Processes and trends of the land use change in Aksu watershed in the central Asia from 1960 to 2008

Land use change (LUC) in trans-boundary watersheds is of great importance to environmental assessment. The Aksu River is the largest trans-boundary river crossing Kyrgyzstan and China,but there was little information on the LUC of the watershed. We quantitatively investigated the processes and trends of its LUC by using analytic models based on the land use data derived from the remote sensing images and topographic maps. The LUC was in the quasi-balanced status with a slight difference between the loss and the gain of the area for most land use types during the period of 1960-1990,whereas transferred to the unbalanced status with significant difference between the loss and gain of the area during the period of 1990-2008. At the same time,land conversion direction changed from two-way transition to one-way transition for the most land use types. The integrated rate of net change of land use during the period of 1990-2008 is 2.1 times of that during the period of 1960-1990. Information on the processes and trends of LUC is valuable for better understanding the environmental changes across the whole trans-boundary watershed,and helpful to decision-making management for Kyrgyzstan and China.     

Uncertainties of snow cover extraction caused by the nature of topography and underlying surface

Manas River,the largest inland river to the north of the Tianshan Mountains,provides important water resources for human production and living.The seasonal snow cover and snowmelt play essential roles in the regulation of spring runoff in the Manas River Basin(MRB).Snow cover is one of the most significant input parameters for obtaining accurate simulations and predictions of spring runoff.Therefore,it is especially important to extract snow-covered area correctly in the MRB.In this study,we qualitatively and quantitatively analyzed the uncertainties of snow cover extraction caused by the terrain factors and land cover types using TM and DEM data,along with the Per(the ratio of the difference between snow-covered area extracted by the Normalized Difference Snow Index(NDSI) method and visual interpretation method to the actual snow-covered area) and roughness.The results indicated that the difference of snow-covered area extracted by the two methods was primarily reflected in the snow boundary and shadowy areas.The value of Per varied significantly in different elevation zones.That is,the value generally presented a normal distribution with the increase of elevation.The peak value of Per occurred in the elevation zone of 3,700–4,200 m.Aspects caused the uncertainties of snow cover extraction with the order of sunny slopesemi-shady and semi-sunny slopeshady slope,due to the differences in solar radiation received by each aspect.Regarding the influences of various land cover types on snow cover extraction in the study area,bare rock was more influential on snow cover extraction than grassland.Moreover,shrub had the weakest impact on snow cover extraction.     

Responses of vegetation cover to the Grain for Green Program and their driving forces in the He-Long region of the middle reaches of the Yellow River

The implementation of the Grain for Green Program is a great breakthrough in the history of China's ecological environment construction,which can control soil erosion effectively,increase land productivity and improve the ecological environment.To investigate the eco-environmental benefits brought by the Grain for Green Program,the spatiotemporal variations of vegetation cover in the growing season from 2000 to 2010 across the Hekou-Longmen(He-Long) region were analyzed by using remote sensing information,meteorological data and land use data.Moreover,the impacts of climate and human activities on vegetation change were evaluated objectively.Annual vegetation cover in the growing season increased very significantly.Increased vegetation cover occurred in 98.7% of the region,of which the area for vegetation cover improved slightly constituted 79.8% of the whole area.Vegetation moderately improved was mainly distributed in the south of the He-Long region,covering 9.6% of the area,and the area for vegetation basically unchanged concentrated in the middle and upper reaches of the Wuding River.Precipitation was found to be an important natural factor influencing vegetation cover change.The area of vegetation cover showing a significantly positive correlation with precipitation occupied 22.14% of the region.As driven by policies from the Grain for Green Program,forestland increased significantly and land use structure became more intensive.Human activities played a positive and effective role in the protection,restoration and improvement of vegetation in the places where vegetation cover was basically unchanged,even though precipitation declined greatly,and vegetation improved moderately with massive increases of forestland and grassland.     

