Simulation of site-scale water fluxes in desert and natural oasis ecosystems of the arid region in Northwest China

The study of water fluxes is important to better understand hydrological cycles in arid regions. Data-driven machine learning models have been recently applied to water flux simulation. Previous studies have built site-scale simulation models of water fluxes for individual sites separately, requiring a large amount of data from each site and significant computation time. For arid areas, there is no consensus as to the optimal model and variable selection method to simulate water fluxes. Using data from seven flux observation sites in the arid region of Northwest China, this study compared the performance of random forest (RF), support vector machine (SVM), back propagation neural network (BPNN), and multiple linear regression (MLR) models in simulating water fluxes. Additionally, the study investigated inter-annual and seasonal variation in water fluxes and the dominant drivers of this variation at different sites. A universal simulation model for water flux was built using the RF approach and key variables as determined by MLR, incorporating data from all sites. Model performance of the SVM algorithm (R² = 0.25–0.90) was slightly worse than that of the RF algorithm (R² = 0.41–0.91); the BPNN algorithm performed poorly in most cases (R² = 0.15–0.88). Similarly, the MLR results were limited and unreliable (R² = 0.00–0.66). Using the universal RF model, annual water fluxes were found to be much higher than the precipitation received at each site, and natural oases showed higher fluxes than desert ecosystems. Water fluxes were highest during the growing season (May–September) and lowest during the non-growing season (October–April). Furthermore, the dominant drivers of water flux variation were various among different sites, but the normalized difference vegetation index (NDVI), soil moisture and soil temperature were important at most sites. This study provides useful insights for simulating water fluxes in desert and oasis ecosystems, understanding patterns of variation and the underlying mechanisms. Besides, these results can make a contribution as the decision-making basis to the water management in desert and oasis ecosystems.     

Mapping of Regional Soil Salinities in Xinjiang and Strategies for Amelioration and Management

Information on the spatial distribution of soil salinity can be used as guidance in avoiding the continued degradation of land and water resources by better informing policy makers. However, most regional soil-salinity maps are produced through a conventional direct-linking method derived from historic observations. Such maps lack spatial details and are limited in describing the evolution of soil salinization in particular instances. To overcome these limitations, we employed a method that included an integrative hierarchical-sampling strategy(IHSS) and the Soil Land Inference Model(So LIM) to map soil salinity over a regional area. A fuzzy c-means(FCM) classifier is performed to generate three measures, comprising representative grade, representative area, and representative level(membership). IHSS employs these three measures to ascertain how many representative samples are appropriate. Through this synergetic assessment, representative samples are obtained and their soil-salinity values are measured. These samples are input to So LIM, which is constructed based on fuzzy logic, to calculate the soil-forming environmental similarities between representative samples and other locations. Finally, a detailed soil-salinity map is produced through an averaging function that is linearly weighted, which is used to integrate the soil salinity value and soil similarity. This case study, in the Uyghur Autonomous Region of Xinjiang of China, demonstrates that the employed method can produce soil salinity map at a higher level of spatial detail and accuracy. Twenty-three representative points are determined. The results show that 1) the prediction is appropriate in Kuqa Oasis(R2 = 0.70, RPD = 1.55, RMSE = 12.86) and Keriya Oasis(R2 = 0.75, RPD = 1.66, RMSE = 10.92), that in Fubei Oasis(R2 = 0.77, RPD = 2.01, RMSE = 6.32) perform little better than in those two oases, according to the evaluation criterion. 2) Based on all validation samples from three oases, accuracy estimation show the employed method(R2 = 0.74, RPD = 1.67, RMSE = 11.18) performed better than the multiple linear regression model(R2 = 0.60, RPD = 1.47, RMSE = 14.45). 3) The statistical result show that approximately half(48.07%) of the study area has changed to salt-affected soil, mainly distributed in downstream of oases, around lakes, on both sides of rivers and more serious in the southern than the northern Xinjiang. To deal with this issue, a couple of strategies involving soil-salinity monitoring, water management, and plant diversification are proposed, to reduce soil salinization. Finally, this study concludes that the employed method can serve as an alternative model for soil-salinity mapping on a large scale.     

