Effects of altered precipitation regimes on plant productivity in the arid region of northern China
(Electronic Supplementary Material)
Hao-jie Xua,b, Xin-ping Wanga
a Shapotou Desert Research and Experiment Station, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, Lanzhou 730000, China
b University of Chinese Academy of Sciences, Beijing 100049, China
Corresponding author
Prof. Xin-ping Wang
Tel: +86-931-4967183
Fax: +86-931-8273894
E-mail:
This supplementary material introduces the datasets and software packages in this study. In addition, an in-depth overview of the methodology is provided (Fig.S1). This material comprises three parts: Data List, Tool List, and Methodology.
1. Data List
1) Remote Sensing Data
l The 16-day composition MODIS NDVI product (MOD13A2) with a spatial resolution of 1 km from 2000 to 2013. The MODIS NDVI dataset is corrected for calibration, view geometry and atmospheric correction (Fensholt et al., 2009).
Available Online: https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mod13a2
l Monthly gridded Tropical Rainfall Measurement Mission (TRMM) product (TRMM 3B43) at 0.25° spatial resolution from 2000 to 2013. TRMM covers the precipitation radar, and the main scientific goals for TRMM are to determine the distribution and variability of precipitation on a monthly average in tropical and sub-tropical regions (Dr. Simpson et al., 1996).
Available Online: http://trmm.gsfc.nasa.gov/
l Terra/MODIS Annual Net Primary Production product (MOD17A3) with a spatial resolution of 1 km from 2000 to 2010. This product produces gross primary productivity of vegetation every day and sum to net primary productivity at the end of the year. A more detailed overview of the MODIS NPP algorithm is introduced by Zhao et al. (2005).
Available Online: https://lpdaac.usgs.gov/dataset_discovery/modis/modis_products_table/mod17a3
2) Meteorological Data
l Monthly meteorological data including temperature, precipitation and solar radiation from 72 permanent meteorological stations from 2000 to 2013 (Table S1). These data are downloaded from the China Meteorological Data Sharing Service System.
Available Online: http://cdc.nmic.cn/dataSetLogger.do?changeFlag=dataLogger
3) Aboveground Biomass Data
l The aboveground biomass data, including alive and standing dead biomass produced in the current year, are collected from 60 grassland sites and 25 shrubland sites in late July or August from 2001 to 2005 (Fig. 1 and Fig 3 in the manuscript). At each grassland site (10×10 m), the aboveground biomass in five quadrates (1×1 m) is measured. All biomass samples are oven-dried at 65 °C for 48 h to constant mass and are weighed to the nearest 0.1 g. At each shrubland sites (10×10 m), shrub plants are categorized into three groups in the light of their sizes (large, medium and small), and an exemplary plant in each group is selected with its current-year leaves and twigs being harvested, oven-dried and weighed. The total weights of the three shrub groups are calculated by multiplying the weight of the representative plants by their quantities in each group. To validate the results of simulated grassland and shrubland NPP, we assume that aboveground biomass represent aboveground NPP and obtain an estimated ratio of aboveground to belowground NPP (Ma et al., 2008). The estimated ratio of aboveground to belowground NPP for grasslands and shrublands are 0.385 and 0.635, respectively.
4) Ancillary data
l 1:250,000 scale Administrative District Map of China.
Available Online:
http://emuch.net/bbs/attachment.php?tid=5095997&aid=12003
l 1: 1:1,000,000 scale Chinese Vegetation Map of 2010.
Available Online: http://data.ess.tsinghua.edu.cn
l 1: 1:1,000,000 scale Chinese Soil Map of 2004.
Available Online: http://westdc.westgis.ac.cn/data/611f7d50-b419-4d14-b4dd-4a944b141175
l Digital Elevation Model (DEM) of China with a spatial resolution of 1 km.
Available Online: https://lta.cr.usgs.gov/GMTED2010
2. Tool List
l MODIS Reprojection Tool (MRT)
It enables users to read data files in HDF-EOS format as input to processing, perform geographic transformation to a different coordinate system/cartographic projection, and write the output to file formats other than HDF-EOS.