Long-term oscillation of drought conditions in the western China:an analysis of PDSI on a decadal scale

Water resource availability is one of the primary limiting factors with regard to ecosystems in the western China.Having a clear understanding of multi-scale drought patterns in this region is a key step for adaption and mitigation to climate change.The Palmer drought severity index(PDSI) is a widely applied index to assess drought conditions.In this study,long-term monthly self-calibrated PDSI data from 1951 to 2012 were examined for drought spatiotemporal variations in the western China.The results clearly indicated that apparent spatial heterogeneities were evidenced between two sub-regions(arid land with annual precipitation less than 200 mm and semiarid land with annual precipitation between 200 to 500 mm) as well as in the entire region of the western China.Ensemble empirical mode decomposition(EEMD) analyses on monthly PDSI and other atmospheric variable time-series obtained from the Department of Civil and Environmental Engineering,Princeton University revealed that all monthly time-series of variables could be completely decomposed into eight intrinsic mode functions(IMFs) and a trend(residual).This indicates that the monthly PDSI and atmospheric variables of the semiarid area in the western China contain eight quasi-period oscillations on various timescale spanning,seasonal to decadal cycles and a trend of a larger timescale from 1951–2012.The multi-scale drought patterns identified in this research could be powerful supports for decision-making regarding coping with droughts in this region.     

Influences of landform as a confounding variable on SOM-NDVI association in semiarid Ordos Plateau

Soil organic matter(SOM) plays an important role in maintaining vegetation cover and thus mitigating land erosion of fragile terrestrial ecosystems such as in the Northern Ordos Plateau of China(NOPC).However,little information is available on whether and how SOM varies spatially as an intrinsic characteristic of landform in NOPC.The objective of this study was to examine the spatial associations of SOM with landform and vegetation cover.The study was conducted in a 23,000-km2 area within NOPC because this area has landforms of mobile dunes(MD),flat dunes(FD),grassy sandy land(GSL),flat sandy bedrocks(FSB),and swamps and salt lakes(SW),which are typical landforms in semiarid ecosystems.SOM was determined using a standard laboratory analysis method for 5 cm topsoil samples collected at 72 locations across the study area.In addition,the 250 m Multitemporal Moderate Resolution Imaging Spectroradiometer(MODIS) imageries taken in the period from August 2006 to August 2010 were used to extract Normalized Difference Vegetation Index(NDVI) which in turn was used as the surrogate of vegetation cover.Classic and geostatistical methods were used to compare SOM concentration across different landforms.The results indicated that an area with a greater value for NDVI(i.e.better vegetation cover) tended to have a higher SOM concentration regardless of the landform types.However,the association between SOM and NDVI varied from one landform to another.The SW and GSL had a highest SOM concentration,while MD had a lowest concentration.For the study area as a whole and the FD,GSL,and MD,SOM was found to be the sole function of NDVI,whereas,for the FSB,SOM was influenced by several intrinsic variables,namely ground surface altitude,slope,and aspect,as well as NDVI.SOM for the SW landform was found to be a function of NDVI.Furthermore,SOM and NDVI exhibited a consistent spatial pattern of increasing from north to south and from west to east.The highest SOM concentration of 3.5% occurred along an east-westward belt,which is adjacent to water pathways,in the mid part of the study area.     

Summer atmospheric boundary layer structure in the hinterland of Taklimakan Desert,China

Understanding the characteristics of the structure of desert atmospheric boundary layer and its land surface process is of great importance to the simulations of regional weather and climate.To investigate the atmospheric boundary layer structure and its forming mechanism of Taklimakan Desert,and to improve the accuracy and precision of regional weather and climate simulations,we carried out a GPS radiosonde observation experiment in the hinterland of Taklimakan Desert from 25 June to 3 July,2015.Utilizing the densely observed sounding data,we analyzed the vertical structures of daytime convective boundary layer and nighttime stable boundary layer in summer over this region,and also discussed the impacts of sand-dust and precipitation events on the desert atmospheric boundary layer structure.In summer,the convective boundary layer in the hinterland of Taklimakan Desert developed profoundly and its maximum height could achieve 4,000 m;the stable boundary layer at nighttime was about 400–800-m thick and the residual mixing layer above it could achieve a thickness over 3,000 m.Sand-dust weather would damage the structures of nighttime stable boundary layer and daytime convective boundary layer,and the dust particle swarm can weak the solar radiation absorbed by the ground surface and further restrain the strong development of convective boundary layer in the daytime.Severe convective precipitation process can change the heat from the ground surface to the atmosphere in a very short time,and similarly can damage the structure of desert atmospheric boundary layer remarkably.Moreover,the height of atmospheric boundary layer was very low when raining.Our study verified the phenomenon that the atmospheric boundary layer with supernormal thickness exists over Taklimakan Desert in summer,which could provide a reference and scientific bases for the regional numerical models to better represent the desert atmospheric boundary layer structure.     

