As a typical ecologically vulnerable area in northern China, the ecologically vulnerable area of the Loess Plateau is significantly affected by extreme rainfall events because of its unique topography and climatic conditions. Twenty-eight meteorological stations in the water-wind erosion crisscross region were selected, and eleven extreme precipitation indices were calculated using the RClimDex model. Linear correlation analysis, the Mann-Kendall trend test, and the wavelet crossing method were used to analyze the spatial-temporal distribution characteristics of extreme precipitation events in the water-wind erosion crisscross region of the Loess Plateau from 1970 to 2020. Driving factors for extreme precipitation events are discussed. The results are as follows: (1) The number of continuous dry days in the water and wind erosion interlacing zone from 1970 to 2020 shows a decreasing trend, whereas the other 10 indices show an increasing trend, reflecting the increasing frequency, magnitude, and intensity of extreme precipitation events in the study area during the past 50 years. There is a close relationship between increases in annual precipitation and extreme precipitation events, and an increase in extreme precipitation events is mainly caused by the number of moderate and heavy rain days. (2) Extreme precipitation events show an increasing trend in the entire region from 1970 to 2020, with significant extreme precipitation events occurring in the central and southwest parts of the crisscross region. The extreme precipitation and intensity in the Shaanxi section show a significant increasing trend, and the extreme degree is very significant. (3) The three extreme precipitation indices, total wet day precipitation, number of heavy rain days, and 5-day maximum precipitation, have varying degrees of influence from El Nino-Southern Oscillation, East Asian Summer Monsoon, and sunspot number (SN), with cross wavelet transform with SN having the greatest influence. This shows that the correlation between SN and extreme precipitation index is the highest among the influencing factors, and SNs have the greatest influence on extreme precipitation events.

Multisource remote-sensing precipitation products play an important role in drought monitoring in regions with few or uneven meteorological stations, such as arid areas in northwest China. In this study, five sets of typical remote-sensing precipitation products (PERSIANN, CHIRPS, CMORPH, TMPA, and MSWEP) were selected. The meteorological drought performance of the precipitation products at three timescales was evaluated based on the standardized precipitation evapotranspiration index (SPEI). The capability of remote-sensing precipitation products to capture drought events was explained by identifying drought events using the run-course theory. The results showed the following: (1) In arid northwest China, the five sets of remote-sensing precipitation products could capture the spatial distribution pattern of annual mean precipitation well, but it was difficult to accurately capture the change trend of precipitation. (2) MSWEP had the best performance in capturing SPEI, followed by TMPA, PERSIANN, and CHIRPS, and CMORPH had the worst performance. SPEI1 was the best timescale for remote-sensing precipitation products to identify meteorological droughts. (3) CHIRPS had the best recognition capability for several drought events, whereas PERSIANN had the worst. MSWEP and TMPA were the best indicators of drought severity, whereas CHIRPS was the worst. Except for CMORPH, the other four sets of products captured the intensity and extreme values of the drought events well. In summary, although the five sets of remote-sensing precipitation products could capture the drought characteristics of the northwest arid region on the whole, finding a precipitation product with the best performance in all aspects of capturing drought characteristics was difficult because of the impact of the inversion algorithm of falling aquatic products, terrain complexity, and density of ground verification stations. The results of this study can provide a reference for the selection of the best precipitation products for regional meteorological drought monitoring and for the improvement of remote-sensing precipitation products in the inversion algorithm of extreme drought environments.