Available Online: https://lpdaac.usgs.gov/tools/modis_reprojection_tool
l ANUSPLIN package
It provides a facility for transparent analysis and interpolation of multi-variate data using thin plate smoothing splines, through comprehensive statistical analyses, data diagnostics and spatially distributed standard errors.
Available Online: http://fennerschool.anu.edu.au/research/products/anusplin-vrsn-44
l TIMESAT package
It is originally intended for handling noisy time-series of AVHRR NDVI data and to extract seasonality information from the data. The program now has the capability to handle different types of remotely sensed time-series, e.g. data from Terra/MODIS at different time resolutions.
Available Online: http://web.nateko.lu.se/timesat/timesat.asp
3. Methodology
l We use a flow chart to provide a detailed overview of the methodology in this study (Fig.S1). The computation formulas are provided in Method section.
² CASA Model
Based on MODIS NDVI data, temperature, precipitation and solar radiation data, as well as vegetation type and soil texture information (Fig.S1), the CASA (Carnegie–Ames–Stanford Approach) model was developed to simulate monthly NPP (Potter et al., 1999). The annual NPP dataset was obtained by calculating accumulated monthly NPP in a year (Zhao et al., 2005).
- In the CASA model, NPP is determined with absorbed photosynthetically active radiation (APAR), multiplied by light utilization efficiency (ε). It is noteworthy that the effect of maintenance respiration (MR) is estimated by time-varying stress scalar terms for temperature and moisture behind the CASA algorithm logic (Potter et al., 1999).
1)
Where PAR is the total incident photosynthetically active radiation (MJ/m2) and accounts for 50% of total solar radiation. fAPAR is the fraction of PAR absorbed by photosynthetic tissues in a canopy. ε* is the maximum light utilization efficiency set as different constant values for different land cover types. In this study, ε* of forests, shrublands, grasslands and croplands was set to 1.004, 0.768, 0.608 and 0.604 g C/MJ, respectively (Zhu et al., 2006). Tε and Wε are time-varying stress scalar terms for temperature and soil water balance.
- fAPAR is estimated from a linear function of NDVI following Myneni and Williams (1994).
2)
Where NDVImin and NDVImax correspond to the 5th and 95th percentile of NDVI in each land cover type. fAPARmin and fAPARmax are set to 0.001 and 0.95.
- Tε is calculated concerning derivation of the optimal temperature (Topt) for plant production.
3)
Where Topt is the monthly mean temperature (T) in the month of maximum NDVI.
- Wε is the monthly water deficit, which is simulated by a comparison of monthly actual evapotranspiration (ET) to monthly potential evapotranspiration (PET) from the method of Rahimi et al. (2015) and Yu et al. (2011).
4)
5)
6)
7)
8)
9)
10)
Where P is the monthly precipitation (mm). Rn is the net solar radiation (MJ/m2). I is the total heat index in a year. Recent studies suggested that the used method for estimating potential evapotranspiration in this study was applicable and satisfactory (Valipour, 2015a, 2015b)
² NPP trend analysis
Based on the NPP simulation, further investigation was conducted to detect long-term annual NPP changing trend for each pixel (Chen et al., 2014).
11)
Where n is the sequential year. NPPi is the annual NPP in the year i. A positive or negative slope value suggests a linear increasing or decreasing trend in NPP within the time.
² NPP percentage change
The total percentage change of annual NPP (TPC) was measured as the ratio of slope to the initial values (Ma and Frank, 2006).
12)
² GSP and SDP calculation
GSP was defined as accumulated precipitation for the growing season (April to August). SDP was quantified by calculating the coefficient of variance (CV) for monthly precipitation from April to August (Guo et al., 2012).
13)
14)
Where Pi is the accumulated precipitation of month i. Pmean is the mean precipitation of the 5 months.
² The effects of GSP and SDP on NPP variability
To establish the relationships between NPP and GSP, and between NPP and SDP, the partial correlation analysis was used to reflect whether SDP affects NPP variability independent of GSP (Mao et al., 2014).