A new parallel framework of distributed SWAT calibration

With the development of large-scale hydrologic modeling, computational efficiency is becoming more and more important. Rapid modeling and analysis are needed to deal with emergency environmental disasters. The Soil and Water Assessment Tool(SWAT) is a popular hydrologic model, which is less applied in large-scale watershed simulation because of its sequential characteristics. For improving the computational efficiency of the SWAT model, we present a new parallel processing solution for hydrologic cycle and calibration based on MPI(Message Passing Interface). We partitioned sub-basins during the processes based on a load balancing method. Then the calibration was parallelized using a master-slave scheme, in which different input parameters were allocated to different processes to run the hydrologic cycle and compute the function value. Because of the slow convergence and local optimization of the SCE-UA(Shuffled Complex Evolution-developed by University of Arizona) algorithm in SWAT calibration, a genetic algorithm(GA) is developed to optimize the calibration step. Then by dividing the default communicator into several sub-communicators, all the hydrologic cycles were parallelized in their own sub-communicators to achieve further acceleration. In this paper the results show speedups for the hydrologic cycle calculations, as well as in the optimized calibration step. In the case study, we tested the parallel hydrologic cycle by four processes, and got a speedup of 3.06. In the calibration section, after applying the GA optimization, with 10 cores, we got a speed increase of 8.0 in our GA parallel framework compared with the GA sequential calibration, which is much better than the original SWAT calibration. After the sub-communicators added, this process was speeded up even further. The study demonstrated that the GA parallel framework with multi-sub-communicators is an effective and efficient solution for the hydrologists in large scale hydrology simulations.     

Factors determining soil water heterogeneity on the Chinese Loess Plateau as based on an empirical mode decomposition method

Soil water is a critical resource, and as such is the focus of considerable physical research. Characterization of the distribution and spatial variability of soil water content(SWC) offers important agronomic and environmental information. Estimation of non-stationary and non-linear SWC distribution at different scales is a research challenge. Based on this context, we performed a case study on the Chinese Loess Plateau, with objectives of investigating spatial variability of SWC and soil properties(i.e., soil particle composition, organic matter and bulk density), and determining multi-scale correlations between SWC and soil properties. A total of 86 in situ sampling sites were selected and 516 soil samples(0–60 cm depth with an interval of 10 cm) were collected in May and June of 2019 along the Yangling-Wugong-Qianxian transect, with a length of 25.5 km, in a typical wheat-corn rotation region of the Chinese Loess Plateau. Classical statistics and empirical mode decomposition(EMD) method were applied to evaluate characteristics of the overall and scale-specific spatial variation of SWC, and to explore scale-specific correlations between SWC and soil properties. Results showed that the spatial variability of SWC along the Yangling-Wugong-Qianxian transect was medium to weak, with a variability coefficient range of 0.06–0.18, and it was gradually decreased as scale increased. We categorized the overall SWC for each soil layer under an intrinsic mode function(IMF) number based on the scale of occurrence, and found that the component IMF1 exhibited the largest contribution rates of 36.45%–56.70%. Additionally, by using EMD method, we categorized the general variation of SWC under different numbers of IMFs according to occurrence scale, and the results showed that the calculated scales among SWC for each soil layer increased in correspondence with higher IMF numbers. Approximately 78.00% of the total variance of SWC was extracted in IMF1 and IMF2. Generally, soil texture was the dominant control on SWC, and the influence of the three types of soil properties(soil particle composition, organic matter and bulk density) was more prominent at larger scales along the sampling transect. The influential factors of soil water spatial distribution can be identified and ranked on the basis of the decomposed signal from the current approach, thereby providing critical information for other researchers and natural resource managers.     

