- Email: [email protected]
The distribution of and factors influencing the vegetation in a gully in the Dry-hot Valley of southwest China
Catena 116 (2014) 60–67
Contents lists available at ScienceDirect
Catena journal homepage: www.elsevier.com/locate/catena
The distribution of and factors influencing the vegetation in a gully in the Dry-hot Valley of southwest China Yifan Dong a, Donghong Xiong a,⁎, Zheng'an Su a, Jiajia Li b, Dan Yang a, Liangtao Shi c, Gangcai Liu a a b c
Key Laboratory of Mountain Hazards and Earth Surface Processes, Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, 610041 Chengdu, China Department of Environmental Engineering, Chengdu University of Information Technology, 610225 Chengdu, China Institute of Tropical Eco-agricultural Sciences, Yunnan Academy of Agricultural Sciences, Yuanmou 651300, Yunnan, China
a r t i c l e
i n f o
Article history: Received 22 April 2013 Received in revised form 8 November 2013 Accepted 18 December 2013 Keywords: Gully Vegetation Soil Topography Dry-hot Valley
a b s t r a c t The Dry-hot Valley of the Jinsha River in southwest China is an ecologically fragile zone in which gully erosion is one of the most important environmental problems due to the high sediment yield from the gully. The vegetation, which impacts the erosion processes of the gully, is important to the ecological environment. In this study, we investigated the vegetation in gully n1 of Yuanmou County, which is a typical area of the Dry-hot Valley. A total of 82 vegetation quadrats on a relatively gentle section of gully n1 (slope was mostly less than 45°) and nine points on a steep slope (slope higher than 70°) were investigated in 2012. The vegetation indices, in addition to the soil conditions and topography, were measured through field investigations and the gully Digital Elevation Model (DEM). On the gully sidewall, almost no vegetation grew in the steep slope, whereas the slope gradient and the slope aspect exhibited a weak relationship with the vegetation indices when the slope was less than 45°. On the gully bed, the runoff path was the most important factor that affected the vegetation. The average vegetation cover inside the runoff path was 5.2%, whereas the average vegetation cover outside the runoff path increased to 56.3%. The vegetation strongly impacted the soil moisture during the rainy season. However, this relationship was not observed during the dry season, which indicates that the water conservation effects exerted by the vegetation cannot last a long time in the Dry-hot Valley. The active part of gully n1 has a vegetation cover (average 16.9%) that is significantly lower compared with the stable parts of the gully (55.1%). This finding indicates that the vegetation in the gully is not only impacted by the soil and the topography in the area of vegetation growth but also influenced by the runoff processes upstream of the gully heads. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Vegetation, which can purify air, filter water, and protect soil, is often impacted by the soil condition and topography in a region (Florinsky and Kuryakova, 1996; Poulos et al., 2007; Xu et al., 2008). The topography can influence the vegetation by affecting the development of roots, the microclimate, and the solar radiation (Alday et al., 2010; Bennie et al., 2006; Li et al., 1991). The vegetation is also affected by various soil properties, such as soil moisture, acidity (pH), and nutrient content. In addition, the growth of vegetation may affect the topography and soil properties through various processes, including respiration, hydrologic control, and soil conservation (Jiao et al., 2009; Osterkamp et al., 2011). The relationship between vegetation, topography, and soil, which may change at different spatial scales, is an important subject of ecological and geographic studies (Solon et al., 2007; Xu et al., 2008).
⁎ Corresponding author at: Key Laboratory of Mountain Hazards and EarthSurface Process, Institute of Mountain Hazards and Environment,Chinese Academy of Sciences, 610041 Chengdu, China. Tel.: +86 28 85259762; fax: +86 28 85222258. E-mail address: [email protected] (D. Xiong). 0341-8162/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.catena.2013.12.009
Gully erosion is an important type of soil erosion, and gully itself is a main source of sediments, which contributes 10% to 94% of the total sediments in different areas of the world (Poesen et al., 2003). The vegetation has beneficial effects in controlling of gully erosion (Munoz-Robles et al., 2010; Rey, 2003), which could result in the reinforcement of steep gully sidewalls and the interception of sediment transport processes on the gully beds. The topography of the gully investigated in this study was quite different from that reported in previous studies. The slope gradient of the vegetation were often reported to be below 30° in previous studies (Solon et al., 2007; Wang et al., 2011; Xu et al., 2008), while slopes reached 35° to 45° thus were described as steep slopes (Wang et al., 2008b). The slope gradient of the gully changed significantly: the gully sidewall was often close to or even more than 90°, whereas the gully bed was often below 5°. The soil composition of the gully was often complicated because it cut deeply into the earth surface, which made the soil type to vary in the longitudinal direction. Thus, the special topography and soil conditions of the gully create a complex environment for vegetation growth. Furthermore, the erosion processes that occurred in different parts of the gully also varied (Leyland and Darby, 2008; Martinez-Casasnovas, 2003; Poesen et al., 2003), and the vegetation conditions, which are strongly correlated with erosion processes, may also be different. However, few studies have focused on
Y. Dong et al. / Catena 116 (2014) 60–67
the vegetation distribution and the factors that influence this distribution in a gully. The Dry-hot Valley of the Jinsha River is an ecologically fragile zone in southwest China that covers an area of approximately 1.2 × 104 km2 (Xiong et al., 2010), and gully erosion is a serious problem in this area. According to previous studies, the distribution density of most gully area ranged from 3 to 5 km/km2, with the highest density being 7.4 km/km2 (Cheng et al., 2011; Wang et al., 2008a). This distribution results in a high sediment yield and a rapid degradation of land. The revegetation of the gullies is very important for the improvement of the ecological environment of this region, although the gullies in this region are often large in scale with a deep incision, steep gully sidewall and a complex soil composition. These special characteristics may influence the vegetation distribution and growth in the gully. The aim of this study was to investigate the vegetation conditions in conjunction with the topography and the soil conditions in the gully and to analyse the distribution of vegetation in different parts of the gully. 2. Materials and methods 2.1. Study area The Yuanmou Dry-hot Valley, which covers an area of more than 2000 km2, is located in the upper and middle reaches of the Jinsha River (an important tributary of the Yangtze River) in Yunnan Province (Fig. 1a) in southwest China (Xiong et al., 2009). The region exhibits a typical southern subtropical climate with an annual average temperature of 21.8 °C. The annual precipitation is 615 mm, and the annual potential evaporation is 3569 mm, which results in the extreme aridness of this region (Wang et al., 2008a). The rainy season, which often lasts from June to early October, accounts for more than 85% of the annual precipitation. Gullies often developed at the quaternary fluvial–lacustrine deposits with a loose structure at an elevation of in the range of 1000 to 1350 m above sea level, and the main soil types at the surface of the region were dry red soil (classified as Ustic Ferrisols in Chinese Taxonomy) and vertisols. Dry red soil has a high sand content (often above 50%) and contains iron and manganese, whereas vertisols are mainly composed of clay (often over 50%) with a strong expansibility. In addition, a sandy soil layer, which could easily be eroded by water, often develops under these two types of soil, and a large amount of sediments are deposited in the middle and downstream of the gully bed. The vegetation in the gully is dominated by the herbs Heteropogon and Bothriochloa pertusa, and a few Dodonaea viscosa (L.) Jacg. shrubs grow in some parts of the gully (Zhang et al., 2003). 2.2. Quadrat investigation of the vegetation We randomly selected 82 1 m × 1 m quadrats in gully n1 (Fig. 1b) at the beginning of the rainy season (June 22th) in 2012, which was the very next day after a three-day period of continuous rainfall (about 15 h after the rainfall stopped). The vegetation indices, including the plant species, the percentage of vegetation cover, the number of plants, and the height of the plants, of each quadrat were determined. According to previous investigation and studies, the length of the herb roots in gully area was mainly ranged between 5 and 30 cm. So a soil depth of 0 to 30 cm was sampled in each quadrat which was mainly concerned with the vegetation condition. A total of 82 samples were oven-dried at 105 °C for 12 h to test the soil moisture by mass in the rainy season. The location of each quadrat was recorded by highprecision RTK GPS (Trimble R8, the horizontal and vertical positioning accuracy were 1 cm ± 2 ppm and 2 cm ± 2 ppm, respectively). During the dry season (November 6th, 46 days without any precipitation) in 2012, we used the lofting function of the RTK GPS to find the locations of the quadrats investigated in June; the navigation accuracy reached the centimetre level. The vegetation indices and the soil moisture were measured in the dry season to obtain the changes in the
61
vegetation conditions and the soil moisture in the same quadrats. In addition, the vegetation conditions at nine points on the steep gully sidewall (slope over 70°) of gully n1 were also investigated. 2.3. Gully topography measurement and data processing Gully n1 was located in the valley floor of a small catchment in Yuanmou (Fig. 1c) and was measured using an RTK GPS in April 2012. A total of 11,280 high-precision topographic points were monitored along the boundary of the gully sides, the edge of the gully bottom where the gully sidewall changes abruptly into a flatter gully bed, and randomly distributed on the sidewall and the bottom of the gully. These data were transferred to ArcGIS 9.3, used to create Delaunay triangulated irregular networks (Tins) using the 3D Analyst tool, and then converted into a DEM grid with 1-m cell size (Wu et al., 2008). The elevation and depth of each cell could be extracted from the DEM grid. The slope gradient and the slope aspect of each cell could be calculated using the Spatial Analyst tool of ArcGIS, and the topography information of each quadrat could be obtained according to its location cell in the DEM grid. According to the results, the area and volume of gully n1 were 16,939.4 m2 and 124,514.5 m3 respectively. The mean depth of the gully was 7.6 m, and maximum depth was 15.9 m. The elevation ranged between 1060.4 m and 1090.9 m, and the slope gradient ranged from 0.45° to 87.8° with an average of 34.3°. 3. Results 3.1. Vegetation, soil, and topography of the quadrats in the gully Herbs were observed to be the major vegetation type in the gully. All of the quadrats with vegetation contained herb plants: 94.7% of the quadrats contained Heteropogon and 37.3% of the quadrats contained B. pertusa. In addition, 4.9% of the quadrats contained shrubs, and the species was D. viscosa (L.) Jacg. The vegetation conditions in November were slightly better than those in June because the herbs in the gully, which had just begun to grow in June and were thus fully grown after the rainy season, were tolerant to drought and had therefore not completely wilted in November (after more than 1 month without precipitation). A total of 12 of the 82 quadrats did not have any vegetation in June, and 5 of these 12 quadrats contained herb plants in November; however, the average vegetation cover in these quadrats was 3.6%, and only one of the five quadrats exhibited a vegetation cover of more than 5%. The average vegetation cover in the 82 quadrats increased by 6.0% from June to November. Although the number of plants did not significantly change, the height of the plants increased by 38.6% (Table 1). The height of the plants enhanced in November, but the leaves of the herbs began to wilt and the colour of the herbs was mostly turned into yellow already. The high coefficient of variation (Cv) of the vegetation cover revealed that the vegetation distribution varied between different locations of the gully, likely due to the complex topography and soil conditions. Because the growing season of the plants was mainly during the rainy season and the vegetation was not fully grown at that point, the vegetation indices in the rest of the manuscript indicate the results obtained in the dry season (November) unless otherwise noted. The analysis of the 82 quadrats revealed a significant relationship between the vegetation indices, including the vegetation cover and the number and the height of the plants (Fig. 2). The soil water content (by mass) in June (SWC1) was significantly higher than that in November (SWC2); the average SWC1 was 7.45%, whereas the SWC2 was only 3.78%. The slope of the quadrats ranged from 1.25° to 61.35° with an average of 22.05°. The slope aspect was converted using the cosine function to a range from − 1 to 1, which indicates south to north (Xu et al., 2008), and the average was − 0.27. The depths of the quadrats ranged from 1.28 m to 15.50 m, and the elevation ranged from 1060.70 m to 1088.44 m (Table 2).
