Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China

Rangeland Ecology & Management xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Rangeland Ecology & Management journal homepage: http://www.elsevier.com/locate/rama

Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China☆ Liangxia Zhang a,b, Jiangwen Fan b,⁎, Decheng Zhou a, Haiyan Zhang b a Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters/Jiangsu Key Laboratory of Agricultural Meteorology, Nanjing University of Information Science and Technology, Nanjing 210044, China b Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China

a r t i c l e

i n f o

Article history: Received 25 November 2015 Received in revised form 13 March 2017 Accepted 4 May 2017 Available online xxxx Key Words: ecological restoration GLO-PEM grassland degradation forage supply livestock reduction

a b s t r a c t The Ecological Protection and Restoration Program (EPRP), initiated in 2005 in the Three-River Headwaters (TRH, the headwaters of the Yangtze, Yellow, and Lantsang rivers) region, is the largest project for nature reserve protection and reconstruction in China. This massive effort was expected to improve the trade-off between grassland productivity and grazing pressure in the region. However, the impacts of EPRP on forage supply and livestock carrying capacity remain poorly understood. Using the Global Production Efficiency Model and grazing pressure index, we investigated the influences of the EPRP by comparing the grassland yield and grazing pressure index before (1988−2004) and after (2005−2012) implementation of the program. Vegetation cover, represented by the annual maximum Normalized Difference Vegetation Index (NDVI), increased by 11.2% after implementation of the EPRP. The increase of NDVI, together with increasing temperature and precipitation, led to a 30.3% increase of the mean annual grassland yield in 2005−2012 relative to that in 1988−2004 (694 kg ha−1 vs. 533 kg ha−1 dry matter). We show that grazing pressure was largely alleviated by the EPRP due to increased grassland yield and decreased livestock number. This was indicated by a 36.1% decline of the grazing pressure index. The effects of the EPRP varied spatially. As examples, there were larger increases of grassland yield in the southeast of the region dominated by alpine meadow and greater reduction of grazing pressure in the central and eastern parts. Nevertheless, the ecological effectiveness of the EPRP may vary with the measures used and is indicated to be coupled with climate change. This calls for more detailed comparison and attribution analyses to predict the ongoing consequences of the EPRP in order to attain sustainable implementation of restoration practices in the TRH region. © 2017 The Society for Range Management. Published by Elsevier Inc. All rights reserved.

Introduction The Three-River Headwaters (TRH) region, located in the Qinghai– Tibet Plateau, is the headwaters of the Yangtze, Yellow, and Lantsang rivers of China. It provides 42.6 billion tons of water every year and is regarded as the water tower of China (Liu et al., 2008; Fan et al., 2010; Li et al., 2012). The ecological condition of the TRH region is critical to water conservation and ecological security in China and southeastern Asia (Myers et al., 2000; Li et al., 2012). Because of the high elevation and harsh environment, ecosystems in the TRH region are sensitive to climate change and anthropogenic ☆ This work is financially supported by the National Natural Science Foundation of China (41601196), National Science and Technology Support Project (2013BAC03B04), and Chinese Academy of Sciences Action Plan for the Development of Western China (KZCX2-XB3-08-01). ⁎ Correspondence: Jiangwen Fan, PhD, Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences 11A, Datun Road, Chaoyang District, Beijing, P. R. China 100101. Tel./fax: +86 10 64889807. E-mail address: [email protected] (J. Fan).

activities. Climate warming and extensive human activities (e.g., overgrazing, wood harvesting, gold mining) have caused a wide range of environmental problems in the region, such as loss of biodiversity, grassland degradation, and soil erosion (Feng et al., 2006; Liu et al., 2008; Li et al., 2013; Shang et al., 2013). Particularly, overgrazing is mostly believed to be the major cause of grassland degradation in the region (Li et al., 2013). Ecological protection and restoration of endangered ecosystems damaged by human activities have been widely applied globally (Wortley et al., 2013). They are today nowhere more evident in the world than in China (Li et al., 2012; Zhang et al., 2016) and the TRH region is a key example. An ecological project, the Ecological Protection and Restoration Program (EPRP), was implemented in the region in 2005. It implemented a series of practices including enclosure, livestock reduction, returning farmland or rangeland to grassland, severely degraded grassland restoration, grassland rodent control, cloud seeding, and hazard reduction management. Livestock reduction was begun in 2003, two yr earlier than implementation of the large-scale EPRP (Li et al., 2012). The EPRP is the largest project for nature reserve protection

http://dx.doi.org/10.1016/j.rama.2017.05.001 1550-7424/© 2017 The Society for Range Management. Published by Elsevier Inc. All rights reserved.

Please cite this article as: Zhang, L., et al., Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China, Rangeland Ecology & Management (2017), http://dx.doi.org/10.1016/j.rama.2017.05.001