基于相关分析和自适应遗传算法的盐渍化建模变量和参数优选

机器学习结合遥感等其他数据反演土壤盐分含量（Soil Salt Content, SSC）较少关注对模型精度影响较大的建模特征变量和模型参数的优选。本文基于自适应遗传算法（Adaptive Genetic Algorithm, AGA）同步优选建模特征变量和模型参数的支持向量回归（Support Vector Regression, SVR）算法反演三工河流域2016年SSC,并分析其在不同土地利用类型的分布特征。建模特征变量和模型参数的同步优选及实验设计如下：首先基于Landsat 8 OLI和SRTM高程数据提取7类共40个盐渍化相关因子,经相关分析初步筛选出候选特征变量,分别代入AGA、遗传算法（Genetic Algorithm, GA）和格网搜索算法（Grid Search, GS）同步优选SVR的建模特征变量和模型参数,并建立盐渍化监测模型（AGA-SVR、GA-SVR、GS-SVR）。结果表明：① AGA-SVR精度最优,GA-SVR次之,GS-SVR最差,相较于GS-SVR,AGA-SVR的R²/RMSE提高了44.65%;② 三工河流域非、轻度、中度、重度盐渍地和盐土的面积占比分别为42.83%、11.02%、15.88%、9.22%、21.05%;③ 草地和未利用地主要以非盐渍地和盐土为主,耕地和林地中非盐渍地分布比例均为最大;不同土地利用类型的SSC均值和标准差均呈现未利用地>草地>耕地>林地的规律。本研究的建模特征变量和模型参数的优选方法可在一定程度上提高盐渍化监测的精度。关键词：盐渍化;遗传算法;机器学习;特征优选;参数优化;土壤盐分含量;土地利用;相关分析     

天山北坡三工河流域中山带森林发育与气候土壤的关系

开展干旱区山地森林发育状况及其土壤因素影响的分析,对干旱区乃至全球山地森林带的成因研究具有重要的理论意义。以天山北坡三工河流域为研究区,主要利用森林调查与遥感影像数据,确定该流域森林带的分布状况,并结合流域气象与土壤采样数据,重点分析气候土壤因素对森林发育的影响。结果表明:1.该流域森林带位于海拔1 510～2 720 m,胸径与树高随海拔增加呈双峰曲线;其中胸径两个峰值分别位于约海拔2 000 m与2 550m,而树高峰值分别为海拔2 100 m与2 600 m,均稍高于胸径的峰值海拔高度;2.在森林带内,年均温随海拔高度增加呈线性下降趋势,最冷月均温(1月)则表现先增加后减小趋势;与其他地区相比,该流域高山林线年均温较高,最冷月均温相差较大,而最热月均温差异不明显;年降水量呈先增加后减小的趋势,且在海拔2 000 m左右达到最大值。土壤属性随海拔递增呈规律性的变化趋势:森林带内海拔约2 000～2 700 m树木发育较好,其有机质、全磷及全氮含量较高;CaCO3,pH值及电导率最小值与海拔2 000 m的最大降水带恰好吻合;土壤A层(0～10 cm)有机质、全磷及全氮含量与B(10～30 cm)、C(>3...     

巴里坤盆地现代气候波动及近250年来的冷暖变化

通过单相关普查找出巴里坤山区年表7的生长限制因子为上年6—7月均温,从而得到了巴里坤近250a来冷暖变化的替代序列。该序列的低通滤波曲线,揭示了近250a来的冷暖变化事实。方差分析指出,巴里坤冷暖变化的主要周期为126a。温度曲线与天山冰川进退时期相吻合。     

地气耦合与灾害的某些讨论

地气耦合是未来灾害学中进一步要研究的问题,下面列举某些问题的讨论。1.冷空气活动朔、望和弦时冷空气活动较多。前人用月球对大气的引力来解释其相关性。我们从地气耦合的观点认为应补充以下的相关机制。月亮的引潮力在中低纬度较大,据观测,在朔望时地下气体易于逸出,故朔望时在中低纬度地区可能有地下热气逸出使中低     

干旱区绿洲土地利用与覆被变化过程

应用遥感和GIS技术及数理统计学方法,以新疆三工河流域冲洪积扇－冲积平原型绿洲为研究区,利用1978年、1987年和1998年的遥感数据,深入研究了70年代中后期以来干旱区绿洲LUCC的过程和趋势。结果表明：(1) 遥感和GIS在绿洲土地利用与覆被变化数据获取和空间分析方面是一种非常强大而有效的技术手段;(2) 通过建立绿洲LUCC的幅度、速度、区域差异和变化趋势数学模型,可以较好的研究绿洲土地LUCC的过程和趋势,如变化的双向性、单向性、正向性和反向性,变化的涨势或落势等,在三工河流域绿洲应用取得良好效果;(3) 自70年代中后期以来三工河流域绿洲LUCC趋缓,并存在显著的时空差异：前期绿洲LUCC处于单向转换过程的非平衡状态,并呈现一定的地段分异,其中绿洲上部处于准平衡状态,而绿洲中部和下部处于非平衡状态。后期绿洲LUCC呈现双向转换为主的发展态势,使绿洲LUCC处于准平衡状态;绿洲上部和中部处于非平衡状态,绿洲下部处于平衡状态,均表现出不同的变化过程和趋势。     