62
Y. Dong et al. / Catena 116 (2014) 60–67
Y. Dong et al. / Catena 116 (2014) 60–67
63
Table 1 The vegetation condition of 82 quadrats in June (during the rainy season) and November (during the dry season) in 2012. Vegetation cover (%)
Average Stdev Cv
Number of plants
Height of plants (cm)
June
November
June
November
June
November
37.0 32.5 88.0%
43.0 35.6 82.9%
6.1 4.3 70.2%
6.3 4.1 67.4%
29.8 20.0 67.1%
41.3 22.8 55.2%
homogeneous slope gradient of often below the response angle (approximately 45° with vegetation). The vegetation distributions in the two types of sidewalls are different and clearly affected by the slope gradient. There were nine points analysed on the cliff part of sidewall type A, where the slope gradient ranged from 71.53° to 83.35°, and there was no vegetation cover because it was difficult for the plant seeds to develop in the surface soil of the cliff. The vegetation was mainly distributed on the slope wash part of the type A and type B sidewalls, which often present slope gradients of less than 45°, and the vegetation cover of the 39 sidewall quadrats ranged from 0 to 96% with an average of 49.0% and a coefficient of variation (Cv) of 78.1%. A total of 43 quadrats were investigated on the gully bed, and the most important factor that impacted the vegetation distribution on the gully bed was whether its location was inside the runoff path. The earth surface of runoff path was often scoured by strong concentrate flow, which could sweep away the plant seeds and reduce the soil water conservation capacity, and finally result to very low vegetation cover on the runoff path. The investigation in June was conducted right after a three-day period of continuous rainfall, and the runoff path could be clearly identified; as a result, it was observed that 17 of the investigated quadrats were located inside the runoff path. We also used the Hydrology tool of ArcGIS to create the runoff path according to the gully DEM, and 16 of the 17 quadrats were in or one cell adjacent to the calculated runoff path. The vegetation cover of these quadrats in November ranged from 0 to 15%; 14 quadrats exhibited a vegetation cover of less than 10%, and the average was only 5.2%. In contrast, the vegetation cover in the other gully bed quadrats ranged from 11% to 98% with an average of 56.3%. Distinct differences in the vegetation distribution could be identified at the boundary of the runoff path: a high vegetation density with substantial growth was observed outside the boundary, whereas almost no vegetation cover was observed inside the runoff path (Fig. 4). Fig. 2. Relationships between vegetation indices including percentage of vegetation cover (a), number and height of the plants (b).
4. Discussion 4.1. The relationship between gully vegetation and topography
3.2. Vegetation on the gully sidewall and the gully bed Gullies can often be divided into the gully sidewall and the gully bed according to their position in the landscape (Bocco, 1991). The sidewall often has a steep slope with different soil types at different depths. In contrast, the gully bed has a relatively gentle slope gradient at the base of the sidewall, and the surface of the gully bed is often covered with sediments that are deposited by the runoff. The various topographies and soil types result in different environments for vegetation in the gully. In general, there are two types of gully sidewalls in our study area (Fig. 3). Type A often exhibits a steep cliff (with a slope of over 70°) at the upper and middle part and a slope wash (below 45°) at the bottom of the sidewall, whereas type B has a long slope with a relatively
Previous studies have demonstrated that the slope gradient, the slope aspect, and the elevation are the most important factors that impact a vegetation habitat (Miller et al., 1996; Ostendorf and Reynold, 1998; Pearson et al., 1999). In contrast, the conditions in a gully are more complicated. The slope gradient was observed to be the dominant factor that determined vegetation growth on the cliff part of the gully sidewall, which was mostly a type A slope. Regardless of the type of soil and the moisture and nutrient contents of the soil, almost no vegetation grew on the cliff part of the gully sidewall. When the slope gradient was less than 45°, the vegetation indices exhibited some correlation with the slope gradient and the slope aspect, which could impact the survival and development of the vegetation as a result of moisture migration and sunlight conditions. However, this relationship was not very strong because the R2 ranged from 0.10 to 0.15 (Table 3). The
Fig. 1. The location of the study area, the distribution of the vegetation quadrats, and the general situation in gully n1. a. Location of the Yuanmou Dry-hot Valley. b. Distribution of the vegetation quadrats in gully n1 of the DEM. c. Boundary of gully n1.
64
Y. Dong et al. / Catena 116 (2014) 60–67
Table 2 Basic information of the soil and topography of the 82 quadrats. Quadrat ID
Position
SWC1a
SWC2b
Slope gradient (°)
YF001 YF002 YF003 YF004 YF005 YF006 YF007 YF008 YF009 YF010 YF011 YF012 YF013 YF014 YF015 YF016 YF017 YF018 YF019 YF020 YF021 YF022 YF023 YF024 YF025 YF026 YF027 YF028 YF029 YF030 YF031 YF032 YF033 YF034 YF035 YF036 YF037 YF038 YF039 YF040 YF041 YF042 YF043 YF044 YF045 YF046 YF047 YF048 YF049 YF050 YF051 YF052 YF053 YF054 YF055 YF056 YF057 YF058 YF059 YF060 YF061 YF062 YF063 YF064 YF065 YF066 YF067 YF068 YF069 YF070 YF071 YF072 YF073 YF074 YF075
Gully sidewall Gully sidewall Gully sidewall Gully sidewall Gully bed Gully bedc Gully sidewall Gully bedc Gully sidewall Gully sidewall Gully sidewall Gully bedc Gully sidewall Gully sidewall Gully bed Gully bedc Gully bedc Gully bed Gully bed Gully sidewall Gully bed Gully bed Gully sidewall Gully bedc Gully bed Gully bedc Gully sidewall Gully bed Gully sidewall Gully bedc Gully sidewall Gully bed Gully bed Gully bedc Gully bed Gully bed Gully bedc Gully bedc Gully sidewall Gully sidewall Gully bedc Gully bed Gully sidewall Gully bedc Gully sidewall Gully bedc Gully sidewall Gully bedc Gully bed Gully sidewall Gully bed Gully sidewall Gully sidewall Gully bed Gully sidewall Gully sidewall Gully sidewall Gully sidewall Gully bed Gully bedc Gully sidewall Gully sidewall Gully sidewall Gully sidewall Gully bed Gully bed Gully bed Gully bed Gully sidewall Gully bed Gully bedc Gully bed Gully sidewall Gully sidewall Gully sidewall
2.36% 8.06% 9.79% 4.03% 4.99% 7.42% 5.18% 8.59% 4.89% 6.22% 9.81% 12.45% 8.98% 5.23% 9.28% 6.65% 9.63% 10.94% 5.49% 3.62% 9.77% 8.18% 4.56% 5.82% 2.64% 5.98% 5.44% 9.52% 12.05% 7.92% 3.82% 11.44% 4.93% 7.42% 9.83% 8.44% 5.65% 5.19% 10.31% 5.94% 6.32% 7.39% 6.02% 5.94% 7.59% 4.68% 8.30% 4.88% 8.33% 7.47% 4.13% 4.11% 4.84% 7.16% 7.00% 9.81% 8.47% 8.88% 12.87% 4.79% 8.79% 4.69% 7.67% 3.88% 9.35% 12.85% 5.84% 9.26% 6.53% 5.68% 8.40% 10.30% 11.61% 10.84% 5.48%
2.37% 2.54% 2.69% 1.29% 2.91% 3.36% 1.14% 2.48% 2.31% 4.93% 3.15% 3.82% 2.19% 3.92% 2.28% 1.85% 4.73% 1.84% 1.12% 4.38% 3.17% 4.34% 2.28% 7.52% 2.03% 2.27% 4.89% 7.23% 5.86% 4.33% 11.08% 4.15% 5.59% 5.56% 1.70% 7.11% 2.97% 2.65% 2.77% 3.55% 3.75% 4.33% 2.70% 1.68% 2.45% 4.52% 2.14% 3.03% 8.19% 2.17% 3.54% 1.37% 1.21% 2.60% 3.23% 4.90% 4.34% 4.53% 4.33% 3.01% 3.94% 2.29% 3.20% 1.58% 2.20% 4.44% 3.92% 2.86% 7.55% 1.72% 2.28% 6.30% 7.36% 1.87% 9.49%