N and P resorption in a pioneer shrub （Artemisia halodendron） inhabiting severely desertified lands of Northern China

Nutrient resorption is an important conservation mechanism for plants to overcome nutrient limitation in the less fertile area of desertified land.In the semi-arid Horqin Sandy Land of Northern China,the shrub Artemisia halodendron usually colonizes into the bare ground of severely desertified land as a pioneer species.It is,therefore,expected that A.halodendron will be less dependent on current nutrient uptake through efficient and proficient resorption of nutrients.In this study,we found that averaged nitrogen(N)and phosphorus(P)concentrations in senesced leaves significantly varied from 12.3 and 1.2 mg/g in the shifting sand dune to 15.9 and 1.9 mg/g in the fixed sand dune,respectively,suggesting that foliar N and P resorption of A.halodendron were more proficient in the shifting sand dune.In particular,positive relationships between nutrient concentrations in senesced leaves and soil nutrient availability indicate that A.halodendron in infertile habitats is more likely to manage with a low level of nutrients in senesced leaves,giving this species an advantage in infertile soil.Moreover,foliar N-and P-resorption efficiencies and proficiencies showed limited inter-annual variability although annual precipitation varied greatly among 2007–2009.However,N and P resorption of A.halodendron were not more efficient and proficient than those previously reported for other shrubs,indicating that the pioneer shrub in sand dune environments does not rely more heavily than other plants on the process of resorption to conserve nutrients.Incomplete resorption of nutrients in A.halodendron suggests that senesced-leaf fall would return litter with high quality to the soil,and thereby would indirectly improve soil nutrient availability.The restoration of desertified land,therefore,may be accelerated after A.halodendron pioneers into shifting sand dunes.     

Spatial pattern of land cover change in China’s semiarid environment

This study seeks a routine to quantify spatial pattern of land cover changes in semiarid environment of China based on post-classification comparison method. The method consists of three major steps: (1) the image classification and unification of classified results based on two-level land cover classification themes, (2) the establishment of land cover change classes based on an unification land cover classification theme, (3) the reclassification and mapping of land cover change classes with three overall classes including no-change, gain and loss based on the unification land cover class. This method was applied to detect the spatial pattern of land cover changes in Yinchuan Plain, one of famous irrigation agricultural zones of the Yellow River, China. The results showed the land cover had undergone a remarkable change from 1991 to 2002 in the study area (the changed area was over 30%). Rapid increase of cropland (12.5%), built-up area (131.4%) and rapid decrease of bare ground (51.7%) were alarming. The spatial pattern of land cover changes showed clear regional difference in the study area and was clearly related to human activities or natural factors. Thus, it obtained a better understanding of the human impact on the fragile ecosystem of China’s semiarid environment.     

干旱区精河流域地表温度的模型反演研究

以精河流域绿洲为研究区,使用Landsat ETM⁺数据,采用单窗算法和普适性单通道算法对研究区地表温度进行反演,并将这两种算法的反演结果与研究区MODIS温度产品(MODIS LST)进行比较。结果表明：(1) 单窗算法和普适性单通道算法反演的结果总体趋势比较接近,研究区整体的平均温度相差约2k;(2) 采用改进型土壤调整植被指数(MSAVI)代替归一植被指数(NDVI)计算地表比辐射率可有效提高反演精度,并且同等条件下单窗算法的反演精度高于普适性单通道算法,两种算法的反演结果与MODIS LST的相关系数分别是0.9255和0.8651;(3) 在城镇区域,普适性单通道算法反演结果与MSAVI的相关性高于单窗算法,相关系数为0.8136,说明普适性单通道算法更适合干旱区大范围城镇地表温度的反演研究。     

用MODIS数据反演地表温度的基本参数估计方法

利用MODIS数据反演地表温度所需的2个基本参数(大气透过率和地表比辐射率)的估计方法,由MODIS的可见光近红外波段计算得到大气水汽含量,根据大气透过率与大气水汽含量的关系,用辐射传输模型MODTRAN模拟得到大气透过率.由于MODIS的扫描带宽为2 330 km,遥感视角和大气温度会对大气透过率有比较大的影响,因此,提出了大气透过率的遥感器视角校正函数和温度校正函数.地表比辐射率主要取决于地表物质构成,尤其是植被覆盖程度,为此,提出了根据地表植被覆盖差异来估计地表比辐射率的方法,其中还引入了辐射比率参数及其确定方法.     