The phenology of lake ice is a sensitive indicator of regional climate change. Through comprehensive analysis of long-term surface reflectance data, meteorological data, and lake information for seven large lakes (Karakul Lake, Balkhash Lake, Aral Sea, Alakol Lake, Zaysan Lake, Chatir Kol Lake, and Markakol Lake) with an area greater than 100 km² in Central Asia from 2000 to 2020, GIS-related technologies were used to explore the characteristics of lake ice phenology and its influencing factors. The results are as follows: (1) Lakes in Central Asia began to freeze from mid-September to early November and completely froze from late November to late December, with an average freezing time of 35 days; lake ice began to melt from late March to mid-May and would completely melt from early April to early June, with an average melting time of 18 days. (2) From 2000 to 2020, the start dates of ice formation in five of the seven lakes in Central Asia exhibited a delayed trend, with an average delay rate of 4.86 days per decade, whereas the start date of Balkhash Lake exhibited an advancing trend, with an advancing rate of 1.44 days per decade. The analysis suggests that this may be due to a decrease in the annual average temperature in the winter half of the year. The complete melting dates showed an advancing trend, with an average advancement rate of 2.90 days per decade. The average ice-covered period for the seven lakes was 171 days, with four of the lakes exhibiting a trend of shortening of the ice-covered period. The complete freezing period shows an overall trend of shortening, with Balkhash Lake exhibiting the most significant reduction, with a rate of 9.02 days per decade. (3) The spatial pattern of the formation and melting of lake ice in the seven lakes in Central Asia can be mainly divided into two categories: the lake water gradually freezes from both sides to the center and melts from the lake shore to the opposite side, or the lake water freezes from the shore to the opposite side and the earlier freezing, lake area melts the sooner. (4) The lake ice phenology changes in Central Asia are influenced by multiple factors such as lake characteristics (altitude and area) and climate (temperature and precipitation). Temperature is the key factor affecting lake ice phenology, and the higher the temperature, the shorter the ice-covered period. The area primarily affects the freezing date of the lake, and the larger the area, the shorter the ice-covered period. As the altitude increases, the ice-covered period of the lake extends.

The bedrock of the Danxia landscape is predominantly cemented by calcareous and ferruginous cements; thus, it has special landscape elements and formation mechanisms. It has become an important tourism resource with much attention because of its high scientific and ornamental values. However, compared with southeast humid areas, little attention has been paid to the characteristics and causes of the Danxia landscape in the arid climate of Northwest China. The Wensu Grand Canyon in the Aksu Prefecture of Xinjiang was selected and investigated to explore the characteristics and causes of the Danxia landscape by field investigation, sample microscope observations, salt chemistry and element geochemistry experiments, and ArcGIS hypsometry. The results are as follows: (1) The study area is characterized by canyon, peak and peak forest landscapes, well-developed mud flow films along slopes, and cap rock columns. The bedrock comprises red fluvial conglomerates and sandstones of Neogene age. The hard conglomerate beds are favorable for the formation of cap rock columns. (2) The study area is located near the Wensu salt dome with a high salt content, and the main salt minerals are probably chlorides, nitrates and sulfates, as inferred from the salt experiment. Most of the major elements of sandy debris samples within caverns migrated compared with the surface rock, indicating active chemical weathering in the arid climate. (3) The hypsometric integral (HI) shows that the HI value of the southern part of the study area is 0.61, which suggests an early stage. The HI value of the northern part is 0.38, which indicates a late stage. Therefore, the geomorphic evolution and development stages differ in the study area. Although the precipitation in the study area is very low, river erosion is obvious. Tectonism is also considered the main controlling factor of Danxia landscape evolution.

To explore the environmental significance of grain size characteristics, modern dunes in the Dinggye area of Xizang, China were sampled to analyze the grain size characteristics of surface sediments from different types of dunes. The results are as follows: (1) The grain size composition was similar between mobile and climbing dunes, and the grain size composition differed significantly between nebkha and the two dune types. The surface sediments of mobile and climbing dunes were dominated by fine and medium sands, whereas those of nebkha were dominated by very fine sand, fine sand, and powdery sands. (2) The grain size parameters of these dunes showed different trends, with the surface sediments of mobile and climbing dunes well sorted, nearly symmetrical, and moderately sharp, and those of nebkha moderately sorted, positively skewed, and very sharp. (3) The frequency distribution curves of mobile and climbing dunes are nearly symmetrical and single-peaked, whereas those of nebkha are bimodal; the differences in the probability accumulation curves of the surface sediments of these dunes are more obvious. (4) The deposition environment of the surface sediments of these dunes is mainly dominated by wind-formed deposition, with a few river deposits. (5) A comparison of particle size characteristics of the surface sediments of these dunes shows that the differences in particle size characteristics are mainly because of the joint action of sand source, vegetation cover, and wind conditions.