18.86 17.11 20.97 18.08 17.12 23.32 45.84 5.72 21.37 29.39 37.31 20.21 33.21 36.06 12.41 17.94 6.20 7.69 10.64 55.92 35.02 31.14 61.35 56.85 15.71 1.25 21.08 27.52 30.99 40.34 48.49 16.62 15.78 14.21 8.34 8.98 9.22 5.15 19.52 19.77 14.09 20.89 19.12 12.56 31.90 6.31 11.99 1.25 8.38 30.24 20.62 36.19 13.08 17.59 30.92 14.35 12.47 31.51 5.36 2.13 19.83 31.66 44.02 40.14 24.11 11.26 4.72 5.08 41.27 16.47 11.56 22.26 38.85 21.97 31.19
Slope aspect −0.24 −0.14 −0.97 −1.00 0.12 −0.10 0.24 −0.85 −0.29 −0.24 0.60 −0.74 −1.00 −1.00 −0.99 0.85 −0.98 0.42 −1.00 −0.99 0.72 −1.00 −1.00 −0.41 0.41 −0.09 0.53 0.44 0.84 0.95 0.45 −0.09 0.07 0.17 −0.66 −1.00 −0.10 −0.19 0.59 −0.96 −0.83 0.10 −0.29 −0.97 −0.42 0.95 −0.98 −0.29 −0.98 −0.75 −1.00 −0.80 −0.96 −0.29 −0.07 −0.31 −0.45 0.63 −0.96 0.77 0.97 0.90 0.39 −0.54 −0.95 −0.96 −0.95 −0.63 0.97 −0.87 0.09 0.10 0.87 −0.28 −0.54
Depth (m)
Elevation (m)
Main soil type
5.87 10.88 13.02 11.11 9.25 7.15 5.80 3.33 5.45 3.81 10.85 13.57 8.36 5.63 13.34 14.57 14.85 14.10 11.88 8.07 10.38 11.73 5.92 6.24 9.67 5.22 1.28 3.38 2.83 3.63 4.33 4.51 11.55 14.58 14.60 14.87 15.29 15.50 13.27 10.65 9.35 5.33 7.07 13.18 14.32 14.01 12.79 13.68 12.93 9.78 11.03 8.83 8.41 6.83 7.17 3.68 5.82 8.18 12.92 11.12 11.25 9.19 6.83 4.26 8.39 13.83 15.19 14.75 8.62 8.37 3.78 3.54 5.20 7.72 2.08
1075.13 1069.11 1066.44 1069.26 1071.95 1073.42 1075.67 1077.91 1075.93 1078.20 1069.74 1067.32 1072.81 1075.12 1067.82 1066.12 1066.90 1068.08 1069.89 1073.31 1071.48 1073.26 1080.25 1081.03 1076.37 1083.31 1088.44 1086.67 1082.87 1083.40 1080.20 1078.06 1069.16 1065.31 1065.71 1064.51 1068.96 1063.30 1064.80 1068.43 1070.49 1073.57 1069.85 1065.08 1063.33 1062.16 1063.22 1061.24 1060.86 1064.32 1063.84 1067.21 1066.61 1068.66 1072.48 1070.87 1067.37 1064.68 1060.79 1060.70 1063.47 1065.67 1068.51 1072.59 1067.45 1060.90 1062.49 1063.68 1070.16 1072.62 1081.93 1081.69 1077.55 1074.28 1081.58
Dry red soil Dry red soil Dry red soil Dry red soil Dry red soil Sediment Sediment Sediment Dry red soil Dry red soil Sandy soil Sandy soil Dry red soil Dry red soil Sediment Sediment Sediment Sediment Sediment Dry red soil Sandy soil Sediment Dry red soil Dry red soil Vertisols Sediment Dry red soil Vertisols Vertisols Sediment Vertisols Dry red soil Dry red soil Sediment Sediment Sediment Sediment Sediment Sandy soil Sandy soil Sediment Dry red soil Sandy soil Sediment Sandy soil Sediment Sandy soil Sediment Sediment Dry red soil Sediment Vertisols Sandy soil Sediment Sandy soil Vertisols Vertisols Vertisols Sediment Sediment Vertisols Vertisols Vertisols Dry red soil Vertisols Sediment Sediment Sediment Vertisols Vertisols Sediment Dry red soil Vertisols Vertisols Vertisols
Y. Dong et al. / Catena 116 (2014) 60–67
65
Table 2 (continued) Quadrat ID
Position
SWC1a
SWC2b
Slope gradient (°)
Slope aspect
Depth (m)
Elevation (m)
Main soil type
YF076 YF077 YF078 YF079 YF080 YF081 YF082 Average Stdev Cv
Gully sidewall Gully sidewall Gully bed Gully sidewall Gully sidewall Gully bed Gully bed – – –
13.52% 11.89% 9.43% 4.60% 2.22% 9.57% 11.22% 7.45% 0.03 36.25%
9.23% 8.57% 3.05% 1.63% 2.57% 3.86% 5.95% 3.78% 0.02 56.14%
23.67 34.46 14.42 20.84 28.67 23.29 14.98 22.05 13.30 60.33%
−0.42 −0.41 −0.12 0.98 −0.09 −0.60 −0.91 −0.24 0.66 −275.54%
4.34 3.14 8.29 7.12 8.08 13.60 2.52 8.99 4.02 44.71%
1079.28 1080.33 1073.42 1073.06 1071.43 1064.59 1081.48 1071.11 6.81 0.64%
Vertisols Dry red soil Vertisols Vertisols Vertisols Vertisols Vertisols – – –
a b c
Soil water content in June. Soil water content in November. Inside the runoff path.
vertical zonality was not distinct in gully n1, i.e., the depth/elevation exhibited a poor relationship with the vegetation indices (Table 3). The depth of the sidewall quadrats ranged from 1.28 to 14.32 m, whereas the elevation of the gully ranged from 1063.22 to 1088.44 m. The amplitude of the variation in the altitude direction was not sufficient to significantly affect the vegetation in the gully. The runoff path was a unique influencing factor on the vegetation distribution along the horizontal direction of the gully bed. Gully n1 was located in the valley of the catchment, which accumulated runoff both on the slope of the catchment and inside the gully itself (Fig. 1c). A large amount of runoff flowed from upstream of the head of gully n1 to the gully end. This path exhibits elevation changes of
approximately 30 m. The potential energy of the runoff was converted to kinetic energy, which leads to a high scouring force on the earth surface of the runoff path. The high-power runoff is able to sweep plant seeds and seedlings away from the soil and thus stop their growth. In addition, the turbulent runoff could also wash away the silt and clay content from the soil, worsening the water conservation properties of the soil on the runoff path. The average soil water content (SWC1) of the runoff path quadrats was 21.67% lower than the average for other places on the gully bed and 1.73% lower than the average in the sidewall quadrats over 16 h after the rainfall stopped, which are often considered to exhibit a lower soil moisture than the gully bed (Table 2). The low survival rate of plants and the poor water retention of the soil resulted in a bare earth surface on the runoff path. 4.2. The relationship between the gully vegetation and the soil moisture Soil moisture is considered to be a vital factor for the survival of vegetation in Dry-hot Valley (Zhang et al., 2003). However, more than 60% of the quadrats that were investigated exhibited a vegetation cover of more than 20%, and the average height of the herb plants was greater than 40 cm, which indicated that the vegetation was fully grown and thus began to impact the soil moisture during the rainy season. The vegetation indices on the gully sidewall, including the vegetation cover and the number and the height of the plants, exhibited a strong correlation with the soil water content in June (SWC1). The vegetation on the gully bed presented results that were similar to those obtained for the sidewall, but the relationship was weak due to the quadrats inside the runoff path (Table 4). In contrast, a weak relationship was observed between the vegetation indices and the soil water content in November (SWC2), which
Fig. 3. Two main types of gully sidewalls in the Yuanmou Dry-hot Valley. Nearly no vegetation grew on the cliff part of the type A sidewalls, which had a depth that was often more than 10 m and a slope gradient that was often higher than 70°.
Fig. 4. The distinct vegetation distributions inside and outside the boundary of the runoff path.
66
Y. Dong et al. / Catena 116 (2014) 60–67
Table 3 Relationship between vegetation indices and topography on gully sidewall and gully bed. Position
Gully sidewall
Gully bed
Vegetation index
Slope gradient
Vegetation cover Number of the plants Height of the plants Vegetation cover Number of the plants Height of the plants
Slope aspect
Depth
Elevation
R2
P
R2
P
R2
P
R2
P
0.092 0.112⁎ 0.146⁎ 0.027 0.016 0.025
0.060 0.037 0.016 0.292 0.417 0.314
0.110⁎ 0.077 0.125⁎ 0.041 0.068 0.065
0.038 0.087 0.027 0.194 0.092 0.099
0.038 0.104 0.004 0.041 0.029 0.046
0.231 0.065 0.223 0.162 0.273 0.168
0.084 0.106 0.036 0.019 0.015 0.040
0.073 0.063 0.247 0.382 0.430 0.202
⁎ Significance reach 0.05 level.
