Zone II • Versions EN1
Abstract: Arid region is known for its terrestrial ecosystems highly sensitive to climate change. The dynamics of ecosystem carbon pools most intuitively reflect regional carbon source/sink characteristics. Studies on the estimation and dynamics of organic carbon pools in arid region ecosystems have important scientific implications for carbon cycle studies at both regional and global scales. The study has an approximate spatial range of 30°N – 50°N and 70°E – 110°E, covering Xinjiang, northwestern Gansu, and southwestern Inner Mongolia of China. This study collects the distributional data of carbon stocks in China's arid regions from 1980 to 2014 (spatial resolution: 50km), obtained from the arid ecosystem model (AEM) simulation based on three sets of meteorological data – MERRA, ERA-Interim, and CFSR. The data obtained are then validated against corresponding data from existing literature. These datasets can be used to understand the changes of carbon stocks in arid regions of China, providing basic data support for regional ecosystem studies. They also serve national needs in the construction of the Silk Road Economic Belt and have important practical significance for the region's socioeconomic stability and sustainable development.
Keywords: arid region; carbon stock; arid ecosystem model (AEM)
|English title||Dynamic spatial datasets of ecosystem carbon stocks in arid regions of western China under climate change, 1980 – 2014|
|Corresponding author||Zhang Chi (email@example.com)|
|Data authors||Fang Xia, Zhang Chi, Zhang Yaonan, Kang Jianfang|
|Time range||From 1980 to 2014|
|Spatial resolution||50 km×50 km|
|Geographical scope||30°N – 50°N, 70°E – 110°E|
|Data format||Grid||Data volume||22.01 MB|
|Data service system||<http://www.sciencedb.cn/dataSet/handle/617>|
|Sources of funding||National Science and Technology Infrastructure Program (Y719H71006), Chinese Academy of Sciences Informatization Program (XXH13506)|
|Dataset composition||Three datasets are included: first, distribution of carbon stocks in China's arid region from 1980 to 2014 obtained through MERRA data-driving AEM model simulation; second, distribution of carbon stocks in China's arid region from 1980 to 2014 obtained through ERA-Interim data-driving AEM model simulation; third, distribution of carbon stocks in China's arid region from 1980 to 2014 obtained through CFSR data-driving AEM model simulation.|
Arid regions are short of water, ecologically fragile, and sensitive to climate change.1 The carbon cycle is a complex process in the Earth system accompanied by various material cycles and energy flows, and its dynamic changes have a major impact on the ecosystem.2 The carbon cycle in terrestrial ecosystems has been the focus of global change research. Therefore, an estimation of organic carbon pools in arid and semi-arid ecosystems and their dynamics has important scientific significance for the study of carbon cycle in regional and global scales.3
Ecosystem modeling is a powerful tool to simulate and predict carbon stocks and their dynamics in regional and global scales.4–6 However, popular models such as Biome-BGC,7 CENTURY,8 LPJ,9 OCHIDEE10 and CEVSA11 oversimplify the vegetation structure and root-water interaction processes,12–14 which make it difficult to accurately simulate arid and semi-arid ecosystems. Dryland ecosystem models like PALS,15 on the other hand, cannot simulate the groundwater uptake process by phreatophytes.16 To make up the gap, Zhang17 developed an AEM (Arid Ecosystem Model), which reflected the special physiological and ecological processes (high root-shoot ratio, vertical root distribution water absorption model, etc.) in the arid area, and can simulate the structure and function of temperate desert vegetation, and the complex interactions of ecosystem carbon-water cycles.
This study uses the AEM to examine the arid region of China, including Xinjiang, northern Gansu and northwestern Inner Mongolia of China. The three sets of MERRA, ERA-Interim and CFSR data are used to drive the AEM for simulating the carbon pool distribution in China's arid regions.
2.1 Data collection and preprocessing
The three data sets used to drive the AEM model are shown in Table 1, including: meteorological data (daily maximum temperature (°C), daily minimum temperature (°C), daily mean temperature (°C), precipitation (mm/day), shortwave radiation (W·m−2/day), relative humidity (%)), topographic data (elevation, aspect and slope), soil data (silt, clay and sand), vegetation functional type map, CO2 concentration, etc. The model input data use daily scale as the time resolution, and the output data use annual scale.
