosion in HLJ showed an increasing trend before 2005 and a weakening trend after 2005, which is basically consistent with the results of this paper, mainly due to the government’s emphasis on soil erosion control. In 2008, policy makers carried out some projects, such as the key erosion control project and the integrated prevention and control pilot project in the black soil region of Northeast China, and the development and implementation of these policies effectively mitigated soil erosion in the second decade in Northeast China. The research results of Li et al. [85] on the changes in wind erosion in IM suggested that wind erosion has a significant decreasing trend among fluctuations, which was also basically consistent with our results. In addition, a study from Yang et al. [86] on the number of days under wind erosion in the three northeastern provinces (HLJ, JL, and LN) suggested that the LRP, the SJP, and the SNP had more days of wind erosion, which was largely consistent with our results as well.

4.2. Influencing Factors of Soil Erosion

Soil erosion in Northeast China develops under various types of complex processes. Its prominent features include multi-force coupling, multi-process overlapping, and significant influences from freezing and thawing [71,87]. The influencing factors of water erosion include rainfall, rainfall intensity, duration [88], and other relevant factors such as vegetation, soil, and terrain. However, these factors may be susceptible to the influence of human activities. For example, land use types, cultivation methods, and crop management may affect vegetation cover, land morphology, and the soil’s physical and chemical properties [89]. In addition, the soil in Northeast China has poor erosion resistance and is prone to being affected by freezing and thawing. Meanwhile, over-reclamation, single land use structures, and exploiting management have aggravated soil erosion there [71].

The results of this study revealed that the sub-regions with the most severe water erosion in Northeast China were the CBMR and the LRP, where the terrain is steep and the LS value is high. In addition, the CBMR is close to the Bohai Sea and has abundant rainfall and the highest R value, and its erosion modulus was larger than most of the sub-regions there. The LRP has the highest K and p values and was the main affected area of erosion at intensive or higher degrees. If values of other factors remain high, a slight alternation in the C factor will inevitably result in a significant change in the erosion modulus. The GKMR includes the Greater Kinggan Mountains and the Inner Mongolia Plateau. This sub-region has steep terrain and covers many areas with high K, LS, C, and p values [83].

Due to the limitations of observation and research methods, wind erosion studies in Northeast China were scarcer than water erosion studies. Multiple factors can cause wind erosion, including wind speed, precipitation, soil water content, soil texture (such as soil structure, particle composition, and others), vegetation cover, and cultivation methods [86,90]. Our results revealed that the wind erosion in the GKMR was more severe than in other sub-regions. The sub-region covers a part of the Inner Mongolia Plateau, which is an arid and semi-arid area and a hot spot for wind erosion studies.

Under the influence of human activities, land use types in Northeast China have changed significantly. Over-reclamation, over-harvesting, and over-grazing have reduced vegetation coverage, and cropland, forests, and grassland have become the primary land use types affected by soil erosion [48]. From 2000 to 2020, the land use type conversion in Northeast China was mainly the conversion between cropland and forest and between cropland and grassland (Figure 6a). From 2000 to 2010, a large area of grassland was converted to cropland and bare land, and there were also forests converted to grassland. The surface vegetation there was destroyed due to over-exploitation and grassland’s reclamation into cropland, and degradation into sandy and saline–alkali land. The vegetation’s destruction loosened the soil, thus aggravating the water and soil losses. Later, under conservation tillage methods such as the conversion of cropland to forests and natural forest protection projects, the area of forest to grassland, grassland to cropland, and grassland to bare land in the second decade reduced remarkably. More bare land has been converted into forests and grasslands, alleviating water and soil losses. However, we found that forests in some areas were also reclaimed as cropland (from the increase in the reclaimed area in the second decade), which aggravated the water and soil losses.

Figure 6b shows the differences between LUCC and soil erosion changes in Northeast China during specific periods. For the areas where the land converted from forest to bare land, their water and wind erosion moduli showed an increasing trend in each period, which means the soil erosion in these areas was aggravated. From 2000 to 2010, the wind erosion modulus decreased significantly in areas where the land was converted from bare land to cropland, to forests, and to grassland. In addition, almost all the water and wind erosion moduli showed a decreasing trend in the areas where the land was converted to forests and grassland. This fact suggests that forests and grassland can assist soil conservation and sand fixation in Northeast China. However, the responses of water erosion to land use changes from 2000 to 2010 also reveal that the areas converted into forests underwent increased water erosion, which was quite different from the above results. This extraordinary result may be related to the rainfall factor during this period.