基于SEBAL模型与MODIS产品的松嫩平原蒸散量研究

利用MODIS的地表温度/比辐射率、植被指数、地表反照率3种产品,结合气象站点的观测资料,通过基于地表能量平衡方程的SEBAL模型估算了松嫩平原2008年5月16日、6月16日、7月13日、8月15日、9月13日不同下垫面的地表蒸散量,并采用涡动相关数据对模型估算结果进行验证,发现估算值与实测值的变化趋势大体一致,平均误差在20%以内,基本上可以满足区域蒸散研究的精度要求。此外,还通过GIS空间分析方法探讨了松嫩平原2008年生长季内不同时相蒸散量的时空分布特征,并分析了蒸散量土地利用类型的关系,结果表明松嫩平原蒸散量的空间分布在很大程度上受气候特征、土壤供水条件及土地利用类型的影响,各时相的蒸散量大致表现为从西南部向东北部逐渐增加的变化趋势,水体、林地、沼泽湿地具有较高的蒸散量,耕地、居工地次之,草地的蒸散量最低。     

A remote sensing-based agricultural drought indicator and its implementation over a semi-arid region,Jordan

The objective of the study was to develop a remote sensing(i.e., Landsat-8 and MODIS)-based agricultural drought indicator(ADI) at 30-m spatial resolution and 8-day temporal resolution and also to evaluate its performance over a heterogeneous agriculture dominant semi-arid region in Jordan. Firstly, we used principal component analysis(PCA) to evaluate the correlations among six commonly used remote sensing-derived agricultural drought related variables. The variables included normalized difference water index(NDWI), normalized difference vegetation index(NDVI), visible and shortwave drought index(VSDI), normalized multiband drought index(NMDI), moisture stress index(MSI), and land surface temperature(LST). Secondly, we integrated the relatively less correlated variables(that were found to be NDWI, VSDI, and LST) to generate four agricultural drought categories/conditions(i.e., wet, mild drought, moderate drought, and severe drought). Finally, we evaluated the ADI maps against a set of 8-day ground-based standardized precipitation index values(i.e., SPI-1, SPI-2, …, SPI-8) by use of confusion matrices and observed the best results for SPI-4(i.e., overall accuracy and Kappa-values were 83% and 76%, respectively) and SPI-5(i.e., overall accuracy and Kappa-values were 85% and 78%, respectively). The results demonstrated that the method would be valuable for monitoring agricultural drought conditions in semi-arid regions at both a reasonably high spatial resolution(i.e., 30-m) and a short time period(i.e., 8-day).     

MODIS数据反演地表温度的应用研究

在近年来提出的两种温度劈窗算法中,研究采用Sobrino等(2003)的辟窗算法和NDVI阈值法获取地表比辐射率,并结合Mao等(2005)劈窗算法中对大气水汽含量的估算,反演了山西省三个时相的地表温度。从反演的LST分布图上发现,该反演算法与LST产品比较符合。同时以其中的一个时相为例,利用LST产品对反演的精度进行了验证,结果发现反演值和产品二者的最大值、最小值和平均值的误差值分别为1.08℃、0.16℃和0.75℃,其均方根误差RMSE达1.30℃,具有较高的精度,从而表明该法反演地表温度的可行性,并化简了地表温度遥感估算的难度,在不依赖于地面实测资料的情况下,可准确快速地大面积遥测农业环境温度,有助于监测农业旱情。     

基于MODIS 的沙漠化地区植被覆盖度提取模型的研究

目前常用的植被覆盖度遥感反演模型,受限条件较多,所用系数精度不高。针对上述问题,提出了利用研究区域的实测最大和最小植被覆盖度确定反演模型系数的方法和基于MODIS影像数据的地面实测验证方法。经用该方法对毛乌素沙地沙漠化地区的植被覆盖度进行反演,并同步监测该区域的地表覆盖度对反演结果进行验证,结果表明地面同步实测植被覆盖度与反演的误差绝对值平均值为0.04,反演方法精度较高。研究表明本文提出的利用MODIS影像数据反演沙漠化地区地表覆盖度的模型、地面验证方法切实可行,适合对沙漠化地区地表植被覆盖度大范围且快速的遥感监测。     