2

L. Zhang et al. / Rangeland Ecology & Management xxx (2017) xxx–xxx

and reconstruction in China, and probably the world (Wang et al., 2010). About 7.5 billion Chinese yuan (≈ 1.1 billion $US) has been invested in this project (Qinghai People's Government, 2005). Therefore, the ecological effects of the project have gained considerable interest among scientists and land managers (Li et al., 2012; Huang et al., 2013; Zhang et al., 2014). Herbage supply (defined by grassland yield or production) and livestock carrying capacity are the two most important indicators used to monitor ecological condition for the TRH region (Fan et al., 2010; Yu et al., 2010; Zhang et al., 2014). This is particularly appropriate as N 70% of the population depends on livestock production for their livelihood in the region (Qinghai People's Government, 2005) and about 26 − 46% of the grassland has deteriorated due to overgrazing (Liu et al., 2008; Li et al., 2013). However, impacts of the EPRP on these two metrics in the TRH region are poorly understood. Relatively limited effort has focused on the effects of the EPRP on grassland cover or on ecosystem services such as water conservation, soil conservation, and biodiversity protection (Shao et al., 2013; Huang et al., 2014; Liu et al., 2014a). As an example of monitoring of change, Liu et al. (2014a) determined that the Normalized Difference Vegetation Index (NDVI) increased by 1.2% per decade in the TRH region after implementation of the EPRP. Study of the balance between livestock numbers and grassland production is urgently needed for the sustainable implementation of EPRP in the TRH region. In this study, we estimated grassland production by application of the Global Production Efficiency Model (GLO-PEM) and quantified livestock carrying capacity via the grazing pressure index over the TRH region from 1988 to 2012. We hypothesized that the EPRP would promote the balance between grassland yield and livestock grazing. Our specific objectives were to investigate 1) whether forage production was promoted and 2) if the grazing pressure was reduced by the EPRP. We did this by comparing grassland yield and grazing pressure before (1988−2004) and after (2005−2012) implementation of the EPRP.

(BARE-RSR, 2007). The region has 16 counties (Maduo, Maqin, Dari, Gande, Jiuzhi, Banma, Chenduo, Zaduo, Zhiduo, Qumalai, Nangqian, Yushu, Xinghai, Tongde, Zeku, and Henan) and one township (Tanggula). Grassland that covers 65.4% of the total area is the main ecosystem type in the region (Fan et al., 2010). The main grassland communities are alpine meadow and alpine steppe, with temperate steppe and alpine desert covering smaller areas (Chen et al., 2014). Related to the region’s altitudinal precipitation gradients, the grasslands show a transition of alpine meadow, alpine steppe, and alpine desert from the southeast to northwest (see Fig. 1). Alpine meadow occurs at 3 500−4 000 m and is dominated by Kobresia spp. Alpine steppe is distributed from 4 000−4 500 m and is dominated by Stipa purpurea, Carex moorcroftii, Festuca rubra, Festuca ovina, and Artemisia arenaria. Alpine desert, dominated by Thylacospermum caespitosum, Androsace tapete, Oxytropis sp., and Saussurea subulata, is mainly located at elevations above 4 500 m. At the end of 2012, livestock number of the TRH region was 16 × 106 sheep units (a sheep unit is equivalent to a 50-kg ewe and lamb, with a daily intake of 1.8 kg forage. Cattle and horses are assumed to be equivalent to four sheep units (Su et al., 2003)). Grasslands at higher altitudes are mainly used for summer grazing and those at lower altitudes for winter grazing (Fan et al., 2010). Winter-grazed pastures account for 45.5% of the grazed area in the region (ECAGRC, 1993). Winter pastures closer to the settlements are grazed for longer periods than summer pastures and consequently have more serious degradation (Zhang et al., 2006).

Materials and methods

where GY is grassland yield (kg DM ha−1 yr−1), ANPP is aboveground net primary production (kg DM ha−1 yr−1), NPP is net primary production (kg DM ha−1 yr−1), and R represents the ratio between measured ANPP and measured belowground net primary production (BNPP) for different grassland types. NPP was estimated by the GLO-PEM (Cao et al., 2004) at an 8-d interval and then aggregated to an annual level for the period 1988−2012. GLO-PEM is a light use efficiency (LUE) model widely used to simulate ecosystem production at both regional and global scales due to its

Study Area The TRH region is located in the center of the Qinghai–Tibet Plateau (31°39′−36°12′N, 89°45′−102°23′E). It has an area of 363 000 km 2 with altitude ranging from 2 800–6 564 m a.s.l. (Fig. 1). Its typical plateau continental climate has an annual mean temperature of − 5.6 to − 3.8°C and annual precipitation between 262.2 and 772.8 mm

Grassland Yield Calculation Grassland yield (GY) in terms of dry matter (DM) was represented by aboveground net primary production (ANPP) (Fan et al., 2010) and calculated as follows: GY ¼ ANPP ¼ NPP=ð1 þ RÞ

ð1Þ

Figure 1. Location of the Three-River Headwaters (TRH, the headwaters of the Yangtze, Yellow, and Lantsang rivers) region, with background indicating grassland types. The meteorological stations in and around the study area are also plotted.

Please cite this article as: Zhang, L., et al., Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China, Rangeland Ecology & Management (2017), http://dx.doi.org/10.1016/j.rama.2017.05.001

L. Zhang et al. / Rangeland Ecology & Management xxx (2017) xxx–xxx

theoretical soundness and practical convenience (Zhang et al., 2015). Inputs for the model include remote sensing (fPAR, fraction of absorbed photosynthetically active radiation), climate (i.e., temperature and precipitation), and soil data (Table 1). Data sources are given in the Data section. The key parameter (potential LUE, ɛmax) for the GLO-PEM was calculated by the quantum yield (Prince and Goward, 1995). Values for the other parameters (such as the minimum, maximum, and optimum air temperature for photosynthetic activity) were obtained from the previous modeling studies in the TRH region (Wang, 2007; Wang et al., 2009, 2016; Fan et al., 2010). Further details about GLO-PEM can be found in Supplementary data Text S1, available online at http://dx.doi.org/10.1016/j.rama.2017.05.001 (Wang et al., 2011; Zhang et al., 2015). The GLO-PEM modeled NPP (in terms of carbon) (kg C ha−1 yr−1) was converted into dry matter (kg DM ha−1 yr−1) by division with a coefficient of 0.45 (Haberl et al., 2001). R was calculated as a ratio of ANPP and BNPP, which were derived from measured aboveground and belowground biomass (AGB and BGB) for different grassland types, respectively. ANPP was represented by the dry measured AGB directly. BNPP was estimated from the dry measured BGB following the method proposed by Gill et al. (2002) as follows: BNPP ¼ BGB  Rlive  Turnover