was mainly due to the slight variation in the SWC2. The soil water content in 69 of the 82 quadrats was less than 5%, whereas only one sample exhibited a soil water content that was higher than 10% in November. The results indicate that, although the vegetation was able to conserve water by intercepting direct sunshine through its aboveground parts and by adsorbing soil water through the plant roots, the effect of water conservation would not last a long time in the Dry-hot Valley. After more than 1 month without water, the soil moisture tended to be uniform throughout the gully. 4.3. The variations in the vegetation in the active and the stable parts of the gully The activity of the erosion processes in different parts of the gully impacted the vegetation distribution in the gully. The active part of gully n1 consisted of five active gully heads that developed quickly (Fig. 1b). Four of the five gully heads were measured by RTK GPS in 2011 and 2012. The average headcut retreat rate was 0.28 m, and the highest headcut retreat rate was higher than 0.6 m in 1 year. In contrast, other parts of the gully were relatively stable and did not exhibit significant changes (Fig. 1b). Ten of the studied quadrats and five of the points analysed in the steep slope were located in the active part of the gully. The average vegetation cover in these quadrants and points was 16.9%, and 7 of the 15 locations did not have any vegetation. In contrast, the average vegetation cover in the stable part of the gully was 43.3%, and 36 of the 77 quadrats exhibited a vegetation cover higher than 50%. A total of 14 of the 17 quadrats along the runoff path, which exhibited an average vegetation cover of 5.1%, were located in the stable part of gully n1. However, these quadrats suffered from strong runoff scouring during the rainy season and can thus not be described as stable. After excluding these quadrats, the average vegetation cover of the stable part of the gully was enhanced to 55.1%. The activity of the gully erosion was mainly impacted by the intensity of the runoff, which was determined by the precipitation, soil, topography, and land use of the drainage area upstream of the gully. Previous studies on vegetation-influencing factors often considered the topography and the soil conditions at the location of the vegetation (Florinsky
Table 4 Relationship between vegetation indices and soil water content on gully sidewall and gully bed. Position
Vegetation index
SWC1a 2
Gully sidewall
Gully bed
a
Vegetation cover Number of the plants Height of the plants Vegetation cover Number of the plants Height of the plants
Soil moisture in rainy season. Soil moisture in dry season. ⁎ significance reach 0.05 level. ⁎⁎ significance reach 0.01 level. b
SWC2b
R
P
R2
P
0.650⁎⁎ 0.666⁎⁎ 0.373⁎⁎ 0.140⁎ 0.096⁎ 0.038
0.000 0.000 0.000 0.013 0.043 0.211
0.073 0.096 0.030 0.030 0.006 0.081
0.095 0.055 0.295 0.268 0.631 0.064
and Kuryakova, 1996; Poulos et al., 2007; Pueyo and Beguería, 2007; Xu et al., 2008). However, the significant relationship between the vegetation distribution and the gully activity revealed that the vegetation distribution in the gully is also influenced by the rainfall and the erosion processes in upstream areas. The typical measurement of vegetation cover, which is generally recorded as the cover area in the projected area (2-D area) of the quadrats, may not be suitable for the analysis of a gully, particularly the active part of the gully. Although the gully sidewalls typically showed a large surface area, because the slope was close to 90°, the gully sidewalls on the 2-D area were quite small. According to the gully n1 DEM and field measurements, the analysis of the cliff sections (slope more than 70°) only took into account 11.2% of the total 2-D area of the active part, and the average vegetation cover of the rest quadrats in the gentle areas was as high as 28.1% (mainly grown on gully bed). If we evaluated the vegetation cover according to the proportion of the 2-D area of the cliff and flatter sections, the average vegetation cover of the active part of the gully would be 24.9%. This result certainly does not reflect the real vegetation conditions in the active part of the gully. The average cliff depth in the active parts was 5.43 m, as measured using tape, and the surface area of the cliff, as estimated by the 2-D area times the average depth, was 1030.20 m2, which took into account approximately 43% of the total surface area of the active part. If we consider the vegetation cover on the surface area, the vegetation cover of the active part decreased to 16.0%, which was very close to the average obtained for the quadrats in the active part of the gully (16.9%). These results demonstrate that the measurement of the vegetation cover in a gully, particularly in an active part of the gully with a steep long slope, which exhibits a small 2-D area but a large surface area, should be performed using the surface area instead of the 2-D area.
5. Conclusions The vegetation in 82 quadrats and nine points on the steep slope of gully n1 were investigated in 2012, and the average vegetation cover was observed to be approximately 40%. There was almost no vegetation grown on the cliff part of the gully sidewall, whereas in relatively gentle parts of the sidewall, the slope gradient and the slope aspect exhibited a weak correlation with the vegetation indices. The most important factor that impacted the vegetation distribution on the gully bed was the runoff: the average vegetation cover was 5.2% inside the runoff path, whereas the average vegetation cover increased to 56.3% outside the runoff path. The vegetation strongly influenced the soil moisture in the rainy season, whereas no relationship was observed during the dry season, which indicates that the water conservation of the vegetation in the Dry-hot Valley cannot last more than 1 month. The gully erosion processes also affected the vegetation. In the active part of the gully, the average vegetation cover was 16.9%, whereas the vegetation cover in the stable parts of the gully was 55.1%. Therefore, the vegetation distribution in the gully is not only related to the soil and the topography in the area where the vegetation grows but is also influenced by runoff processes upstream of the gully heads.
Y. Dong et al. / Catena 116 (2014) 60–67
Acknowledgements Financial support for this study was provided by the Youth Talent Team Program of IMHE, Chinese Academy of Sciences (SDSQB-201101), the National Natural Science Foundation of China (Grant No. 41201009), the Knowledge Innovation Program of the Chinese Academy of Sciences (KZCXZ-EW-QN317), the West Light Foundation of the Chinese Academy of Sciences, and the Foundation of Key Laboratory of Mountain Hazards and Earth Surface Processes, Chinese Academy of Sciences. This research was supported by the Key Laboratory of Mountain Hazards and Earth Surface Process, Chinese Academy of Sciences. The authors would like to thank Professor Zhang Jianhui for the valuable suggestion for this research, and also wish to thank Xiao Yang and Yang Donglin for their assistance in the fieldwork. The language of the article was improved by Elsevier Webshop.
References Alday, J.G., Marrs, R.H., Martinez-Ruiz, C., 2010. The importance of topography and climate on short-term revegetation of coal wastes in Spain. Ecol. Eng. 36, 579–585. Bennie, J., Hill, M.H., Baxter, R., Huntley, B., 2006. Influence of slope and aspect on longterm vegetation change in British chalk grasslands. J. Ecol. 94 (2), 355–368. Bocco, G., 1991. Gully erosion: processes and models. Prog. Phys. Geogr. 15 (4), 392–406. Cheng, A.Q., Zhang, D., Fan, J.R., Liu, G.C., 2011. Relationship between gullying development stage and collapse type in Yuanmou Dry-hot Valley area. Sci. Soil Water Conserv. 9 (4), 1–6 (in Chinese). Florinsky, I.V., Kuryakova, G.A., 1996. Influence of topography on some vegetation cover properties. Catena 27, 123–141. Jiao, J.Y., Zou, H.Y., Jia, Y.F., Wang, N., 2009. Research progress on the effects of soil erosion on vegetation. Acta Ecol. Sin. 29, 85–91. Leyland, J., Darby, S.E., 2008. An empirical–conceptual gully evolution model for channelled sea cliffs. Geomorphology 102, 419–434. Li, Y., Zhu, X.M., Tian, J.Y., 1991. Effectiveness of plant-roots to increase the antiscourability of soil on the Loess Plateau. Chin. Sci. Bull. 36 (12), 935–938. Martinez-Casasnovas, J.A., 2003. A spatial information technology approach for the mapping and quantification of gully erosion. Catena 50, 293–308. Miller, J.R., Joyce, L.A., Knight, R.L., Rudy, W., 1996. Forest roads and landscape structure in the southern Rocky Mountains. Landsc. Ecol. 11 (2), 115–127.
67
Munoz-Robles, C., Reid, N., Frazier, P., Tighe, M., Briggs, S.V., Wilson, B., 2010. Factors related to gully erosion in woody encroachment in south-eastern Australia. Catena 83, 148–157. Ostendorf, B., Reynolds, J.F., 1998. A model of arctic tundra vegetation derived from topographic gradients. Landsc. Ecol. 13 (3), 187–201. Osterkamp, W.R., Hupp, C.R., Stoffel, M., 2011. The interactions between vegetation and erosion: new directions for research at the interface of ecology and geomorphology. Earth Surf. Process. Landf. 37 (1), 23–36. Pearson, S.M., Turner, M.G., Drake, J.B., 1999. Landscape change and habitat availability in the southern Appalachian Highlands and Olympic Peninsula. Ecol. Appl. 9 (4), 1288–1304. Poesen, J., Nachtergaele, J., Verstraeten, G., 2003. Gully erosion and environmental change: importance and research needs. Catena 50, 91–133. Poulos, H.M., Taylor, A.H., Beaty, R.M., 2007. Environmental controls on dominance and diversity of woody plant species in a Madrean, Sky Island ecosystem, Arizona, USA. Plant Ecol. 193, 15–30. Pueyo, Y., Beguería, S., 2007. Modelling the rate of secondary succession after farmland abandonment in a Mediterranean mountain area. Landsc. Urban Plan. 83 (4), 245–254. Rey, F., 2003. Influence of vegetation distribution on sediment yield in forested marly gullies. Catena 50, 549–562. Solon, J., Degorski, M., Roo-Zielinska, E., 2007. Vegetation response to a topographical soil gradient. Catena 71, 309–320. Wang, X.D., Zhong, X.H., Liu, S.Z., Li, M.H., 2008a. A non-linear technique based on fractal method for describing gully-head changes associated with land-use in an arid environment in China. Catena 72, 106–112. Wang, Y.P., Shao, M.A., Zhang, X.C., 2008b. Soil moisture ecological environment of artificial vegetation on steep slope of loess region in North Shaanxi Province, China. Acta Ecol. Sin. 28 (8), 3769–3778. Wang, L., Wei, S.P., Horton, R., Shao, M.A., 2011. Effects of vegetation and slope aspect on water budget in the hill and gully region of the Loess Plateau of China. Catena 87, 90–100. Wu, Y.Q., Zheng, Q.H., Zhang, Y.G., Liu, B.Y., Cheng, H., Wang, Y.Z., 2008. Development of gullies and sediment production in the black soil region of northeastern China. Geomorphology 101, 683–691. Xiong, D.H., Long, Y., Yan, D.C., Lu, X.N., Ji, Z.H., Fang, H.D., 2009. Surface morphology of soil cracks in Yuanmou Dry-hot Valley Region, Southwest China. J. Mt. Sci. 6, 373–379. Xiong, D.H., Yan, D.C., Long, Y., Lu, X.N., Han, J.N., Han, X.Q., Shi, L.T., 2010. Simulation of morphological development of soil cracks in Yuanmou Dry-hot Valley Region, Southwest China. Chin. Geogr. Sci. 20 (2), 112–122. Xu, X.L., Ma, K.M., Fu, B.J., Song, C.J., Liu, W., 2008. Relationships between vegetation and soil and topography in a dry warm river valley, SW China. Catena 75, 138–145. Zhang, J.H., Tian, G., Li, Y., Lindstrom, M., 2003. Requirements for success of reforestation projects in a semiarid low-mountain region of the Jinsha River Basin, Southeastern China. Land Degrad. Dev. 13, 395–401.