During the processing of meteorological data, the downloaded NetCDF (Network Common Data Format) file for China's Arid Zone 1980 – 2014 is first converted into grid format using Python scripts. Then, the data of hour scales are downloaded and integrated into daily scales using the CDO (Climate Data Operators) tool and Python scripts for obtaining relative humidity. The data are checked, reprojected, masked, and converted into bin format using the AML language. The data are integrated into annual scales and saved as the final model input data.
The soil data required for the model are derived from the World Soil Database version 1.2,18 with a spatial resolution of 1 km × 1 km. The data are resampled to obtain a 50 km × 50 km spatial resolution, in order to be consistent with other input data.
Topographic data include information on slope, aspects, and elevation. Elevation is derived from the 30m resolution ASTER (Advanced Space-borne Thermal Emission and Reflection Radiometer).19 The downloaded data are used to generate the spatial distribution of slope and aspect assisted by the ArcGIS surface analysis tool. Then the data are aggregated or resampled into a spatial resolution of 50 km × 50 km.
The plant functional types (PFTs) in the arid region of China constitute the basic model input dataset, which determines the vegetation distributional pattern of the model input and output. Seven major plant functional types in the arid region of China were identified, including: temperate evergreen needleleaf forest, temperate deciduous broadleaf forest (Populus euphratica in the case of Xinjiang), grassland, nonphreatophytic shrub (represented by Haloxylon), phreatophytic shrub (represented by Tamarix), irrigated cropland, and meadow. The distribution of major plant functional types (PFTs) is based on the 1:1,000,000 vegetation map of the People's Republic of China,20 while the CO2 dataset is mainly derived from the global annual average concentration data supplied by the US Mauna Loa Observing Carbon site (http://co2now.org).
|1||MERRA||1980 – 2014||NASA|
|2||ERA-Interim||1980 – 2014||European Mesoscale Weather Forecasting Center|
|3||CFSR||1980 – 2014||US National Center for Environmental Prediction Reanalysis Climate Prediction System|
|4||Elevation, Aspect, Slope||2008||ASTER Global Digital Elevation Model dataset|
|5||Distribution of major plant functional types in the arid region of China||2007||Vegetation Map of the People's Republic of China|
|6||Soil maps in the arid region of China||2008||World Soil Database version 1.2|
2.2 Data processing steps
2.2.1 AEM model
The arid ecosystem model (AEM) is a process-based ecosystem model that incorporates physiological and ecological processes of vegetation in dryland ecosystems, including the root water uptake process of deep root desert shrubs, deep root effects and the high photosynthetic efficiency of drought-tolerant functional vegetation. The model can be used to simulate, mainly, carbon and water processes in arid ecosystems to study the effects of climate change on carbon stocks in the arid regions of China.17 AEM integrates all the carbon-nitrogen-water cycles, including biophysical modules, vegetation physiology modules, soil physics, biogeochemistry modules, and vegetation dynamic modules. Compared to other common ecosystem models, AEM addresses specific physiological and ecological processes in arid regions. AEM uses data concerning meteorology, soil, topography, vegetation distribution, atmospheric composition, and land management as the driving factors to estimate changes in carbon and nitrogen fluxes and stocks of vegetation and soil on a daily scale. It can be used to study the effects of climate change on ecosystem structure (leaf area, biomass, etc.) and function (net primary productivity, evapotranspiration, etc.) in arid regions.17
2.2.2 Carbon pool simulation
Firstly, collect and sort out data on meteorology, soil, topography and vegetation functional types in the arid regions of western China, and convert them into a uniform spatial resolution of 50 km×50 km. Secondly, adjust the key physiological and ecological parameters of the AEM model based on major vegetable functional types in the study area. The AEM model is debugged and optimized as its simulation accuracy for the arid regions is verified; based on the multi-source meteorological driving datasets and the optimal physiological and ecological parameters, the organic carbon pool in arid regions of western China is simulated and analyzed and its dynamic spatiotemporal features are obtained. The carbon pool simulation process in arid regions of western China is shown in Figure 1.
Using the three meteorological data sets of MERRA, ERA-interim and CFSR to drive the AEM model, we obtained the distribution grid of carbon stock in arid regions of western China from 1980 to 2014. The spatial resolution was 50 km × 50 km, and the geographic coordinate system was WGS1984. The data results are shown in Figure 2.