ð2Þ

Turnover ¼ 0:0009  ANPP þ 0:25

ð3Þ

where BGB is measured dry BGB (kg DM ha−1 yr−1), Rlive represents the ratio of living BGB to total BGB, and Turnover is grassland root turnover rate. The value for Rlive was set to 0.8 based on in-situ measurements made by Zhou (2001) in the TRH region. For comparison, we also investigated the dynamics of grassland vegetation cover in the TRH region. Following previous conventions (Huang et al., 2014), the vegetation cover was represented by the annual maximum NDVI, which was generated by a Maximum Value Compositing (MVC) method (Holben, 1986) from AVHRR (1988−2000) and SPOTVGT (1998−2012) NDVI products in this study. Grazing Pressure Index Calculation The grazing pressure index (Ip), defined as the ratio of actual and theoretical carrying capacities (Fan et al., 2010), was calculated at the county level as: Ip ¼

Ca Ct

ð4Þ

where Ca and Ct are the actual and theoretical carrying capacities (sheep units), respectively. The livestock number is above the theoretical carrying capacity when Ip is larger than 1 (i.e., overload) and vice versa. The Ca for each county was calculated as follows: Ca ¼

C n  ð1 þ C h Þ  Gt 365

ð5Þ

where Cn is the livestock number (sheep units) at the end of the year. Ch is the livestock slaughter rate (%), which was set to 30% according to local statistical data. Gt is the grazing period (days [d]), which was set to 240 d for winter pasture and 125 d for summer pasture during 1988−1999. In order to reduce the grazing pressure on winter pasture

3

in 2000−2012, Gt for winter pasture was decreased to 210 days while that for summer pasture was increased to155 days (Fan et al., 2010). The Ct for each county was calculated as follows: Ct ¼

GY  U t  C o  H a  Ar S f  D f  Gt

ð6Þ

where Ut, Co, Ha, Sf, Df, and Ar represent the utilization rate of herbage (%), area available for pasture (%), edible herbage (%), daily intake (kg fresh forage per sheep unit), dry weight of fresh herbage (%), and grassland area (ha), respectively. The values of Ut, Co, Ha, Sf, and Df were set to 70%, 91.5%, 80%, 4 kg fresh forage, and 33%, respectively (Su et al., 2003). Data The fPAR data (8-d interval, 1-km spatial resolution) for the period 2000 − 2012 were obtained from Moderate Resolution Imaging Spectroradiometer (MODIS) products directly (MOD15A2, available from https://lpdaac.usgs.gov). The fPAR data for the period 1988 − 2004 were estimated based on NOAA’s Advanced Very High Resolution Radiometers (AVHRR) images via the canopy radiation transfer algorithm (Liu et al., 2007). The AVHRR images were received and processed by the National Satellite Meteorological Center, China Meteorological Administration, with corrections for geolocation, cloud screening, atmosphere, and sun-sensor viewing geometry (Wessels et al., 2004). In order to keep accordance with MODIS data, we calibrated the AVHRR fPAR on the basis of the relationship between MODIS fPAR and AVHRR fPAR in the overlapped period of 2000 − 2004 (Poulter et al., 2013). The NDVI data (1-km spatial resolution) for the periods 1988 − 2000 and 1998 − 2012 were collected from AVHRR (15-d composite, available from http://daac.gsfc.nasa.gov) and SPOTVGT (10-d composite, available from http://free.vgt.vito.be), respectively. The annual maximum NDVI values were produced using the Maximum Value Compositing (MVC) method (Holben, 1986). To keep accordance with SPOT-VGT data, we corrected the AVHRR NDVI based on the relationship between SPOT-VGT NDVI and AVHRR NDVI in the overlapped period of 1998−2000. Temperature, precipitation, relative humidity, and solar radiation data were obtained from China Meteorological Observation (http:// cdc.cma.gov.cn/) and Qinghai Meteorological Administration. About 160 meteorological stations located in and around the TRH region were used in the study (see Fig. 1). Climate data were interpolated into spatial data with a spatial resolution of 1 km by the widely used ANUSPLIN algorithm (Huang et al., 2013, 2014; Chen et al., 2014). Soil properties were obtained from China’s National Soil Inventory (http:// www.geodata.cn/). The grassland distribution map (winter and summer pastures) was acquired from the Editorial Committee of Atlas of Grassland Resources of China (ECAGRC, 1993, http://www.geodata.cn/). Livestock numbers for each county in the region were obtained from the statistical data provided by local governments. To validate the GLO-PEM model, field measurements of AGB and BGB along the actual aridity gradient in the TRH region were made in the 2004 summer season. Limited by budgets and extreme environments, we collected grassland biomass in 31 representative sites, covering the four main grassland types of alpine meadow, alpine steppe, alpine desert, and temperate steppe. Three replications of 1 m × 1 m

Table 1 Data inputs for Global Production Efficiency Model (GLO-PEM). Data fPAR Climate Grassland distribution Soil properties

Source NOAA/AVHRR (National Satellite Meteorological Center, China Meteorological Administration); MODIS FPAR product (MOD15A2, http://lpdaac.usgs.gov) China Meteorological Observation (http://cdc.cma.gov.cn/) and the Qinghai Meteorological Administration Editorial Committee of Atlas of Grassland Resources of China (http://www.geodata.cn/) The Second State Soil Survey of China (http://www.geodata.cn/)

Resolution

Time period

1 × 1 km

1988-2012

1 × 1 km 1:1 000 000 1:4 000 000

1988-2012 1980s 1980s

Please cite this article as: Zhang, L., et al., Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China, Rangeland Ecology & Management (2017), http://dx.doi.org/10.1016/j.rama.2017.05.001

4

L. Zhang et al. / Rangeland Ecology & Management xxx (2017) xxx–xxx

Results Validation of Grassland Yield The estimated grassland yields were validated by measured in-situ grassland dry matter yields made for this study. As shown in Figure 2, the GLO-PEM estimates agreed well with in-situ observations (n = 31, r = 0.73, P b 0.001), with the relative error (defined as the average ratio of the absolute difference between simulated and observed GY to the observed GY) of 26.7%. Changes of Grassland Yield

Figure 2. Comparison of simulated and observed grassland yield (GY) in the yr 2004.

sample quadrats for each site were arranged randomly in a 40 m × 50 m plot area. AGB was harvested to ground level and oven-dried for weighing. BGB was sampled by taking nine soil cores (3.1-cm inner diameter) at 0- to 30-cm-deep layers (depth intervals of 0−10, 10−20, and 20−30 cm) within each quadrat. Roots contained within the excavated soil were separated by washing with water through a 0.3-mm mesh sieve and dried in an oven at 80°C for 24 hours.