Contents lists available at ScienceDirect
Catena journal homepage: www.elsevier.com/locate/catena
The distribution of and factors influencing the vegetation in a gully in the Dry-hot Valley of southwest China Yifan Dong a, Donghong Xiong a,⁎, Zheng'an Su a, Jiajia Li b, Dan Yang a, Liangtao Shi c, Gangcai Liu a a b c
Key Laboratory of Mountain Hazards and Earth Surface Processes, Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, 610041 Chengdu, China Department of Environmental Engineering, Chengdu University of Information Technology, 610225 Chengdu, China Institute of Tropical Eco-agricultural Sciences, Yunnan Academy of Agricultural Sciences, Yuanmou 651300, Yunnan, China
a r t i c l e
i n f o
Article history: Received 22 April 2013 Received in revised form 8 November 2013 Accepted 18 December 2013 Keywords: Gully Vegetation Soil Topography Dry-hot Valley
a b s t r a c t The Dry-hot Valley of the Jinsha River in southwest China is an ecologically fragile zone in which gully erosion is one of the most important environmental problems due to the high sediment yield from the gully. The vegetation, which impacts the erosion processes of the gully, is important to the ecological environment. In this study, we investigated the vegetation in gully n1 of Yuanmou County, which is a typical area of the Dry-hot Valley. A total of 82 vegetation quadrats on a relatively gentle section of gully n1 (slope was mostly less than 45°) and nine points on a steep slope (slope higher than 70°) were investigated in 2012. The vegetation indices, in addition to the soil conditions and topography, were measured through field investigations and the gully Digital Elevation Model (DEM). On the gully sidewall, almost no vegetation grew in the steep slope, whereas the slope gradient and the slope aspect exhibited a weak relationship with the vegetation indices when the slope was less than 45°. On the gully bed, the runoff path was the most important factor that affected the vegetation. The average vegetation cover inside the runoff path was 5.2%, whereas the average vegetation cover outside the runoff path increased to 56.3%. The vegetation strongly impacted the soil moisture during the rainy season. However, this relationship was not observed during the dry season, which indicates that the water conservation effects exerted by the vegetation cannot last a long time in the Dry-hot Valley. The active part of gully n1 has a vegetation cover (average 16.9%) that is significantly lower compared with the stable parts of the gully (55.1%). This finding indicates that the vegetation in the gully is not only impacted by the soil and the topography in the area of vegetation growth but also influenced by the runoff processes upstream of the gully heads. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Vegetation, which can purify air, filter water, and protect soil, is often impacted by the soil condition and topography in a region (Florinsky and Kuryakova, 1996; Poulos et al., 2007; Xu et al., 2008). The topography can influence the vegetation by affecting the development of roots, the microclimate, and the solar radiation (Alday et al., 2010; Bennie et al., 2006; Li et al., 1991). The vegetation is also affected by various soil properties, such as soil moisture, acidity (pH), and nutrient content. In addition, the growth of vegetation may affect the topography and soil properties through various processes, including respiration, hydrologic control, and soil conservation (Jiao et al., 2009; Osterkamp et al., 2011). The relationship between vegetation, topography, and soil, which may change at different spatial scales, is an important subject of ecological and geographic studies (Solon et al., 2007; Xu et al., 2008).
⁎ Corresponding author at: Key Laboratory of Mountain Hazards and EarthSurface Process, Institute of Mountain Hazards and Environment,Chinese Academy of Sciences, 610041 Chengdu, China. Tel.: +86 28 85259762; fax: +86 28 85222258. E-mail address: [email protected] (D. Xiong). 0341-8162/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.catena.2013.12.009
Gully erosion is an important type of soil erosion, and gully itself is a main source of sediments, which contributes 10% to 94% of the total sediments in different areas of the world (Poesen et al., 2003). The vegetation has beneficial effects in controlling of gully erosion (Munoz-Robles et al., 2010; Rey, 2003), which could result in the reinforcement of steep gully sidewalls and the interception of sediment transport processes on the gully beds. The topography of the gully investigated in this study was quite different from that reported in previous studies. The slope gradient of the vegetation were often reported to be below 30° in previous studies (Solon et al., 2007; Wang et al., 2011; Xu et al., 2008), while slopes reached 35° to 45° thus were described as steep slopes (Wang et al., 2008b). The slope gradient of the gully changed significantly: the gully sidewall was often close to or even more than 90°, whereas the gully bed was often below 5°. The soil composition of the gully was often complicated because it cut deeply into the earth surface, which made the soil type to vary in the longitudinal direction. Thus, the special topography and soil conditions of the gully create a complex environment for vegetation growth. Furthermore, the erosion processes that occurred in different parts of the gully also varied (Leyland and Darby, 2008; Martinez-Casasnovas, 2003; Poesen et al., 2003), and the vegetation conditions, which are strongly correlated with erosion processes, may also be different. However, few studies have focused on
Y. Dong et al. / Catena 116 (2014) 60–67
the vegetation distribution and the factors that influence this distribution in a gully. The Dry-hot Valley of the Jinsha River is an ecologically fragile zone in southwest China that covers an area of approximately 1.2 × 104 km2 (Xiong et al., 2010), and gully erosion is a serious problem in this area. According to previous studies, the distribution density of most gully area ranged from 3 to 5 km/km2, with the highest density being 7.4 km/km2 (Cheng et al., 2011; Wang et al., 2008a). This distribution results in a high sediment yield and a rapid degradation of land. The revegetation of the gullies is very important for the improvement of the ecological environment of this region, although the gullies in this region are often large in scale with a deep incision, steep gully sidewall and a complex soil composition. These special characteristics may influence the vegetation distribution and growth in the gully. The aim of this study was to investigate the vegetation conditions in conjunction with the topography and the soil conditions in the gully and to analyse the distribution of vegetation in different parts of the gully. 2. Materials and methods 2.1. Study area The Yuanmou Dry-hot Valley, which covers an area of more than 2000 km2, is located in the upper and middle reaches of the Jinsha River (an important tributary of the Yangtze River) in Yunnan Province (Fig. 1a) in southwest China (Xiong et al., 2009). The region exhibits a typical southern subtropical climate with an annual average temperature of 21.8 °C. The annual precipitation is 615 mm, and the annual potential evaporation is 3569 mm, which results in the extreme aridness of this region (Wang et al., 2008a). The rainy season, which often lasts from June to early October, accounts for more than 85% of the annual precipitation. Gullies often developed at the quaternary fluvial–lacustrine deposits with a loose structure at an elevation of in the range of 1000 to 1350 m above sea level, and the main soil types at the surface of the region were dry red soil (classified as Ustic Ferrisols in Chinese Taxonomy) and vertisols. Dry red soil has a high sand content (often above 50%) and contains iron and manganese, whereas vertisols are mainly composed of clay (often over 50%) with a strong expansibility. In addition, a sandy soil layer, which could easily be eroded by water, often develops under these two types of soil, and a large amount of sediments are deposited in the middle and downstream of the gully bed. The vegetation in the gully is dominated by the herbs Heteropogon and Bothriochloa pertusa, and a few Dodonaea viscosa (L.) Jacg. shrubs grow in some parts of the gully (Zhang et al., 2003). 2.2. Quadrat investigation of the vegetation We randomly selected 82 1 m × 1 m quadrats in gully n1 (Fig. 1b) at the beginning of the rainy season (June 22th) in 2012, which was the very next day after a three-day period of continuous rainfall (about 15 h after the rainfall stopped). The vegetation indices, including the plant species, the percentage of vegetation cover, the number of plants, and the height of the plants, of each quadrat were determined. According to previous investigation and studies, the length of the herb roots in gully area was mainly ranged between 5 and 30 cm. So a soil depth of 0 to 30 cm was sampled in each quadrat which was mainly concerned with the vegetation condition. A total of 82 samples were oven-dried at 105 °C for 12 h to test the soil moisture by mass in the rainy season. The location of each quadrat was recorded by highprecision RTK GPS (Trimble R8, the horizontal and vertical positioning accuracy were 1 cm ± 2 ppm and 2 cm ± 2 ppm, respectively). During the dry season (November 6th, 46 days without any precipitation) in 2012, we used the lofting function of the RTK GPS to find the locations of the quadrats investigated in June; the navigation accuracy reached the centimetre level. The vegetation indices and the soil moisture were measured in the dry season to obtain the changes in the
61
vegetation conditions and the soil moisture in the same quadrats. In addition, the vegetation conditions at nine points on the steep gully sidewall (slope over 70°) of gully n1 were also investigated. 2.3. Gully topography measurement and data processing Gully n1 was located in the valley floor of a small catchment in Yuanmou (Fig. 1c) and was measured using an RTK GPS in April 2012. A total of 11,280 high-precision topographic points were monitored along the boundary of the gully sides, the edge of the gully bottom where the gully sidewall changes abruptly into a flatter gully bed, and randomly distributed on the sidewall and the bottom of the gully. These data were transferred to ArcGIS 9.3, used to create Delaunay triangulated irregular networks (Tins) using the 3D Analyst tool, and then converted into a DEM grid with 1-m cell size (Wu et al., 2008). The elevation and depth of each cell could be extracted from the DEM grid. The slope gradient and the slope aspect of each cell could be calculated using the Spatial Analyst tool of ArcGIS, and the topography information of each quadrat could be obtained according to its location cell in the DEM grid. According to the results, the area and volume of gully n1 were 16,939.4 m2 and 124,514.5 m3 respectively. The mean depth of the gully was 7.6 m, and maximum depth was 15.9 m. The elevation ranged between 1060.4 m and 1090.9 m, and the slope gradient ranged from 0.45° to 87.8° with an average of 34.3°. 3. Results 3.1. Vegetation, soil, and topography of the quadrats in the gully Herbs were observed to be the major vegetation type in the gully. All of the quadrats with vegetation contained herb plants: 94.7% of the quadrats contained Heteropogon and 37.3% of the quadrats contained B. pertusa. In addition, 4.9% of the quadrats contained shrubs, and the species was D. viscosa (L.) Jacg. The vegetation conditions in November were slightly better than those in June because the herbs in the gully, which had just begun to grow in June and were thus fully grown after the rainy season, were tolerant to drought and had therefore not completely wilted in November (after more than 1 month without precipitation). A total of 12 of the 82 quadrats did not have any vegetation in June, and 5 of these 12 quadrats contained herb plants in November; however, the average vegetation cover in these quadrats was 3.6%, and only one of the five quadrats exhibited a vegetation cover of more than 5%. The average vegetation cover in the 82 quadrats increased by 6.0% from June to November. Although the number of plants did not significantly change, the height of the plants increased by 38.6% (Table 1). The height of the plants enhanced in November, but the leaves of the herbs began to wilt and the colour of the herbs was mostly turned into yellow already. The high coefficient of variation (Cv) of the vegetation cover revealed that the vegetation distribution varied between different locations of the gully, likely due to the complex topography and soil conditions. Because the growing season of the plants was mainly during the rainy season and the vegetation was not fully grown at that point, the vegetation indices in the rest of the manuscript indicate the results obtained in the dry season (November) unless otherwise noted. The analysis of the 82 quadrats revealed a significant relationship between the vegetation indices, including the vegetation cover and the number and the height of the plants (Fig. 2). The soil water content (by mass) in June (SWC1) was significantly higher than that in November (SWC2); the average SWC1 was 7.45%, whereas the SWC2 was only 3.78%. The slope of the quadrats ranged from 1.25° to 61.35° with an average of 22.05°. The slope aspect was converted using the cosine function to a range from − 1 to 1, which indicates south to north (Xu et al., 2008), and the average was − 0.27. The depths of the quadrats ranged from 1.28 m to 15.50 m, and the elevation ranged from 1060.70 m to 1088.44 m (Table 2).