Terrestrial evapotranspiration (ET) is an important process of land-atmosphere exchange and an important link in the global water migration and energy transfer system. Identifying the temporal and spatial dynamic characteristics of ET is of great significance in the study of regional water cycle and energy conversion. In this study, we compared the appropriateness of the three different ET products (GLEAM, MOD16, and PML-V2) in China using the flux data of nine tower stations in the country. We used the dataset of ET products between 2003 and 2020 to analyze the temporal and spatial dynamic characteristics of evapotranspiration in China. The results showed the following: (1) The ET products of PML-V2 is the most suitable for China. (2) On the time scale, ET gradually increased during the research period. On the spatial scale, ET exhibited an increasing tendency during 2003—2020 from northwest to southeast of China. (3) According to the Hurst index, the future ET of the whole nation is expected to show the opposite trend compared to the past. This means that ET increased in the past but is expected to decrease in the future. This study analyzed the spatiotemporal dynamic characteristics of ET in China, which could provide a reference for regional water resources utilization and optimal allocation.

The meteorological data of the daily maximum and minimum temperatures of Hami City National Meteorological Reference Station, Xinjiang, China from 1961 to 2019 were selected to predict future temperature changes in the city. The spatiotemporal variation characteristics of extreme temperature in Hami City were analyzed using the univariate linear regression method, 10 a moving average, Mann-Kendall method, moving t test, and principal component analysis method, which provided a certain basis for determining the future temperature trend, enhancing the city’s ability to respond to extreme temperature events, and reducing the harm caused by meteorological disasters to agricultural production in Hami City. The results showed the following: (1) The change in the extreme temperature index in Hami City is asymmetric, the change in cold and warm indexes is opposite, and the warming amplitude of the night index is greater than that of the day index. Seasonally, most extreme temperature indices vary more widely in summer and autumn. (2) In the past 60 years, the frequency of extreme low-temperature events in Hami City decreased significantly, among which frost days decreased most significantly, with a decrease of -4.59 d·(10a)^-1. The frequency and intensity of extreme heat events have increased significantly. The growth season length tends to be consistent with the change of the thermal persistence index, while the change of the cold persistence index is the opposite. (3) The extreme temperature and cold indices are more sensitive to climate change. The abrupt changes occurred in the mid-1980s and the late 1990s, and the abrupt changes in the warm and persistent indices occurred in the late 1990s. (4) The cumulative variance contribution rate of the two principal components was 76.453%, the correlation between extreme temperature events and temperature warming was high, and the cold and warm index was negatively correlated. Cold nights, cold days, frost days, warm nights, and warm days were the main factors for the temperature increase in Hami City.

Geohazards such as landslides, rock fall, debris flow, and ground collapse occur frequently in the Ili Valley, Xinjiang, China, and there are various influencing factors for the formation of these geohazards. Based on the database of geohazards in the Ili Valley, this paper uses statistical analysis methods such as the frequency ratio and receiver operating characteristic curve to study the distribution characteristics and main controlling factors of geohazards in the Ili Valley. The results show the following: (1) The overall development degree of geohazards in the Ili Valley is greater in the east than in the west, and they are distributed in the middle and low mountains with 500-3500-m elevation. In terms of time of occurrence, they are mainly clustered in the spring and summer snowmelt and rainfall seasons. (2) The development of geohazards is affected by stratigraphic lithology, geological structure, topography and geomorphology, precipitation, and other factors. However, the main controlling factors for different types of geohazards are different. Landslides are most obviously affected by stratigraphic lithology, elevation, slope, and annual precipitation, while rock falls are mainly controlled by slope, elevation, and distance from faults. Ground collapses are obviously affected by elevation, distance from fault, and precipitation, and are basically induced by underground mining activities. The distance and elevation from faults are the main controlling factors for the development of debris flows.