Both the annual mean temperature (r = 0.69, P b 0.01) and total precipitation (r = 0.46, P b 0.05) increased significantly for the TRH region from 1988 to 2012 (Fig. 3, a and b). Grassland vegetation cover, represented by annual maximum NDVI, also increased significantly over the whole study period (r = 0.81, P b 0.05) (Fig. 3, c). Consequently, the annual GY increased dramatically through time (Fig. 4, a) and the annual growth rate (represented by the slope of the linear trend) after implementation of the EPRP (i.e., 2005−2012) was clearly greater than that before implementation of the project (i.e., 1988 − 2004) (13.3 vs. 2.6 kg DM ha − 1 yr − 1). On average, mean annual GY in 2005 − 2012 was 30.3% higher than that in 1988−2004 (694 vs. 533 kg DM ha−1). GY varied greatly spatially in the TRH region, with the mean annual GY ranging from zero to 4471.1 kg DM ha−1 over the whole study period (Fig. 5). Larger increases were observed in southeastern parts of the region. Similar to the GY patterns, the GY increase (i.e., difference of

Figure 3. Interannual variations of (a) mean annual temperature, (b) total precipitation, and (c) annual maximum Normalized Difference Vegetation Index (NDVI) in the Three-River Headwaters (TRH) region during 1988 −2012. The horizontal dashed lines indicate the mean value averaged over the period before (1988 −2004) and after (2005− 2012) implementation of the Ecological Protection and Restoration Program (EPRP).

Figure 4. a, Interannual variations of grassland yield in the Three-River Headwaters (TRH) region during 1988 and 2012. b, Comparison of the grassland yield before and after the implementation of the Ecological Protection and Restoration Program over the TRH region. *Indicates the linear regression slope was significant at P = 0.05.

Please cite this article as: Zhang, L., et al., Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China, Rangeland Ecology & Management (2017), http://dx.doi.org/10.1016/j.rama.2017.05.001

L. Zhang et al. / Rangeland Ecology & Management xxx (2017) xxx–xxx

5

grassland types, temperate steppe had the greatest increase of 461.8 kg DM ha−1, but it occupies only a small part of the region (b 0.5%). Notably, GY decreased in a few areas, especially in southern parts of Tanggula, Zaduo, Nangqian, and Yushu counties. Changes of Grazing Pressure Since 1988, livestock numbers of the TRH have shown three typical developmental phases. Livestock number was about 20 × 10 6 sheep units in 1988 − 1996, dropped sharply to about 17 × 10 6 sheep units in 1997 − 2002, and then declined slowly to about 15 × 10 6 sheep units in 2003−2012 (Fig. 7, a). The average livestock number after implementation of livestock reduction (one major management practice of the EPRP in 2003 − 2012) was 21.3% lower than that in 1988 − 2002 (15.4 × 10 6 vs. 19.6 × 10 6 sheep units). The decrease of livestock number also had large spatial variations, with the greatest decrease in the central part of the TRH (Fig. 8, a). For example, livestock number for the Maduo, Chenduo, Yushu, and Nangqian counties decreased by N 30% after the project started. However, note that livestock numbers in Xinghai and Henan counties increased after implementation of the livestock reduction recommendations. The actual carrying capacity (Ca) decreased gradually from 1988 to 2012 while the theoretical carrying capacity (Ct) increased slowly in the same period (see Fig. 7, b). Specifically, the Ca and Ct in 2003 − 2012 (after implementation of livestock reduction) were respectively 21.3% lower (20 × 10 6 vs. 25.4 × 10 6 sheep units) and 22.9% higher (12.9 × 10 6 vs. 11.3 × 10 6 sheep units) than in 1988−2002. As a result, the grazing pressure index (Ip) declined substantially over the 1988−2012 period (Fig. 7, c). On average, Ip before livestock reduction was 2.29 (i.e., the grassland was overloaded by 1.29 times) and then it dropped to 1.46 (by 36.1%) after implementation of livestock reduction. Further, in general Ip was higher for winter pasture than summer pasture, yet the difference between them decreased gradually during the study period (from 1.93 in 1988 − 2002 to 0.87 in 2003−2012) (Fig. 7, d). Moreover, county changes of Ip varied greatly (Fig. 8, b). Larger decreases were observed in the central and eastern parts of the TRH region. For example, Ip for Maduo, Chenduo, Zaduo, Gande, and Jiuzhi counties were reduced by N 50%. In contrast, Ip decreased slightly in Zaduo county (b 10%). Discussion Increased Grassland Yield in the TRH Region

Figure 5. Spatial pattern of the annual grassland yield (a) before and (b) after implementation of the Ecological Protection and Restoration Program and (c) their difference in the Three-River Headwaters region.