62
Y. Dong et al. / Catena 116 (2014) 60–67
Y. Dong et al. / Catena 116 (2014) 60–67
63
Table 1 The vegetation condition of 82 quadrats in June (during the rainy season) and November (during the dry season) in 2012. Vegetation cover (%)
Average Stdev Cv
Number of plants
Height of plants (cm)
June
November
June
November
June
November
37.0 32.5 88.0%
43.0 35.6 82.9%
6.1 4.3 70.2%
6.3 4.1 67.4%
29.8 20.0 67.1%
41.3 22.8 55.2%
homogeneous slope gradient of often below the response angle (approximately 45° with vegetation). The vegetation distributions in the two types of sidewalls are different and clearly affected by the slope gradient. There were nine points analysed on the cliff part of sidewall type A, where the slope gradient ranged from 71.53° to 83.35°, and there was no vegetation cover because it was difficult for the plant seeds to develop in the surface soil of the cliff. The vegetation was mainly distributed on the slope wash part of the type A and type B sidewalls, which often present slope gradients of less than 45°, and the vegetation cover of the 39 sidewall quadrats ranged from 0 to 96% with an average of 49.0% and a coefficient of variation (Cv) of 78.1%. A total of 43 quadrats were investigated on the gully bed, and the most important factor that impacted the vegetation distribution on the gully bed was whether its location was inside the runoff path. The earth surface of runoff path was often scoured by strong concentrate flow, which could sweep away the plant seeds and reduce the soil water conservation capacity, and finally result to very low vegetation cover on the runoff path. The investigation in June was conducted right after a three-day period of continuous rainfall, and the runoff path could be clearly identified; as a result, it was observed that 17 of the investigated quadrats were located inside the runoff path. We also used the Hydrology tool of ArcGIS to create the runoff path according to the gully DEM, and 16 of the 17 quadrats were in or one cell adjacent to the calculated runoff path. The vegetation cover of these quadrats in November ranged from 0 to 15%; 14 quadrats exhibited a vegetation cover of less than 10%, and the average was only 5.2%. In contrast, the vegetation cover in the other gully bed quadrats ranged from 11% to 98% with an average of 56.3%. Distinct differences in the vegetation distribution could be identified at the boundary of the runoff path: a high vegetation density with substantial growth was observed outside the boundary, whereas almost no vegetation cover was observed inside the runoff path (Fig. 4). Fig. 2. Relationships between vegetation indices including percentage of vegetation cover (a), number and height of the plants (b).
4. Discussion 4.1. The relationship between gully vegetation and topography
3.2. Vegetation on the gully sidewall and the gully bed Gullies can often be divided into the gully sidewall and the gully bed according to their position in the landscape (Bocco, 1991). The sidewall often has a steep slope with different soil types at different depths. In contrast, the gully bed has a relatively gentle slope gradient at the base of the sidewall, and the surface of the gully bed is often covered with sediments that are deposited by the runoff. The various topographies and soil types result in different environments for vegetation in the gully. In general, there are two types of gully sidewalls in our study area (Fig. 3). Type A often exhibits a steep cliff (with a slope of over 70°) at the upper and middle part and a slope wash (below 45°) at the bottom of the sidewall, whereas type B has a long slope with a relatively
Previous studies have demonstrated that the slope gradient, the slope aspect, and the elevation are the most important factors that impact a vegetation habitat (Miller et al., 1996; Ostendorf and Reynold, 1998; Pearson et al., 1999). In contrast, the conditions in a gully are more complicated. The slope gradient was observed to be the dominant factor that determined vegetation growth on the cliff part of the gully sidewall, which was mostly a type A slope. Regardless of the type of soil and the moisture and nutrient contents of the soil, almost no vegetation grew on the cliff part of the gully sidewall. When the slope gradient was less than 45°, the vegetation indices exhibited some correlation with the slope gradient and the slope aspect, which could impact the survival and development of the vegetation as a result of moisture migration and sunlight conditions. However, this relationship was not very strong because the R2 ranged from 0.10 to 0.15 (Table 3). The
Fig. 1. The location of the study area, the distribution of the vegetation quadrats, and the general situation in gully n1. a. Location of the Yuanmou Dry-hot Valley. b. Distribution of the vegetation quadrats in gully n1 of the DEM. c. Boundary of gully n1.
64
Y. Dong et al. / Catena 116 (2014) 60–67
Table 2 Basic information of the soil and topography of the 82 quadrats. Quadrat ID
Position
SWC1a
SWC2b
Slope gradient (°)
YF001 YF002 YF003 YF004 YF005 YF006 YF007 YF008 YF009 YF010 YF011 YF012 YF013 YF014 YF015 YF016 YF017 YF018 YF019 YF020 YF021 YF022 YF023 YF024 YF025 YF026 YF027 YF028 YF029 YF030 YF031 YF032 YF033 YF034 YF035 YF036 YF037 YF038 YF039 YF040 YF041 YF042 YF043 YF044 YF045 YF046 YF047 YF048 YF049 YF050 YF051 YF052 YF053 YF054 YF055 YF056 YF057 YF058 YF059 YF060 YF061 YF062 YF063 YF064 YF065 YF066 YF067 YF068 YF069 YF070 YF071 YF072 YF073 YF074 YF075
Gully sidewall Gully sidewall Gully sidewall Gully sidewall Gully bed Gully bedc Gully sidewall Gully bedc Gully sidewall Gully sidewall Gully sidewall Gully bedc Gully sidewall Gully sidewall Gully bed Gully bedc Gully bedc Gully bed Gully bed Gully sidewall Gully bed Gully bed Gully sidewall Gully bedc Gully bed Gully bedc Gully sidewall Gully bed Gully sidewall Gully bedc Gully sidewall Gully bed Gully bed Gully bedc Gully bed Gully bed Gully bedc Gully bedc Gully sidewall Gully sidewall Gully bedc Gully bed Gully sidewall Gully bedc Gully sidewall Gully bedc Gully sidewall Gully bedc Gully bed Gully sidewall Gully bed Gully sidewall Gully sidewall Gully bed Gully sidewall Gully sidewall Gully sidewall Gully sidewall Gully bed Gully bedc Gully sidewall Gully sidewall Gully sidewall Gully sidewall Gully bed Gully bed Gully bed Gully bed Gully sidewall Gully bed Gully bedc Gully bed Gully sidewall Gully sidewall Gully sidewall
2.36% 8.06% 9.79% 4.03% 4.99% 7.42% 5.18% 8.59% 4.89% 6.22% 9.81% 12.45% 8.98% 5.23% 9.28% 6.65% 9.63% 10.94% 5.49% 3.62% 9.77% 8.18% 4.56% 5.82% 2.64% 5.98% 5.44% 9.52% 12.05% 7.92% 3.82% 11.44% 4.93% 7.42% 9.83% 8.44% 5.65% 5.19% 10.31% 5.94% 6.32% 7.39% 6.02% 5.94% 7.59% 4.68% 8.30% 4.88% 8.33% 7.47% 4.13% 4.11% 4.84% 7.16% 7.00% 9.81% 8.47% 8.88% 12.87% 4.79% 8.79% 4.69% 7.67% 3.88% 9.35% 12.85% 5.84% 9.26% 6.53% 5.68% 8.40% 10.30% 11.61% 10.84% 5.48%
2.37% 2.54% 2.69% 1.29% 2.91% 3.36% 1.14% 2.48% 2.31% 4.93% 3.15% 3.82% 2.19% 3.92% 2.28% 1.85% 4.73% 1.84% 1.12% 4.38% 3.17% 4.34% 2.28% 7.52% 2.03% 2.27% 4.89% 7.23% 5.86% 4.33% 11.08% 4.15% 5.59% 5.56% 1.70% 7.11% 2.97% 2.65% 2.77% 3.55% 3.75% 4.33% 2.70% 1.68% 2.45% 4.52% 2.14% 3.03% 8.19% 2.17% 3.54% 1.37% 1.21% 2.60% 3.23% 4.90% 4.34% 4.53% 4.33% 3.01% 3.94% 2.29% 3.20% 1.58% 2.20% 4.44% 3.92% 2.86% 7.55% 1.72% 2.28% 6.30% 7.36% 1.87% 9.49%