Fig.2 Distribution of carbon stocks in arid regions of western China, 1980 – 2014
4.1 Collection model evaluation
Due to the lack of field observation data, this study used data from exisiting literature to verify the simulation results. Compare vegetation carbon (VEGC) and soil organic carbon (SOC) density data of the same area under the same climatic conditions generated by the model and selected from literature. The VEGC and SOC site information and references used for the verification are shown in Table 2 and Table 3.
|Simulated VEGC (MERRA) (gC/m2)||Simulated VEGC (ERA-Inteim) (gC/m2)||Simulated VEGC (CFSR) (gC/m2)||Source|
|Simulated SOC(MERRA) (gC/m2)||Simulated SOC(ERA-Interim) (gC/m2)||Simulated SOC(CFSR) (gC/m2)||Source|
4.2 Data quality assessment
The validation shows a great consistency between existing literature data and the AEM simulation results. Among the three driving datasets, MERRA dataset generated the best simulation results, with R² greater than 0.65, followed by ERA-Interim and CFSR datasets (Figures 3 & 4).
Fig.3 VEGC simulation and validation
Fig.4 SOC simulation and validation
4.3 Simulation results evaluation
The three sets of simulation results show a relatively high carbon storage in the part of Gansu province south of Jiayuguan and the northern part of Xinjiang province in 1980-2014, especially in the Altay and Tianshan mountains of northern Xinjiang which are abundant in water, superior in natural conditions and flourishing in vegetation. This is very similar to the findings of Liu Weiguo31 and Li Chaofan.32
This study uses the arid ecosystem model AEM to simulate the distribution of carbon stocks in arid regions of western China from 1980 to 2014. Three datasets of MERRA, ERA-Interim and CFSR were used to drive the arid ecosystem model AEM and to optimize the model parameters. The results were then validated against corresponding measured data from existing literature. Finally, three datasets were obtained for the distribution of carbon pool in arid regions of western China.
The datasets provide basic data for ecosystem research in the region, and can also serve as a reference for carbon stock changes in arid regions of western China in the context of global climate change. When analyzed in combination with climate and ecological data, the datasets can be of great significance to the sustainable development of the region's ecological environment.
The carbon pool distribution data sets for arid regions of western China from 1980 to2014 are saved in a grid format. They can be read by ArcGIS, or processed in batches by using computer languages such as Python.
Thanks to NASA for providing the MERRA meteorological reanalysis data, the US National Environmental Forecast Center for providing the CFSR meteorological reanalysis data, and the European Mesoscale Weather Forecast Center for providing the ERA-Interim meteorological reanalysis data. Global annual average CO2 concentration data are provided by the Mauna Loa Observatory of the United States.
Dai A. Drought under global warming: a review. Wiley Interdisciplinary Reviews: Climate Change 2(2011): 45 – 65. DOI:10.1002/wcc.81
Tao B, Ge Q, Li K et al. Progress in the studies on carbon cycle in terrestrial ecosystem. Geographical Research 5(2001): 005.
Piao S, Ciais P, Lomas M et al. Contribution of climate change and rising CO2 to terrestrial carbon balance in East Asia: a multi-model analysis. Global and Planetary Change 75(2011): 133 – 142.
Yu GR, Zhu XJ, Fu YL et al. Spatial patterns and climate drivers of carbon fluxes in terrestrial ecosystems of China. Global Change Biology 19(2013): 798 – 810.
Fang X, Zhang C, Wang Q et al. Isolating and quantifying the effects of climate and CO2 changes (1980 – 2014) on the net primary productivity in arid and semiarid China. Forests 8(2017): 60.
Parton WJ, Stewart JWB & Cole CV. Dynamics of C, N, P and S in grassland soils: a model. Biogeochemistry 5(1988): 109 –131.
Running SW & Coughlan JC. A general model of forest ecosystem processes for regional applications I. Hydrologic balance, canopy gas exchange and primary production processes. Ecological Modelling 42(1988): 125 –154.
Prentice IC, Cramer W, Harrison SP et al. A global biome model based on plant physiology and dominance, soil properties and climate. Journal of Biogeography 19(1992): 117 –134.
Krinner G, Viovy N, De Noblet-Ducoudré N et al. A dynamic global vegetation model for studies of the coupled atmosphere–biosphere system. Global Biogeochemical Cycles 19(2005): GB1015. DOI: 10.1029/2003GB002199
Cao M, Prince SD, Li K et al. Response of terrestrial carbon uptake to climate interannual variability in China. Global Change Biology 9(2003): 536 –546.