mean annual GY between 2005 and 2012 and 1988 and 2004) was clearly larger in the southeast than northwest parts of the region, with overall difference N 225 kg DM ha−1. More specifically, the mean annual GYs in 2005 − 2012 were 54.0%, 52.9%, 50.8%, and 40.4% higher than those in 1988 − 2004 for Qumalai, Xinghai, Tongde, and Zeku counties, respectively (Fig. 6). In contrast, GY increased slightly in northwestern areas (e.g., Yushu, Nangqian, and Zaduo counties), with increases b 75 kg DM ha − 1. The large spatial variations of GY changes can also be viewed in terms of grassland types (see Fig. 4, b). For example, comparing the two most widely distributed grassland types, alpine meadow (184 kg DM ha−1) had a much larger increase than alpine steppe (61 kg DM ha−1). Of the

Our results indicate that grassland yield increased significantly after implementation of the EPRP in the TRH region, agreeing well with previous in-situ observations (Qinghai Grassland Station, 2013; Wu et al., 2013; Xiong et al., 2014) and the findings in the other regions such as the Inner Mongolia and Loess Plateaus of China (Mu et al., 2013; Zhou et al., 2014; Hao et al., 2014). The increase of grassland yield can at least be partly attributed to the implementation of the EPRP. On one hand, the grassland protection practices (e.g., enclosure), can improve vegetation cover directly (see Fig. 3, c) and thus forage production. This can be verified by the overall significant and positive relationship between vegetation cover and grassland yield across the TRH region (Fig. 9). On the other hand, cloud seeding (with a 38.9 billion m 3 increase from 2006 to 2011), a major practice implemented by the EPRP, could have increased annual precipitation in the region and thus grassland yield (QAWO (Qinghai Artificial Weather Office)., 2011; Xiao et al., 2013; Liu et al., 2014a, 2014b). Climate change, coupled with the EPRP, could also have substantially influenced grassland productivity. For example, the annual mean temperature increased considerably through the period of implementation of EPRP in the TRH region (see Fig. 3, a). This resulted in an advance of the vegetation growing season on the Qinghai-Tibet Plateau (Zhang et al., 2013) and may have been a factor increasing grassland yield. By

Please cite this article as: Zhang, L., et al., Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China, Rangeland Ecology & Management (2017), http://dx.doi.org/10.1016/j.rama.2017.05.001

6

L. Zhang et al. / Rangeland Ecology & Management xxx (2017) xxx–xxx

Figure 6. Grassland yield and annual growth rate before and after implementation of the Ecological Protection and Restoration Program for each county in the Three-River Headwaters region.

comparing the results from a process-based ecosystem model (climatic effect) and a LUE model (climatic and anthropogenic effects), Chen et al. (2014) concluded that the contribution of climate change to the rise of grassland yield is similar to that of ecological projects over the QinghaiTibet Plateau (53.6% vs. 46.4%). Apparently, large uncertainties existed in their findings since the two models used are different in model structure and parameter settings. Further studies are needed to partition the

effects of climate change and EPRP on the increased grassland yield over the TRH region. Further, we found that changes of grassland yield varied greatly spatially with relatively larger increases in southeastern parts of the TRH region dominated by alpine meadow (see Figs. 5 − 6). This might be mainly due to the higher annual precipitation in the southeastern relative to northwestern areas. For instance, Xiong et al. (2014) indicated

Figure 7. Changes of (a) livestock numbers, (b) actual and theoretical carrying capacities, (c) grazing pressure indices, and (d) grazing pressure indices for winter and summer pastures from 1988 to 2012 in the Three-River Headwaters region.

Please cite this article as: Zhang, L., et al., Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China, Rangeland Ecology & Management (2017), http://dx.doi.org/10.1016/j.rama.2017.05.001

L. Zhang et al. / Rangeland Ecology & Management xxx (2017) xxx–xxx

7

Figure 8. (a) Livestock number and decrease rate and (b) the grazing pressure index and decrease rate before and after the implementation of Ecological Protection and Restoration Program (EPRP) for each county in the Three-River Headwaters region. The white area in the map indicates the lack of data for livestock number of the Tanggula township.

that background precipitation was the major driver for the biomass enhancement caused by the ecological projects in the Qinghai-Tibet Plateau. Notably, our results indicate that vegetation production decreased in a few areas, particularly in some southern counties. This can mainly be attributed to the difficulty of implementing the EPRP in very high altitude and harsh climatic conditions (Yao et al., 2000). Decreased Grazing Pressure in the TRH Region Our results demonstrate that grazing pressure has been greatly reduced by the EPRP in the study area, as indicated by a 36.1% decline of grazing pressure index (i.e., Ip) after implementation of the project. This can be explained by increase of grassland yield (improved theoretical carrying capacity) and decrease of livestock number (reduced actual carrying capacity). Therefore, larger decreases of grazing pressure were observed in the central and eastern parts of the TRH region (see Fig. 8, b), which had greater increases of grassland yield (see Fig. 6) and decreases of livestock number (see Fig. 8, a). Note that factors besides EPRP also contribute to the livestock reduction in the TRH region.

For example, natural disasters usually lead to a sharp decrease of livestock number and the decrease could be amplified by the Rangeland Household Contract Policy (RHCP, implemented in the TRH region since 1984) that constrained the livestock mobility (Gongbuzeren et al., 2015). This might mainly explain the drop of livestock number in 1996−1997 when a serious snow disaster happened (Chen, 1996). The decrease of grazing pressure was critical for grassland recovery in the TRH region (Qiao and Wang, 2010; Li et al., 2013; Chen et al., 2014). However, livestock numbers are still very high even though they have been reduced by the ERPP (see Fig. 7, a). For example, the grazing pressure index (Ip) was estimated to be 1.32 by the end of our study period, indicating that the grassland is still overstocked given that livestock are mainly pasture-fed. As expected, grazing pressure was found to be much higher for winter than for summer pastures. This can be attributed to the fact that local herdsmen are more likely to graze winter pastures due to the proximity of these to their settlements and watering facilities. The longer grazing period, together with higher grazing intensity, usually leads to more serious degradation of winter pastures than that for summer pastures

Please cite this article as: Zhang, L., et al., Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China, Rangeland Ecology & Management (2017), http://dx.doi.org/10.1016/j.rama.2017.05.001

8

L. Zhang et al. / Rangeland Ecology & Management xxx (2017) xxx–xxx

Figure 9. Spatial distribution of the correlation coefficient (r) between NDVI and grassland yield between 1988 and 2012 in the Three-River Headwaters region.