18.86 17.11 20.97 18.08 17.12 23.32 45.84 5.72 21.37 29.39 37.31 20.21 33.21 36.06 12.41 17.94 6.20 7.69 10.64 55.92 35.02 31.14 61.35 56.85 15.71 1.25 21.08 27.52 30.99 40.34 48.49 16.62 15.78 14.21 8.34 8.98 9.22 5.15 19.52 19.77 14.09 20.89 19.12 12.56 31.90 6.31 11.99 1.25 8.38 30.24 20.62 36.19 13.08 17.59 30.92 14.35 12.47 31.51 5.36 2.13 19.83 31.66 44.02 40.14 24.11 11.26 4.72 5.08 41.27 16.47 11.56 22.26 38.85 21.97 31.19
Slope aspect −0.24 −0.14 −0.97 −1.00 0.12 −0.10 0.24 −0.85 −0.29 −0.24 0.60 −0.74 −1.00 −1.00 −0.99 0.85 −0.98 0.42 −1.00 −0.99 0.72 −1.00 −1.00 −0.41 0.41 −0.09 0.53 0.44 0.84 0.95 0.45 −0.09 0.07 0.17 −0.66 −1.00 −0.10 −0.19 0.59 −0.96 −0.83 0.10 −0.29 −0.97 −0.42 0.95 −0.98 −0.29 −0.98 −0.75 −1.00 −0.80 −0.96 −0.29 −0.07 −0.31 −0.45 0.63 −0.96 0.77 0.97 0.90 0.39 −0.54 −0.95 −0.96 −0.95 −0.63 0.97 −0.87 0.09 0.10 0.87 −0.28 −0.54
Depth (m)
Elevation (m)
Main soil type
5.87 10.88 13.02 11.11 9.25 7.15 5.80 3.33 5.45 3.81 10.85 13.57 8.36 5.63 13.34 14.57 14.85 14.10 11.88 8.07 10.38 11.73 5.92 6.24 9.67 5.22 1.28 3.38 2.83 3.63 4.33 4.51 11.55 14.58 14.60 14.87 15.29 15.50 13.27 10.65 9.35 5.33 7.07 13.18 14.32 14.01 12.79 13.68 12.93 9.78 11.03 8.83 8.41 6.83 7.17 3.68 5.82 8.18 12.92 11.12 11.25 9.19 6.83 4.26 8.39 13.83 15.19 14.75 8.62 8.37 3.78 3.54 5.20 7.72 2.08
1075.13 1069.11 1066.44 1069.26 1071.95 1073.42 1075.67 1077.91 1075.93 1078.20 1069.74 1067.32 1072.81 1075.12 1067.82 1066.12 1066.90 1068.08 1069.89 1073.31 1071.48 1073.26 1080.25 1081.03 1076.37 1083.31 1088.44 1086.67 1082.87 1083.40 1080.20 1078.06 1069.16 1065.31 1065.71 1064.51 1068.96 1063.30 1064.80 1068.43 1070.49 1073.57 1069.85 1065.08 1063.33 1062.16 1063.22 1061.24 1060.86 1064.32 1063.84 1067.21 1066.61 1068.66 1072.48 1070.87 1067.37 1064.68 1060.79 1060.70 1063.47 1065.67 1068.51 1072.59 1067.45 1060.90 1062.49 1063.68 1070.16 1072.62 1081.93 1081.69 1077.55 1074.28 1081.58
Dry red soil Dry red soil Dry red soil Dry red soil Dry red soil Sediment Sediment Sediment Dry red soil Dry red soil Sandy soil Sandy soil Dry red soil Dry red soil Sediment Sediment Sediment Sediment Sediment Dry red soil Sandy soil Sediment Dry red soil Dry red soil Vertisols Sediment Dry red soil Vertisols Vertisols Sediment Vertisols Dry red soil Dry red soil Sediment Sediment Sediment Sediment Sediment Sandy soil Sandy soil Sediment Dry red soil Sandy soil Sediment Sandy soil Sediment Sandy soil Sediment Sediment Dry red soil Sediment Vertisols Sandy soil Sediment Sandy soil Vertisols Vertisols Vertisols Sediment Sediment Vertisols Vertisols Vertisols Dry red soil Vertisols Sediment Sediment Sediment Vertisols Vertisols Sediment Dry red soil Vertisols Vertisols Vertisols
Y. Dong et al. / Catena 116 (2014) 60–67
65
Table 2 (continued) Quadrat ID
Position
SWC1a
SWC2b
Slope gradient (°)
Slope aspect
Depth (m)
Elevation (m)
Main soil type
YF076 YF077 YF078 YF079 YF080 YF081 YF082 Average Stdev Cv
Gully sidewall Gully sidewall Gully bed Gully sidewall Gully sidewall Gully bed Gully bed – – –
13.52% 11.89% 9.43% 4.60% 2.22% 9.57% 11.22% 7.45% 0.03 36.25%
9.23% 8.57% 3.05% 1.63% 2.57% 3.86% 5.95% 3.78% 0.02 56.14%
23.67 34.46 14.42 20.84 28.67 23.29 14.98 22.05 13.30 60.33%
−0.42 −0.41 −0.12 0.98 −0.09 −0.60 −0.91 −0.24 0.66 −275.54%
4.34 3.14 8.29 7.12 8.08 13.60 2.52 8.99 4.02 44.71%
1079.28 1080.33 1073.42 1073.06 1071.43 1064.59 1081.48 1071.11 6.81 0.64%
Vertisols Dry red soil Vertisols Vertisols Vertisols Vertisols Vertisols – – –
a b c
Soil water content in June. Soil water content in November. Inside the runoff path.
vertical zonality was not distinct in gully n1, i.e., the depth/elevation exhibited a poor relationship with the vegetation indices (Table 3). The depth of the sidewall quadrats ranged from 1.28 to 14.32 m, whereas the elevation of the gully ranged from 1063.22 to 1088.44 m. The amplitude of the variation in the altitude direction was not sufficient to significantly affect the vegetation in the gully. The runoff path was a unique influencing factor on the vegetation distribution along the horizontal direction of the gully bed. Gully n1 was located in the valley of the catchment, which accumulated runoff both on the slope of the catchment and inside the gully itself (Fig. 1c). A large amount of runoff flowed from upstream of the head of gully n1 to the gully end. This path exhibits elevation changes of
approximately 30 m. The potential energy of the runoff was converted to kinetic energy, which leads to a high scouring force on the earth surface of the runoff path. The high-power runoff is able to sweep plant seeds and seedlings away from the soil and thus stop their growth. In addition, the turbulent runoff could also wash away the silt and clay content from the soil, worsening the water conservation properties of the soil on the runoff path. The average soil water content (SWC1) of the runoff path quadrats was 21.67% lower than the average for other places on the gully bed and 1.73% lower than the average in the sidewall quadrats over 16 h after the rainfall stopped, which are often considered to exhibit a lower soil moisture than the gully bed (Table 2). The low survival rate of plants and the poor water retention of the soil resulted in a bare earth surface on the runoff path. 4.2. The relationship between the gully vegetation and the soil moisture Soil moisture is considered to be a vital factor for the survival of vegetation in Dry-hot Valley (Zhang et al., 2003). However, more than 60% of the quadrats that were investigated exhibited a vegetation cover of more than 20%, and the average height of the herb plants was greater than 40 cm, which indicated that the vegetation was fully grown and thus began to impact the soil moisture during the rainy season. The vegetation indices on the gully sidewall, including the vegetation cover and the number and the height of the plants, exhibited a strong correlation with the soil water content in June (SWC1). The vegetation on the gully bed presented results that were similar to those obtained for the sidewall, but the relationship was weak due to the quadrats inside the runoff path (Table 4). In contrast, a weak relationship was observed between the vegetation indices and the soil water content in November (SWC2), which
Fig. 3. Two main types of gully sidewalls in the Yuanmou Dry-hot Valley. Nearly no vegetation grew on the cliff part of the type A sidewalls, which had a depth that was often more than 10 m and a slope gradient that was often higher than 70°.
Fig. 4. The distinct vegetation distributions inside and outside the boundary of the runoff path.
66
Y. Dong et al. / Catena 116 (2014) 60–67
Table 3 Relationship between vegetation indices and topography on gully sidewall and gully bed. Position
Gully sidewall
Gully bed
Vegetation index
Slope gradient
Vegetation cover Number of the plants Height of the plants Vegetation cover Number of the plants Height of the plants
Slope aspect
Depth
Elevation
R2
P
R2
P
R2
P
R2
P
0.092 0.112⁎ 0.146⁎ 0.027 0.016 0.025
0.060 0.037 0.016 0.292 0.417 0.314
0.110⁎ 0.077 0.125⁎ 0.041 0.068 0.065
0.038 0.087 0.027 0.194 0.092 0.099
0.038 0.104 0.004 0.041 0.029 0.046
0.231 0.065 0.223 0.162 0.273 0.168
0.084 0.106 0.036 0.019 0.015 0.040
0.073 0.063 0.247 0.382 0.430 0.202
⁎ Significance reach 0.05 level.