4.3. Methods to Moderate Soil Erosion

Soil erosion control and the protection of black soil in Northeast China is urgent and this has become a consensus among people. Based on the above-mentioned factors influencing soil erosion, we can find some methods to moderate soil erosion. For example, we should prioritize improving vegetation coverage in LRP since it is a prominent factor in countering soil erosion. In addition, to address erosion in GKMR, we should first reduce the impact of rainfall on erosion by reducing slope cropland and implementing conservation tillage methods such as zero tillage. In terms of LUCC, changes in cropland and bare land (i.e., cultivation reversion and wasteland elimination) can help reduce soil erosion. Conversely, the loss of forests and grasslands aggravates soil erosion. In recent years, policymakers have attached great importance to cropland protection in Northeast China. In 2021, the government issued a document stating, “we will implement the national black soil protection project and promote conservation tillage methods”, which upgraded the protection of black soil in Northeast China into a national strategy. The Ministry of Water Resources has compiled the Outline of the Black Soil Protection Plan in Northeast China (2017–2030), and the region has gradually applied conservation tillage methods such as zero tillage, less tillage, cross ridge tillage, and straw mulching to address soil erosion and water and soil losses there. Previous studies have suggested that zero tillage with straw mulching has multiple advantages. For example, it can improve soil structure and structural stability and increase water conservation capacity and overall soil quality to reduce water erosion [91]. In addition, compared to longitudinal ridge tillage, cross ridge tillage can better retain water and soil, thus reducing organic carbon loss and increasing crop yield [92,93]. For wind erosion, conservation tillage methods, such as zero tillage, subsoiling, ridge cultivation, and straw mulching, are available. They act to alleviate wind erosion by reducing soil disturbance and increasing stubble-mulching on the soil surface. Furthermore, these prevention and control measures for soil erosion (strip cropping, contour farming, contour bunds, and terracing) are evidenced in the Ukrainian plain [94], which is also one of the four major black soil regions in the world and very similar to Northeast China.

4.4. Limitations and Future Research

Our study still has some uncertainties and pending questions: (1) Soil erosion was caused by multiple factors such as climate, topography, soil, vegetation, and human activities. Factors such as climate change and cultivation methods may also impact changes in soil erosion. Our study only discussed the temporal and spatial evolution of soil erosion in Northeast China over the past two decades and the reasons behind the changes. The quantitative analysis of the driving factors of those changes in Northeast China requires systematic and in-depth research. (2) Our study only considered water and wind erosion, and the freeze–thaw cycle was not included. As a prominent process of soil erosion in Northeast China, the freeze–thaw cycle can aggravate water erosion by affecting soil hydrological properties and reducing erosion resistance. (3) Soil erosion may be affected by temporal alternation or synchronization and spatial intersection or overlapping. Thus, various erosion types such as water and wind erosion in Northeast China coexist and interact with each other. Herein, we discussed water and wind erosion in Northeast China separately, but our study did not include the interaction between them. (4) Although this study reveals the changes of soil erosion in Northeast China to some extent, experimental data are lacking to verify the accuracy. To improve the estimation accuracy of the amount of soil erosion, researchers should further localize the parameters of the management factors in the erosion equation in combination with field investigation. (5) Vegetation and crop growth are also affected by the seasonal distribution of temperature and precipitation. Thus, there are significant seasonal changes in erosion intensity in Northeast China. This feature also deserves further study [71]. (6) There are differences in soil status, erosion characteristics, and planting patterns in different climatic sub-regions, which lead to differences in the applicability of conservation tillage methods in different areas in Northeast China. Future studies may need to include the climates and their changes and soil and agricultural production status of each sub-region in Northeast China to screen and propose conservation tillage methods suitable for specific regions.

5. Conclusions

Based on the RUSLE and the RWEQ models, this paper assessed the water and wind erosion on the soil in Northeast China from 2000 to 2020 and analyzed the current situation, as well as the spatial and temporal evolution of soil erosion. The results suggest that:

(1) Northeast China’s multi-year mean water and wind erosion moduli were 1.52 t·hm⁻² and 1.13 t·hm⁻², respectively, and the water and wind erosion amounts were 1.88 × 10⁸ t and 1.4 × 10⁸ t, respectively. The degree of soil erosion in Northeast China was mainly slight, and water erosion was more severe than wind erosion. Water and wind erosion in bare land and grassland were more intensive than in cropland and forests. The LRP has undergone relatively intensive water erosion, while the wind erosion in the GKMR was more intensive than in other sub-regions.

(2) There was a slight intensifying trend in water erosion in Northeast China from 2000 to 2020, which can be found in its increased water erosion modulus at 0.001 t·hm⁻² per year on average. From a spatial view, the area of slight water erosion decreased, and that of light and intensive water erosion increased. The water erosion in the CBMR, the SJP, and the SNP intensified, while the LRP underwent water erosion slower than before. The water erosion in bare land and cropland intensified, while the water erosion in grassland and forests slowed down. On a temporal scale, the results reveal that compared to the first decade, the second decade in the timespan of the study had a reversed change trend of water erosion from intensifying to moderating, which means the soil retention capacity in Northeast China has improved, and water erosion of the soil there has been alleviated to some degree.

(3) There was a moderating trend in wind erosion in Northeast China from 2000 to 2020, which can be found in its decreased wind erosion modulus at 0.023 t·hm⁻² per year on average. The area of slight wind erosion increased, and that of light, moderate, and intensive wind erosion decreased. The wind erosion in the LRP had a pronounced decrease and the wind erosion in bare land and cropland also decreased considerably. The decrease in wind erosion in grassland and forest was relatively slight. The amount of wind erosion in the second decade decreased by 18.2% compared to the first decade. The rate of wind erosion decreased faster in the first decade, but in the second decade, it decreased slowly or even increased. The two facts indicate that wind erosion in Northeast China has alleviated, but this trend is gradually slowing down.

Soil erosion is caused by multiple factors, such as climate, topography, soil, and human activities. This study has important implications for our understanding of soil erosion control and management in Northeast China. For example, in sub-regions with severe erosion, such as the LRP and the GKMR, we can adopt methods such as zero tillage, cross ridge tillage, and straw mulching according to the local characteristics of soil erosion to slow down the process.