(Zhang et al., 2006). However, we show that the grazing pressure for winter pastures decreased gradually (see Fig. 7, d) after implementation of the EPRP due to the reduction of both livestock number and grazing period. The increases of grassland cover and production and the decrease of grazing pressure are beneficial for soil conservation and carbon mitigation potential (Conant et al., 2001; Dean et al., 2015) and therefore help mitigate climate change (Ogle et al., 2004; IPCC, 2006). For example, Shao et al. (2011) showed that increased vegetation cover reduced soil erosion rates significantly for alpine meadow in the study region. Huang et al. (2014) estimated that soil and BGB and AGB of degraded grasslands in the TRH region can sequester 515.1, 44.6, and 21.7 Mt C, respectively, provided that they have been restored to a healthy status. Implications This study shows that the EPRP improved the balance between grassland productivity and grazing pressure, endorsing the importance of the EPRP in improving the ecological conditions of the TRH region. The effectiveness of the project varied greatly across space and with grassland type, as indicated by the greater enhancement of grassland yield in southeastern parts and the larger reductions of grazing pressure in central and eastern parts of the TRH region. This suggests that ecorestoration practices should be preferentially implemented in areas of the region with relatively warmer and wetter climates as these areas will have greater increases of grassland productivity (Xiong et al., 2014). However, viewed from their ecosystem vulnerability, drier regions also need to be preferentially covered by implementation of ecorestoration practices (Wortley et al., 2013; Liu et al., 2016). Further, we emphasize the importance of further reduction of livestock number in the TRH region as the grasslands are still overstocked. We emphasize that the effects of the EPRP may differ substantially according to the specific management measures applied (Hu et al., 2016). More detailed comparisons and attribution analysis are needed for formulating effective eco-restoration plans and strategies in the TRH region. Acknowledgments We thank Dr. Quanqin Shao, Dr. Junbang Wang, Dr. Zhuoqi Chen, Dr. Wei Cao, Dr. Lin Huang, Dr. Longfei Bing, Dr. Jun Zhai, and Fengpei Tang for their

valuable help in the field survey and data analysis. We thank the local governments and Environmental Protection Agency of Qinghai province for providing livestock number data for each county. We thank two anonymous reviewers for their valuable comments to improve this manuscript. We also thank Dr. Warwick Harris (Emeritus Research Associate, Landcare Research, New Zealand) for his valuable comments and for improving the text. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.rama.2017.05.001. References BARE-RSR, 2007. A comprehensive background assessment report on ecosystem in river source region. IGSNRR, Beijing, China [in Chinese]. Cao, M., Prince, S.D., Small, J., Goetz, S.J., 2004. Remotely sensed interannual variations and trends in terrestrial net primary productivity 1981−2000. Ecosystems 7, 233–242. Chen, B., Zhang, X., Tao, J., Wu, J., Wang, J., Shi, P., Zhang, Y., Yu, C., 2014. The impact of climate change and anthropogenic activities on alpine grassland over the Qinghai-Tibet Plateau. Agricultural and Forest Meteorology 189, 11–18. Chen, Q.G., 1996. The snowdisaster and its defence strategy in Yushu Tibetan prefecture of Qinghai province. Pratacultural Science 13, 60–63 [in Chinese with English abstract]. Conant, R.T., Paustian, K., Elliott, E.T., 2001. Grassland management and conversion into grassland: effects on soil carbon. Ecological Applications 11, 343–355. Dean, C., Kirkpatrick, J.B., Harper, R.J., Eldridge, D.J., 2015. Optimising carbon sequestration in arid and semiarid rangelands. Ecological Engineering 74, 148–163. ECAGRC (Editorial Committee of Atlas of Grassland Resources of China), 1993. Atlas of grassland resources of China (1:1M). China Atlas, Beijing, China [in Chinese]. Fan, J.W., Shao, Q.Q., Liu, J.Y., Wang, J.B., Harris, W., Chen, Z.Q., Zhong, H.P., Xu, X.L., Liu, R.G., 2010. Assessment of effects of climate change and grazing activity on grassland yield in the Three Rivers Headwaters Region of Qinghai-Tibet Plateau, China. Environmental Monitoring and Assessment 170, 571–584. Feng, J., Wang, T., Xie, C., 2006. Eco-Environmental Degradation in the Source Region of the Yellow River, Northeast Qinghai-Xizang Plateau. Environmental Monitoring and Assessment 122, 125–143. Gill, R.A., Kelly, R.H., Parton, W.J., Day, K.A., Jackson, R.B., Morgan, J.A., Scurlock, J.M.O., Tieszen, L.L., Castle, J.V., Ojima, D.S., 2002. Using simple environmental variables to estimate below-ground productivity in grasslands. Global Ecology and Biogeography 11, 79–86. Gongbuzeren, Li, Y., Li, W., 2015. China’s rangeland management policy debates: what have we learned? Rangeland Ecology & Management 68, 305–314. Hao, L., Sun, G., Liu, Y., Gao, Z., He, J., Shi, T., Wu, B., 2014. Effects of precipitation on grassland ecosystem restoration under grazing exclusion in Inner Mongolia, China. Landscape Ecology 29, 1657–1673.