was mainly due to the slight variation in the SWC2. The soil water content in 69 of the 82 quadrats was less than 5%, whereas only one sample exhibited a soil water content that was higher than 10% in November. The results indicate that, although the vegetation was able to conserve water by intercepting direct sunshine through its aboveground parts and by adsorbing soil water through the plant roots, the effect of water conservation would not last a long time in the Dry-hot Valley. After more than 1 month without water, the soil moisture tended to be uniform throughout the gully. 4.3. The variations in the vegetation in the active and the stable parts of the gully The activity of the erosion processes in different parts of the gully impacted the vegetation distribution in the gully. The active part of gully n1 consisted of five active gully heads that developed quickly (Fig. 1b). Four of the five gully heads were measured by RTK GPS in 2011 and 2012. The average headcut retreat rate was 0.28 m, and the highest headcut retreat rate was higher than 0.6 m in 1 year. In contrast, other parts of the gully were relatively stable and did not exhibit significant changes (Fig. 1b). Ten of the studied quadrats and five of the points analysed in the steep slope were located in the active part of the gully. The average vegetation cover in these quadrants and points was 16.9%, and 7 of the 15 locations did not have any vegetation. In contrast, the average vegetation cover in the stable part of the gully was 43.3%, and 36 of the 77 quadrats exhibited a vegetation cover higher than 50%. A total of 14 of the 17 quadrats along the runoff path, which exhibited an average vegetation cover of 5.1%, were located in the stable part of gully n1. However, these quadrats suffered from strong runoff scouring during the rainy season and can thus not be described as stable. After excluding these quadrats, the average vegetation cover of the stable part of the gully was enhanced to 55.1%. The activity of the gully erosion was mainly impacted by the intensity of the runoff, which was determined by the precipitation, soil, topography, and land use of the drainage area upstream of the gully. Previous studies on vegetation-influencing factors often considered the topography and the soil conditions at the location of the vegetation (Florinsky
Table 4 Relationship between vegetation indices and soil water content on gully sidewall and gully bed. Position
Vegetation index
SWC1a 2
Gully sidewall
Gully bed
a
Vegetation cover Number of the plants Height of the plants Vegetation cover Number of the plants Height of the plants
Soil moisture in rainy season. Soil moisture in dry season. ⁎ significance reach 0.05 level. ⁎⁎ significance reach 0.01 level. b
SWC2b
R
P
R2
P
0.650⁎⁎ 0.666⁎⁎ 0.373⁎⁎ 0.140⁎ 0.096⁎ 0.038
0.000 0.000 0.000 0.013 0.043 0.211
0.073 0.096 0.030 0.030 0.006 0.081
0.095 0.055 0.295 0.268 0.631 0.064
and Kuryakova, 1996; Poulos et al., 2007; Pueyo and Beguería, 2007; Xu et al., 2008). However, the significant relationship between the vegetation distribution and the gully activity revealed that the vegetation distribution in the gully is also influenced by the rainfall and the erosion processes in upstream areas. The typical measurement of vegetation cover, which is generally recorded as the cover area in the projected area (2-D area) of the quadrats, may not be suitable for the analysis of a gully, particularly the active part of the gully. Although the gully sidewalls typically showed a large surface area, because the slope was close to 90°, the gully sidewalls on the 2-D area were quite small. According to the gully n1 DEM and field measurements, the analysis of the cliff sections (slope more than 70°) only took into account 11.2% of the total 2-D area of the active part, and the average vegetation cover of the rest quadrats in the gentle areas was as high as 28.1% (mainly grown on gully bed). If we evaluated the vegetation cover according to the proportion of the 2-D area of the cliff and flatter sections, the average vegetation cover of the active part of the gully would be 24.9%. This result certainly does not reflect the real vegetation conditions in the active part of the gully. The average cliff depth in the active parts was 5.43 m, as measured using tape, and the surface area of the cliff, as estimated by the 2-D area times the average depth, was 1030.20 m2, which took into account approximately 43% of the total surface area of the active part. If we consider the vegetation cover on the surface area, the vegetation cover of the active part decreased to 16.0%, which was very close to the average obtained for the quadrats in the active part of the gully (16.9%). These results demonstrate that the measurement of the vegetation cover in a gully, particularly in an active part of the gully with a steep long slope, which exhibits a small 2-D area but a large surface area, should be performed using the surface area instead of the 2-D area.
5. Conclusions The vegetation in 82 quadrats and nine points on the steep slope of gully n1 were investigated in 2012, and the average vegetation cover was observed to be approximately 40%. There was almost no vegetation grown on the cliff part of the gully sidewall, whereas in relatively gentle parts of the sidewall, the slope gradient and the slope aspect exhibited a weak correlation with the vegetation indices. The most important factor that impacted the vegetation distribution on the gully bed was the runoff: the average vegetation cover was 5.2% inside the runoff path, whereas the average vegetation cover increased to 56.3% outside the runoff path. The vegetation strongly influenced the soil moisture in the rainy season, whereas no relationship was observed during the dry season, which indicates that the water conservation of the vegetation in the Dry-hot Valley cannot last more than 1 month. The gully erosion processes also affected the vegetation. In the active part of the gully, the average vegetation cover was 16.9%, whereas the vegetation cover in the stable parts of the gully was 55.1%. Therefore, the vegetation distribution in the gully is not only related to the soil and the topography in the area where the vegetation grows but is also influenced by runoff processes upstream of the gully heads.
Y. Dong et al. / Catena 116 (2014) 60–67
Acknowledgements Financial support for this study was provided by the Youth Talent Team Program of IMHE, Chinese Academy of Sciences (SDSQB-201101), the National Natural Science Foundation of China (Grant No. 41201009), the Knowledge Innovation Program of the Chinese Academy of Sciences (KZCXZ-EW-QN317), the West Light Foundation of the Chinese Academy of Sciences, and the Foundation of Key Laboratory of Mountain Hazards and Earth Surface Processes, Chinese Academy of Sciences. This research was supported by the Key Laboratory of Mountain Hazards and Earth Surface Process, Chinese Academy of Sciences. The authors would like to thank Professor Zhang Jianhui for the valuable suggestion for this research, and also wish to thank Xiao Yang and Yang Donglin for their assistance in the fieldwork. The language of the article was improved by Elsevier Webshop.
References Alday, J.G., Marrs, R.H., Martinez-Ruiz, C., 2010. The importance of topography and climate on short-term revegetation of coal wastes in Spain. Ecol. Eng. 36, 579–585. Bennie, J., Hill, M.H., Baxter, R., Huntley, B., 2006. Influence of slope and aspect on longterm vegetation change in British chalk grasslands. J. Ecol. 94 (2), 355–368. Bocco, G., 1991. Gully erosion: processes and models. Prog. Phys. Geogr. 15 (4), 392–406. Cheng, A.Q., Zhang, D., Fan, J.R., Liu, G.C., 2011. Relationship between gullying development stage and collapse type in Yuanmou Dry-hot Valley area. Sci. Soil Water Conserv. 9 (4), 1–6 (in Chinese). Florinsky, I.V., Kuryakova, G.A., 1996. Influence of topography on some vegetation cover properties. Catena 27, 123–141. Jiao, J.Y., Zou, H.Y., Jia, Y.F., Wang, N., 2009. Research progress on the effects of soil erosion on vegetation. Acta Ecol. Sin. 29, 85–91. Leyland, J., Darby, S.E., 2008. An empirical–conceptual gully evolution model for channelled sea cliffs. Geomorphology 102, 419–434. Li, Y., Zhu, X.M., Tian, J.Y., 1991. Effectiveness of plant-roots to increase the antiscourability of soil on the Loess Plateau. Chin. Sci. Bull. 36 (12), 935–938. Martinez-Casasnovas, J.A., 2003. A spatial information technology approach for the mapping and quantification of gully erosion. Catena 50, 293–308. Miller, J.R., Joyce, L.A., Knight, R.L., Rudy, W., 1996. Forest roads and landscape structure in the southern Rocky Mountains. Landsc. Ecol. 11 (2), 115–127.
67
Munoz-Robles, C., Reid, N., Frazier, P., Tighe, M., Briggs, S.V., Wilson, B., 2010. Factors related to gully erosion in woody encroachment in south-eastern Australia. Catena 83, 148–157. Ostendorf, B., Reynolds, J.F., 1998. A model of arctic tundra vegetation derived from topographic gradients. Landsc. Ecol. 13 (3), 187–201. Osterkamp, W.R., Hupp, C.R., Stoffel, M., 2011. The interactions between vegetation and erosion: new directions for research at the interface of ecology and geomorphology. Earth Surf. Process. Landf. 37 (1), 23–36. Pearson, S.M., Turner, M.G., Drake, J.B., 1999. Landscape change and habitat availability in the southern Appalachian Highlands and Olympic Peninsula. Ecol. Appl. 9 (4), 1288–1304. Poesen, J., Nachtergaele, J., Verstraeten, G., 2003. Gully erosion and environmental change: importance and research needs. Catena 50, 91–133. Poulos, H.M., Taylor, A.H., Beaty, R.M., 2007. Environmental controls on dominance and diversity of woody plant species in a Madrean, Sky Island ecosystem, Arizona, USA. Plant Ecol. 193, 15–30. Pueyo, Y., Beguería, S., 2007. Modelling the rate of secondary succession after farmland abandonment in a Mediterranean mountain area. Landsc. Urban Plan. 83 (4), 245–254. Rey, F., 2003. Influence of vegetation distribution on sediment yield in forested marly gullies. Catena 50, 549–562. Solon, J., Degorski, M., Roo-Zielinska, E., 2007. Vegetation response to a topographical soil gradient. Catena 71, 309–320. Wang, X.D., Zhong, X.H., Liu, S.Z., Li, M.H., 2008a. A non-linear technique based on fractal method for describing gully-head changes associated with land-use in an arid environment in China. Catena 72, 106–112. Wang, Y.P., Shao, M.A., Zhang, X.C., 2008b. Soil moisture ecological environment of artificial vegetation on steep slope of loess region in North Shaanxi Province, China. Acta Ecol. Sin. 28 (8), 3769–3778. Wang, L., Wei, S.P., Horton, R., Shao, M.A., 2011. Effects of vegetation and slope aspect on water budget in the hill and gully region of the Loess Plateau of China. Catena 87, 90–100. Wu, Y.Q., Zheng, Q.H., Zhang, Y.G., Liu, B.Y., Cheng, H., Wang, Y.Z., 2008. Development of gullies and sediment production in the black soil region of northeastern China. Geomorphology 101, 683–691. Xiong, D.H., Long, Y., Yan, D.C., Lu, X.N., Ji, Z.H., Fang, H.D., 2009. Surface morphology of soil cracks in Yuanmou Dry-hot Valley Region, Southwest China. J. Mt. Sci. 6, 373–379. Xiong, D.H., Yan, D.C., Long, Y., Lu, X.N., Han, J.N., Han, X.Q., Shi, L.T., 2010. Simulation of morphological development of soil cracks in Yuanmou Dry-hot Valley Region, Southwest China. Chin. Geogr. Sci. 20 (2), 112–122. Xu, X.L., Ma, K.M., Fu, B.J., Song, C.J., Liu, W., 2008. Relationships between vegetation and soil and topography in a dry warm river valley, SW China. Catena 75, 138–145. Zhang, J.H., Tian, G., Li, Y., Lindstrom, M., 2003. Requirements for success of reforestation projects in a semiarid low-mountain region of the Jinsha River Basin, Southeastern China. Land Degrad. Dev. 13, 395–401.