Yang R, Friedl MA & Ni W. Parameterization of shortwave radiation fluxes for nonuniform vegetation canopies in land surface models. J. Geophys. Res. 106(2001): 14275 –14286.
Zeng X, Zeng X & Barlage M. Growing temperate shrubs over arid and semiarid regions in the Community Land Model–Dynamic Global Vegetation Model. Global Biogeochemical Cycles 22(2008): GB3003. DOI:10.1029/2007GB003014
Zhang C & Ren W. Complex Climatic and CO2 Controls on Net Primary Productivity of Temperate Dryland Ecosystems over Central Asia during 1980 – 2014. Journal of Geophysical Research 122(2017): 2356 – 2374.
Reynolds JF, Kemp PR & Tenhunen JD. Effects of long-term rainfall variability on evapotranspiration and soil water distribution in the Chihuahuan Desert: A modeling analysis. Plant Ecology 150(2000): 145 –159.
Lai CT & Katul G. The dynamic role of root-water uptake in coupling potential to actual transpiration. Advances in Water Resources 23(2000): 427 –439.
Zhang C, Li CF, Luo GP et al. Modeling plant structure and its impacts on carbon and water cycles of the Central Asian arid ecosystem in the context of climate change. Ecological Modelling 267(2013): 158 – 179.
Nachtergaele F, Van Velthuizen H, Verelst L et al. Harmonized world soil database v1.2. Food and Agriculture Organization of the United Nations, 2008.
NASA, METI. ASTER GDEM Readme File-ASTER GDEM Version 1. 2009. Available at: <http://www.ersdac.or.jp/GDEM/E/image/ASTER%20GDEM%
Zhang X, Sun S, Yong S et al. Vegetation Map of the People’s Republic of China (1: 1000000). Beijing: Geological Publishing House, 2007.
Huang J & Ji F. Effects of rising temperatures in green house and variation of irrigation on seed cotton yield, biomass and water use efficiency. Chinese Agricultural Science Bulletin 30(2014): 152 – 157.
Mu M, Shi X, Wang L et al. Measurement of biomass of cotton stalk in each part. Chinese Cotton 33(2006): 16.
Luo X, Bing C & Zhang J. Effects of nitrogen applied levels on the dynamics of biomass, nitrogen accumulation of cotton plant in different soil textures. Chinese Journal of Soil Science 41(2010): 904 – 910.
Zhang H, Luo H, Li L et al. Characteristics of root and shoot biomass accumulation in high-yield cotton fields with mulch-drip irrigation. Cotton Science 27(2015): 427 – 434.
Zhang X, Liu S, Wang P & Zhou L. Effects of different fertilizations on cotton dry matter accumulation, nutrients uptake and yield. Acta Agriculturae Boreali: Occidentalis Sinica 21(2012): 107 – 113.
Xu H, Tian L, Lin T et al. Study on effect of dry matter accumulation and distribute on the nitrogen fertilizer applied to upland cotton using drip irrigation under plastic film in southern Xinjiang. Xinjiang Agricultural Sciences 49(2012): 1765 – 1772.
Wang X, Wei C & Zhang J. Effects of irrigation methods and N application level on cotton growth and nitrogen use efficiency. Cotton Science 24(2012): 554 – 561.
Han Q, Luo G, Li C et al. Simulated grazing effects on carbon emission in Central Asia. Agricultural and Forest Meteorology 216(2016): 203 – 214.
Yang Y, Fang J, Ma W et al. Large-scale pattern of biomass partitioning across China’s grasslands. Global Ecology and Biogeography 19(2010): 268 – 277.
Yang Y, Fang J, Ma W et al. Soil carbon stock and its changes in northern China’s grasslands from 1980s to 2000s. Global Change Biology 16(2010): 3036 – 3047.
Liu W. Research of Carbon and NPP Changes of the Land Ecosystem in Xinjiang. Doctoral Dissertation, Xinjiang University, 2007.
1. Fang X, Zhang C, Zhang Y et al. Dynamic spatial datasets of ecosystem carbon stocks in arid regions of western China under climate change, 1980 – 2014. Science Data Bank. DOI: 10.11922/sciencedb.617
How to cite this article
Fang X, Zhang C, Zhang Y et al. Dynamic spatial datasets of ecosystem carbon stocks in arid regions of western China under climate change, 1980 – 2014. China Scientific Data 3(2018). DOI: 10.11922/csdata.2018.0039.zh