Soil properties at the tree limits of the coniferous forest in response to varying environmental conditions in the Tianshan Mountains,Northwest China

The continuous coniferous forest in the Tianshan Mountains primarily consists of Picea schrenkiana. The forest forms an unbroken 1,000 km forest zone extending from west to east on the northern slope of the Tianshan Mountains, where environmental conditions such as precipitation, temperature and terrain vary greatly. To gain insight into the differences between soil properties at the upper and lower limits of P. schrenkiana, soil underlying the forests was studied along 4 transects located at Jinghe, Manasi, Fukang, and Balikun, where the natural conditions are typical and therefore representative of western, mid-western, mid-eastern and eastern portions of the Tianshan Mountains. The soil type under the coniferous forest on the northern slope of the Tianshan Mountains is gray cinnamon forest soil. For each transect, soil samples were collected from three different depths (0–5, 5–30 and 30–70 cm) in the area near the upper and lower limits of P. schrenkiana. A total of 48 soil samples were collected from 16 soil profiles and analyzed in two different laboratories. The soil pH, organic matter (OM), total N, total P, CaCO₃, and electrical conductivity (EC) were analyzed using common pedological methods. The continuous distribution of P. schrenkiana on the northern slope of the Tianshan Mountains was caused by combined water-heat conditions and pedological factors. At the upper and lower tree limits, where differences existed in the soil parent materials, precipitation, topography, combined water-heat conditions, and local atmospheric pattern, the soil properties showed a close relationship with the local environmental conditions. Due to the presence of high precipitation over the Manasi transect, the soil was fertile and had a low CaCO₃ content. In contrast, low soil fertility with high CaCO₃ was observed in the soil along the Jinghe and Balikun transects, where there was low precipitation. Although the soil properties varied on the northern slope of the Tianshan Mountains due to the presence of different environmental conditions across the 1,000-km forest zone, this region was suitable for the growth of P. schrenkiana. Taken together, the results of this study indicated that soil played an important role in controlling the continuous distribution of the P. schrenkiana along the northern slope of the Tianshan Mountains.     

柴达木盆地昆北油田路乐河组辫状三角洲沉积特征

柴达木盆地昆北油田路乐河组主要为砂砾岩、含砾砂岩、砂岩、粉砂岩及泥岩构成的陆源碎屑岩组合,通过详细的岩心特征和区域地质分析,将其确定为辫状三角洲沉积,其微相包括(水下)分流河道、河口坝、席状砂、泛滥平原、分流间湾和前三角洲泥等,其中以(水下)分流河道沉积为主体。该辫状三角洲的纵向岩石组合序列反映为典型的退积型三角洲,即发育于可容纳空间与沉积物补给速率比值(A/S比值)逐渐增大的湖侵环境。沉积旋回早期发育辫状三角洲平原,旋回中后期发育辫状三角洲前缘和前三角洲。辫状三角洲整体呈自南西向北东的扇形展布,平面上发育两期分支物源体系,分别位于Q6和Q16井区。实验分析测试、试油等资料揭示辫状三角洲(水下)分流河道砂体储集性能较好,它的发育控制路乐河组油气的纵向与平面分布。     

新疆夏季降水时空分布的适用性评估

新疆气象站点稀疏且分布不均,高精度时空气象数据缺乏。基于数据同化的再分析资料,可成为解决这一问题的有效途径。利用美国国家环境预报中心再分析数据(CFSR)、欧洲中期数值预报中心再分析数据(ERA-Interim)和美国国家航空航天局再分析数据(MERRA)中的降水数据,分别与1979-2007年新疆气象观测数据和日本气象厅高分辨率亚洲陆地降水数据(APHRO)进行数理统计分析,评估了这3套再分析数据在新疆的适用性。3套再分析数据可有效表征新疆大部分地区年内降水的时空分布特征,夏季降水偏差小于100%;但未能捕捉到夏季降水的长期趋势。夏季降水的偏差与高程具有显著的相关性,这可为订正3套再分析数据、提高降水数据的精度提供技术支撑。