Please cite this article as: Zhang, L., et al., Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China, Rangeland Ecology & Management (2017), http://dx.doi.org/10.1016/j.rama.2017.05.001

L. Zhang et al. / Rangeland Ecology & Management xxx (2017) xxx–xxx Haberl, H., Erb, K.H., Krausmann, F., Loibl, W., Schulz, N., Weisz, H., 2001. Changes in ecosystem processes induced by land use: human appropriation of aboveground NPP and its influence on standing crop in Austria. Global Biogeochemical Cycles 15, 929–942. Holben, B.N., 1986. Characteristics of maximum-value composite images from temporal AVHRR data. International Journal of Remote Sensing 7, 1417–1434. Hu, Z., Li, S., Guo, Q., Niu, S., He, N., Li, L., Yu, G., 2016. A synthesis of the effect of grazing exclusion on carbon dynamics in grasslands in China. Global Change Biology 22 (4), 1385–1393. Huang, L., Xiao, T., Zhao, Z.P., Sun, C., Liu, J.Y., Shao, Q.Q., Fan, J.W., Wang, J.B., 2013. Effects of grassland restoration programs on ecosystems in arid and semiarid China. Journal of Environmental Management 117, 268–275. Huang, L., Xu, X.L., Shao, Q.Q., Liu, J.Y., 2014. Improving carbon mitigation potential through grassland ecosystem restoration under climatic change in northeastern Tibetan Plateau. Advances in Meteorology 6, 154. IPCC, 2006. 2006 IPCC guidelines for national greenhouse gas inventories Volume 4, Agriculture, Forestry and Other Land Uses (AFOLLI). IPCC/IGES, Hayama, Japan. Li, X.L., Gao, J., Brierley, G., Qiao, Y.M., Zhang, J., Yang, and Y. W., 2013. Rangeland degradation on the Qinghai-Tibet plateau: implications for rehabilitation. Land Degradation & Development 24, 72–80. Li, X.L., Brierley, G., Shi, D., Xie, Y., Sun, H., 2012. Ecological protection and restoration in Sanjiangyuan National Nature Reserve, Qinghai Province, China. Perspectives on Environmental Management and Technology in Asian River Basins. Springer, Berlin, Germany, pp. 93–120. Liu, D., Cao, C., Dubovyk, O., Tian, R., Chen, W., Zhuang, Q., Zhao, Y., Menz, G., 2016. Using fuzzy analytic hierarchy process for spatio-temporal analysis of eco-environmental vulnerability change during 1990–2010 in Sanjiangyuan region, China. Ecological Indicators 73, 612–625. Liu, J.Y., Xu, X.L., Shao, Q.Q., 2008. Grassland degradation in the “Three-River Headwaters” region, Qinghai Province. Journal of Geographical Sciences 18, 259–273. Liu, R., Chen, J.M., Liu, J., Deng, F., Sun, R., 2007. Application of a new leaf area index algorithm to China's landmass using MODIS data for carbon cycle research. Journal of Environmental Management 85, 649–658. Liu, X., Zhang, J., Zhu, X., Pan, Y., Liu, Y., Zhang, D., Lin, Z., 2014a. Spatiotemporal changes in vegetation coverage and its driving factors in the Three-River Headwaters Region during 2000–2011. Journal of Geographical Sciences 24, 288–302. Liu, X., Zhu, X., Zhu, W., Pan, Y., Zhang, C., Zhang, D., 2014b. Changes in spring phenology in the Three-Rivers Headwater Region from 1999 to 2013. Remote Sensing 6, 9130–9144. Mu, S., Zhou, S., Chen, Y., Li, J., Ju, W., Odeh, I., 2013. Assessing the impact of restorationinduced land conversion and management alternatives on net primary productivity in Inner Mongolian grassland, China. Global and Planetary Change 108, 29–41. Myers, N., Mittermeier, R.A., Mittermeier, C.G., Fonseca, G.A.B.D., Kent, J., 2000. Biodiversity hotspots for conservation priorities. Nature 403, 853–858. Ogle, S.M., Conant, R.T., Paustian, K., 2004. Deriving grassland management factors for a carbon accounting method developed by the intergovernmental panel on climate change. Environmental Management 33, 474–484. Prince, S.D., Goward, S.N., 1995. Global primary production: a remote sensing approach. Journal of Biogeography 22, 815–835. Poulter, B., Pederson, N., Liu, H., Zhu, Z., D’Arrigo, R., Ciais, P., Davi, N., Frank, D., Leland, C., Myneni, R., 2013. Recent trends in Inner Asian forest dynamics to temperature and precipitation indicate high sensitivity to climate change. Agricultural and Forest Meteorology 178, 31–45. QAWO (Qinghai Artificial Weather Office)., 2011. The statistical analysis report of artificial precipitation effects in Three-River Headwaters region and summer-autumn season in Qinghai province. QAWO, Xining, China [in Chinese]. Qiao, Y., Wang, Q., 2010. Driving forces of alpine meadow degradation in the Yellow River headwaters. In: Brierley, G.J., Li, X., Chen, G. (Eds.), Landscape and environmental science and management in the Sanjiangyuan Region. Qinghai People Press, Xining, China, pp. 149–165. Qinghai Grassland Station, 2013. Grassland monitoring and analysis report in 2002 in the ecological monitoring program of general planning for Ecological Conservation and Construction in Sanjiangyuan Nature Reserve. Qinghai Grassland Station, Xining, China [in Chinese]. Qinghai People's Government, 2005. General Planning for Ecological Conservation and Construction in Sanjiangyuan Nature Reserve. QPG, Xining, China [in Chinese].