基于MODIS数据的中国热环境时空变化研究

以2011年1、4、7、10月份的MODIS陆地表面温度产品(MOD11A2)、地表覆盖类型产品(MOD12Q1)为依据,采用GIS的空间分析,方差分析和统计分析方法,研究了中国陆地表面温度(LST)的季节变化规律,以及LST与不同下垫面类型之间的相互关系。结果表明:(1)冬季,整个区域昼夜呈现出陆地表面温度随纬度递减而逐渐上升的趋势;夏季白天,北方大部分地区陆地表面温度要比东南沿海地区高,晚上,则相反。(2)不同下垫面类型对陆地表面温度的影响较为显著,整体表现为,白天,林地、草地、水体、耕地、未利用土地、建设用地的陆地表面温度依次升高;黑夜,水体、林地、草地、耕地、建设用地、未利用土地的陆地表面温度依次降低。这种变化趋势和下垫面类型的热容量性质具有密切的关系。     

黑河流域土壤含水量遥感反演及不同地类土壤水分效应分析

利用NASA提供的MOD IS数据产品和黑河流域的野外实测数据,在充分考虑研究区的下垫面特征、数据的时间特性和遥感反演土壤含水量的模型与算法的基础上,采用了热惯量法,计算了表观热惯量,并与实测数据进行回归分析建模,反演了整个黑河流域的土壤含水量。随后构建了三个区域土壤水分效应评价指标:土壤水分单元权重指数、区域土壤水分单元权重指数和土壤水分效应贡献度,结合黑河流域的土地利用状况,定量评价了不同土地利用类型的土壤水分效应。研究表明:利用MOD IS数据产品,反演参量获取简单,可降低反演土壤含水量的复杂性,有利于大、中尺度的实际应用;利用当地积累的季节标准日变化曲线将温差修正到最大值,使反演的表观热惯量更接近真实,可提高模型精度;用多个地面采样点的土壤含水量均值与一个像元对应建模,能改善因MOD IS数据空间分辨率低而存在大量混合像元致使反演精度降低的问题;在黑河流域,中覆盖度草地、有林地、高覆盖度草地、水田等用地类型的土壤水分效应最为明显,贡献最大。     

基于MapReduce模型的生物量遥感并行反演方法研究

MapReduce模型是一种基于云计算平台下新型的并行编程模型。文中将MapReduce并行编程模型应用到遥感影像并行化处理中,以2005-2009年5a生长季期(5-10月)MODIS13Q1数据产品为数据源,对青海省三江源地区的生物量(草地总生物量和可食草量)进行并行化反演,研究基于该模型的生物量遥感并行反演方法。实验分析结果表明:基于该模型的并行生物量遥感反演结果与经过精度验证的串行反演结果一致,并行化反演结果准确、可信;并行化反演效率较串行化反演效率有大幅提高,并随着计算节点的增加,并行效率不断提高。     

中国干旱区积雪面积产品去云处理方法验证与评估

针对MODIS每日积雪产品中云覆盖现象严重这一问题,以中国干旱区作为研究对象,结合AMSR-E被动微波雪深数据,采用多时相、多传感器数据融合的方法进行去云处理,获取MODIS每日,4 d,8 d和MODIS与AMSR-E融合后的每日,4 d与8 d共6种新的积雪产品,并分别提取其积雪持续日数（SCD）。对比结果显示,MODIS与AMSR-E多传感器的阈值法4日融合产品在融合算法效率、云去除效果和融合后保持较高分类精度方面均有较好表现,其融合后的无云产品在全天气条件下具有96%的整体分类精度、80%的雪分类精度和99%的陆地分类精度,大大高于研究区原MODIS Terra-Aqua每日融合积雪产品全天候条件下64%,32%和70%的整体、雪、陆分类精度;并且由其提取的积雪持续日数不仅在最大程度上保持了原MODIS产品高的空间分辨率,而且精度较高,对研究区积雪的空间分布状况有很好反映。