15)
Where rxy is the Pearson correlation coefficient. xi and yi represent precipitation variables (GSP or SDP) and annual NPP in the year i. xmean and ymean are the multi-year mean values for x and y.
16)
Where r*xy.z is the partial correlation coefficient, showing the relationship between variable x and variable y after excluding the effect of variable z. y is NPP. x and z are two different precipitation variables (GSP or SDP).
² The relative contribution of GSP and SDP to NPP variability
The stepwise regression model was adopted to quantify the relative contribution of GSP and SDP to the temporal variability in NPP (Graham, 2003).
Table S1 The geographical location of 72 meteorological stations.
Station ID / Name / Longitude / Altitude / Elevation52895 / Jingyuan / 104.68 / 36.57 / 1398.2
51931 / Yutian / 81.65 / 36.85 / 1422.0
51839 / Minfeng / 82.72 / 37.07 / 1409.3
51828 / Hotan / 79.93 / 37.13 / 1374.6
52797 / Jingtai / 104.05 / 37.18 / 1630.5
53705 / Zhongning / 105.67 / 37.48 / 1183.3
51818 / Pishan / 78.28 / 37.62 / 1375.4
52679 / Wuwei / 102.67 / 37.92 / 1530.9
51855 / Qiemo / 85.55 / 38.15 / 1247.5
52674 / Yongchang / 101.97 / 38.23 / 1976.1
51811 / Shache / 77.27 / 38.43 / 1231.2
53614 / Yinchuan / 106.22 / 38.48 / 1111.4
52681 / Minqin / 103.08 / 38.63 / 1367.0
52661 / Shandan / 101.08 / 38.80 / 1764.6
53615 / Taole / 106.70 / 38.80 / 1101.6
53602 / Alxa Left Banner / 105.67 / 38.83 / 1561.4
52652 / Zhangye / 100.43 / 38.93 / 1482.7
51777 / Ruoqiang / 88.17 / 39.03 / 888.3
52576 / Alxa Right Banner / 101.68 / 39.22 / 1510.1
53519 / Huinong / 106.77 / 39.22 / 1091.0
52546 / Gaotai / 99.83 / 39.37 / 1332.2
51705 / Wuqia / 75.25 / 39.72 / 2175.7
52533 / Jiuquan / 98.48 / 39.77 / 1477.2
53502 / Jilantai / 105.75 / 39.78 / 1031.8
52418 / Dunhuang / 94.68 / 40.15 / 1139.0
52436 / Yumen / 97.03 / 40.27 / 1526.0
52446 / Dingxin / 99.52 / 40.30 / 1177.4
51720 / Keping / 79.05 / 40.50 / 1161.8
51730 / Alar / 81.05 / 40.50 / 1012.2
52424 / Anxi / 95.77 / 40.53 / 1170.8
51765 / Tikanlik / 87.70 / 40.63 / 846.0
52495 / Bayan Mod / 104.50 / 40.75 / 1328.1
53513 / Linhe / 107.42 / 40.75 / 1039.3
51628 / Aksu / 80.23 / 41.17 / 1103.8
52378 / Guaizi Lake / 102.37 / 41.37 / 960.0
52313 / Hongliu River / 94.67 / 41.53 / 1700.0
53336 / Urat Banner / 108.52 / 41.57 / 1288.0
52323 / Mazong Mountain / 96.88 / 41.58 / 1962.7
51644 / Kuqa / 82.95 / 41.72 / 1099.0
51656 / Korla / 86.13 / 41.75 / 931.5
51633 / Baicheng / 81.90 / 41.78 / 1229.2
51642 / Luntai / 84.25 / 41.78 / 976.1
52267 / Ejin Banner / 101.07 / 41.95 / 940.5
51567 / Karasahr / 86.57 / 42.08 / 1055.8
51526 / Kumush / 88.22 / 42.23 / 922.4
53276 / Zhurihe / 112.90 / 42.40 / 1150.8
53149 / Mandula / 110.13 / 42.53 / 1225.2
51467 / Balguntay / 86.33 / 42.67 / 1752.5
52203 / Hami / 93.52 / 42.82 / 737.2
51573 / Turpan / 89.20 / 42.93 / 34.5
51477 / Dabancheng / 88.32 / 43.35 / 1103.5
51495 / Qijiaojing / 91.63 / 43.48 / 873.2