Taking the five kinds of underlying surfaces i.e. mobile sand, sand sealing grass belt, wind and sand barrier forest, farmland shelter forest and farmland of Hobq Desert as the research objects, the particle size and elemental characteristics of sediments in the surface and 0-100 cm aeolian sand flow were analyzed by field observation and indoor analysis. The results showed that: (1) The surface roughness of farmland and vegetation-covered land increased significantly compared with moving sand, the wind speed at 10 cm decreased by more than 18%, and the total sand transport decreased by 85.6% on average. (2) The content of Cu, Fe, Mn and Zn in different grain sizes of sand is obviously different, Cu and Zn elements have the highest content in silt sand, Mn elements have the highest content in very fine sand, and Fe elements have the highest content in fine sand. (3) Under the redistribution of aeolian sand flow materials, the content of silt and very fine sand in aeolian sand flow increases with the uplift of height, which increases by about 14 times on average compared with the surface, and the content of Cu, Zn and Mn elements increases accordingly, the content of fine sand in aeolian sand flow increases first and then decreases, and the Fe element content also shows a trend of first increasing and then decreasing, which is closely related to the element content in the original surface and the element content characteristics of different particle sizes in surface sediments.

Climate change during the Holocene epoch in the arid Central Asia (ACA) has been one of the research hotspots in paleoclimate and global change research communities. Compared with the Asian monsoon region, the history of the change in humidity and the combination of moisture-temperature in the ACA region in the Holocene epoch are still controversial. In this study, we present the results of sediment element geochemical records taken from the Wenquan wetland of western Tianshan Mountains in Xinjiang, China. The core was collected in September 2017 using a Russian peat corer. The dated AMS ¹⁴C ages were calibrated to calendar years before present using the IntCall3 calibration dataset. Concentrations of Rb, Ti, Sr, Zr, Mg, and Ca were determined using an inductively coupled plasma atomic emission spectrometer, and the error of parallel analysis was <±5%. Holocene environmental evolution was reconstructed using the Rb/Sr, Sr/Ca, Ti/Sr, Mg/Ca, and Zr/Sr chemical element ratios. Through comprehensive analysis of the climatic and environmental indicators of the five ratios, and contrastive analysis with adjacent areas, this paper provides a useful information for a better understanding of the Holocene moisture-temperature relationship, and to identify patterns that drove the Holocene climate change in the Wenquan wetland. The results show that the Wenquan wetland is able to provide a reliable record of the Holocene climate change in Xinjiang. Based on analyses of the chronology records, the five element ratios reveal that the climate of the Wenquan wetland region during the last 10300 cal a BP has experienced a warm dry period (10300-7700 cal a BP), a warm dry period to warm wet period (7700-7000 cal a BP), a warm wet period (7000-4200 cal a BP), a warm dry period (4200-2900 cal a BP), and a cold and wet period (2900-81 cal a BP). This process is consistent with the climate change records and model simulations of neighboring regions, which verifies the warm/dry and cold/wet climate change patterns in Xinjiang during the Holocene. This indicates that the climate environment change in Xinjiang is similar to the westerly domination pattern and exhibits an “out-of-phase” relationship with the pattern of monsoonal evolution in eastern monsoonal Asia. Moreover, the cooling process indicated by the ratio of the core elements in the Wenquan wetland sediments during 7700-7000 cal a BP may correspond to the global cooling event of 8.2 cal ka BP. The warming process observed during 4200-2900 cal a BP may be consistent with a Holocene sub-high-temperature event reflected by Dunde ice cores from 3.0 to 2.9 cal ka BP. In recent years, an increasing number of researches has concentrated on ACA region moisture changes and the possible mechanisms responsible for these changes, based on a westerly dominated regime. Compared with other proxy indexes in the surrounding region, it was found that the trend of gradual wetting in Xinjiang may be the result of the combined effect of decreasing temperature and increasing precipitation in the Holocene.