9

Shang, Z., Feng, Q., Wu, G., Ren, G., Long, R., 2013. Grasslandification has significant impacts on soil carbon, nitrogen and phosphorus of alpine wetlands on the Tibetan Plateau. Ecological Engineering 58, 170–179. Shao, Q.Q., Liu, J.Y., Huang, L., Fan, J.W., Xu, X.L., Wang, J.B., 2013. Integrated assessment on the effectiveness of ecological conservation in Sanjiangyuan National Nature Reserve. Geographical Research 32, 1645–1656 [in Chinese with English abstract]. Shao, Q.Q., Xiao, T., Liu, J.Y., Qi, Y.Q., 2011. Soil erosion rates and characteristics of typical alpine meadow using 137Cs technique in Qinghai-Tibet Plateau. Chinese Science Bulletin 56, 1708–1713. Su, D.X., Meng, Y.D., Wu, B.G., 2003. Agricultural industry standards of P. R. China—calculation of proper carrying capacity of rangelands (NY/T 635–2002). Standards Press of China, Beijing, China, pp. 2–6 [in Chinese]. Wang, J.B., 2007. Modeling carbon fluxes of terrestrial ecosystem on regional scale through coupling a remote sensing model with an ecosystem process model. Postdoctoral work report, Institute of Geographical Sciences and Nature Resources Research. Chinese Academy of Sciences, Beijing, China [in Chinese]. Wang, J.B., Liu, J.Y., Shao, Q.Q., Liu, R.G., Fan, J.W., Chen, Z.Q., 2009. Spatial-temporal patterns of net primary productivity for 1988-2004 based on GLOPEM-CEVSA model in the “Three-River Headwaters” Region of Qinghai province, China. Chinese Journal of Plant Ecology 33, 254–269 [in Chinese with English abstract]. Wang, J., Liu, J., Cao, M., Liu, Y., Yu, G., Li, G., Qi, S., Li, K., 2011. Modelling carbon fluxes of different forests by coupling a remote-sensing model with an ecosystem process model. International Journal of Remote Sensing 32, 6539–6567. Wang, Z., Song, K., Hu, L., 2010. China’s Largest Scale Ecological Migration in the ThreeRiver Headwater Region. Ambio 39, 443–446. Wang, X., Cheng, G., Li, X., Lu, L., Ma, M., Su, P., Zhu, G., Tan, J., 2016. A comparison of two photosynthesis parameterization schemes for an alpine meadow site on the QinghaiTibetan Plateau. Theoretical and Applied Climatology 126, 751–764. Wessels, K., Prince, S., Frost, P., Van Zyl, D., 2004. Assessing the effects of human-induced land degradation in the former homelands of northern South Africa with a 1 km AVHRR NDVI time-series. Remote Sensing of Environment 91, 47–67. Wortley, L., Hero, J.M., Howes, M., 2013. Evaluating ecological restoration success: a review of the literature. Restoration Ecology 21, 537–543. Wu, J., Zhang, X., Shen, Z., Shi, P., Xu, X., Li, X., 2013. Grazing-exclusion effects on aboveground biomass and water-use efficiency of alpine grasslands on the northern Tibetan Plateau. Rangeland Ecology & Management 66, 454–461. Xiao, J., Sun, G., Chen, J., Chen, H., Chen, S., Dong, G., Gao, S., Guo, H., Guo, J., Han, S., 2013. Carbon fluxes, evapotranspiration, and water use efficiency of terrestrial ecosystems in China. Agricultural and Forest Meteorology 182, 76–90. Xiong, D., Shi, P., Sun, Y., Wu, J., Zhang, X., 2014. Effects of grazing exclusion on plant productivity and soil carbon, nitrogen storage in alpine meadows in northern Tibet, China. Chinese Geographical Science 24, 488–498. Yao, T., Liu, X., Wang, N., Shi, Y., 2000. Amplitude of climatic changes in Qinghai-Tibetan Plateau. Chinese Science Bulletin 45, 1236–1243. Yu, L., Zhou, L., Liu, W., Zhou, H.K., 2010. Using remote sensing and GIS technologies to estimate grass yield and livestock carrying capacity of alpine grasslands in Golog Prefecture, China. Pedosphere 20, 342–351. Zhang, G., Zhang, Y., Dong, J., Xiao, X., 2013. Green-up dates in the Tibetan Plateau have continuously advanced from 1982 to 2011. Proceedings of the National Academy of Sciences 110, 4309–4314. Zhang, J., Zhang, L., Liu, W., Qi, Y., Xiao, W., 2014. Livestock-carrying capacity and overgrazing status of alpine grassland in the Three-River Headwaters region, China. Journal of Geographical Sciences 24, 303–312. Zhang, L.X., Zhou, D.C., Fan, J.W., Hu, Z.M., 2015. Comparison of four light use efficiency models for estimating terrestrial gross primary production. Ecological Modelling 300, 30–39. Zhang, Y.L., Liu, L.S., Bai, W.Q., Shen, Z.X., Yan, J.Z., Ding, M.J., Li, S.C., Zheng, D., 2006. Grassland degradation in the source region of the Yellow River. Acta Geographica Sinica 61, 3–14 [in Chinese with English abstract]. Zhang, L.X., Cao, W., Fan, J.W., 2016. Soil organic carbon dynamics in Xilingol grassland of northern China induced by the Beijing-Tianjin Sand Source Control Program. Front Earth Science http://dx.doi.org/10.1007/s11707-016-0589-9. Zhou, D., Zhao, S., Liu, S., Zhang, L., 2014. Modeling the effects of the Sloping Land Conversion Program on terrestrial ecosystem carbon dynamics in the Loess Plateau: a case study with Ansai County, Shaanxi province, China. Ecological Modelling 288, 47–54. Zhou, X.M., 2001. Chinese Kobresia meadow. Science, Beijing, China, p. 158 [in Chinese].

Please cite this article as: Zhang, L., et al., Ecological Protection and Restoration Program Reduced Grazing Pressure in the Three-River Headwaters Region, China, Rangeland Ecology & Management (2017), http://dx.doi.org/10.1016/j.rama.2017.05.001