52101 / Barkol / 93.00 / 43.60 / 1650.0
53068 / Erenhot / 111.97 / 43.65 / 964.7
53195 / Sonid Left Banner / 113.72 / 43.83 / 1111.4
51379 / Qitai / 89.57 / 44.02 / 793.5
53192 / Abag Banner / 114.95 / 44.02 / 1126.1
51365 / Caijia Lake / 87.53 / 44.20 / 440.5
51346 / Wusu / 84.67 / 44.43 / 478.7
51334 / Jinghe / 82.90 / 44.62 / 320.1
53083 / Naran bulag / 114.15 / 44.62 / 1181.6
51232 / Alataw Pass / 82.58 / 45.18 / 284.8
51288 / Baytik Mountain / 90.53 / 45.37 / 1653.7
51243 / Karamay / 84.85 / 45.60 / 427.3
51241 / Tuoli / 83.60 / 45.93 / 1077.8
51186 / Qinggil / 90.38 / 46.67 / 1218.2
51156 / Hoboksar / 85.72 / 46.78 / 1291.6
51068 / Fuhai / 87.47 / 47.12 / 500.9
51059 / Jeminay / 85.87 / 47.43 / 984.1
51076 / Aletai / 88.08 / 47.73 / 735.3
51053 / Haba River / 86.40 / 48.05 / 532.6
50603 / Xin Barag Right Banner / 116.82 / 48.67 / 554.2
Fig. S1 The flow chart of the methodology.
References
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. Agric. For. Meteorol. 189, 11-18.
Dr. Simpson, J., Kummerow, C., Tao, W.K., Adler, R.F., 1996. On the Tropical Rainfall Measuring Mission (TRMM). Meteorol. Atmos. Phys. 60, 19-36.
Fensholt, R., Rasmussen, K., Nielsen, T.T., Mbow, C., 2009. Evaluation of earth observation based long term vegetation trends-Intercomparing NDVI time series trend analysis consistency of Sahel from AVHRR GIMMS, Terra MODIS and SPOT VGT data. Remote Sens. Environ. 113, 1886-1898.
Graham, M.H., 2003. Confronting multicollinearity in ecological multiple regression. Ecology 84, 2809-2815.
Guo, Q., Hu, Z., Li, S., Li, X., Sun, X., Yu, G., 2012. Spatial variations in aboveground net primary productivity along a climate gradient in Eurasian temperate grassland: effects of mean annual precipitation and its seasonal distribution. Glob. Change Biol. 18, 3624-3631.
Ma, M.G., Frank, V., 2006. Interannual variability of vegetation cover in the Chinese Heihe river Basin and its relation to meteorological parameters. Int. J. Remote Sens. 27, 3473-3486.
Ma, W., Yang, Y., He, J., Hui, Z., Fang, J., 2008. Above- and belowground biomass in relation to environmental factors in temperate grasslands, Inner Mongolia. Sci. China-Life Sci. 51, 263-270.
Mao, D., Wang, Z., Li, L., Ma, W., 2014. Spatiotemporal dynamics of grassland aboveground net primary productivity and its association with climatic pattern and changes in Northern China. Ecol. Indic. 41, 40-48.
Myneni, R.B., Williams, D.L., 1994. On the relationship between FAPAR and NDVI. Remote Sens. Environ. 49, 200-211.
Potter, C.S., Klooster, S., Brooks, V., 1999. Interannual variability in terrestrial net primary production: Exploration of trends and controls on regional to global scales. Ecosystems 2, 36-48.
Rahimi, S., Sefidkouhi, M.A.G., Raeini-Sarjaz, M., Valipour, M., 2015. Estimation of actual evapotranspiration by using MODIS images (a case study: Tajan catchment). Arch. Agron. Soil Sci. 61, 695-709.