Search

Achieving land degradation neutrality (LDN) is considered critical to achieving sustainable development goals. However, further research and an improved large-scale desertification assessment framework are needed. We constructed a comprehensive land desertification status index (LDSI) to assess desertification and its driving mechanisms in Central Asia from 2001 to 2020. Our spatial distance models and principal component analysis were verified by combined field data. The results indicate that desertification levels in Central Asia fluctuated, decreasing from south to north and increasing from west to east. The extent of increased desertification (45.75%) and recovery (45.65%) were nearly equivalent. According to optimal parameters-based geographical detectors, climate factors are the main contributors to desertification, the most significant being reference evapotranspiration. We also found regional differences in the driving mechanisms of desertification. Based on Theil–Sen, Mann–Kendall, and Hurst tests, future desertification intensification of 9.69% is possible, with hotspots mainly in northwestern Central Asia, while 9.73% restoration may occur in other areas. The findings have relevance for achieving LDN. Policymakers should monitor the impact of global climate change and develop recovery measures based on comprehensive consideration of local climate, soil, and terrain conditions. [ABSTRACT FROM AUTHOR]

Abstract In arid and semi-arid climatic areas, lakes are extremely essential for fragile ecological environment and regional sustainable development. Ebinur Lake is an important component of the ecological barrier of Junggar Basin, Xinjiang Uyghur Autonomous Region (XUAR), China. Due to the tremendous changes in Ebinur Lake and surrounding marshes during the last decades, Ebinur Lake becomes a representative ecological degradation region in northwestern China. The detection of the intra-annual variations of water body and its responses to changing environment are critical for regional ecological security and rehabilitation of degraded ecosystem. To extract more accurate water information using Synthetic Aperture Radar (SAR) data and further fill the gap of inter-month dynamic monitoring of Ebinur Lake, a new SAR water index (modified Sentinel 1A water index, MSWI) was proposed based on the relationship between normalized difference water index (NDWI) imageries and Sentinel 1A data. The dynamic thresholds of classification were selected using Otsu method and the results showed that the classification results were acceptable with the optimal overall accuracy of 99.94% and kappa coefficient of 0.9971, respectively. We conduct a time series analysis of surface areas of Ebinur Lake using S1A data from February 9th, 2017 to February 4th, 2018. The maximum lake surface area was 965.29 km2 in April 22nd, 2017, while the minimum value was 750.37 km2 in September 1st, 2017, and the mean area was 831.51 km2. The seasonal variations showed the stages of "sharp rising" – "significant decreasing" – "gradual stabilizing" in the study period. The water surface area was highly correlated with inflow water volume (correlation coefficient = 0.72, P < 0.001). The variation of Ebinur Lake's water surface area is crucial to monitor the resulting eco-environmental impacts under the changing environmental conditions in the arid and semi-arid areas. Graphical abstract Unlabelled Image Highlights • Proposed a new SAR water index (MSWI). • The 12-day interval spatial-temporal distributions of Ebinur Lake. • Variation of lake in arid areas and its eco-environmental impacts. • Examined the driving forces of the dynamic variations. [ABSTRACT FROM AUTHOR]

Cities in arid and semi-arid regions often confront harsher natural environments and severer industrial pollution. So these are frequently subjected to heavy air pollution, which has critical consequences for the health and production activities and well-being of local citizens. To curb the aggravation of air pollution, the Chinese government conducted an air pollution control action plan with unprecedented strength in September 2013. In this study, based on the fine-resolution aerosol optical depth (AOD) data, combined meteorological and other auxiliary data, utilizing information entropy, backward trajectory, source analysis, and multi-layer perceptron methods, we comprehensively analyzed the spatiotemporal characteristics and variations of AOD in typical arid and semi-arid areas megacities (Lanzhou and Urumqi) from 2010 to 2019, and further explore the potential pollution sources and nature and anthropic influences on AOD. The main results indicate that Lanzhou AOD remains higher than Urumqi in almost all different time windows, especially in the spring. Specifically, the AOD shows a single-peaked distributed trend on the monthly scale, but the peak occurred one month later in Lanzhou than in Urumqi. In the seasons, Lanzhou and Urumqi all show a trend that spring and winter AOD are higher than summer and fall. Lanzhou and Urumqi overall show a tortuous decreasing and then rebounded trend between 2010 and 2019, but the AOD variation in the Urumqi is more dramatic than in Lanzhou. The high AOD mainly concentrates in low altitude areas and there is a downward tendency of AOD in high-AOD areas, but an upward trend in low-AOD areas. The AOD intensity of change is higher in Lanzhou than in Urumqi, but Lanzhou has more concentration. The seasonal differences in backward trajectories are stronger in Lanzhou than in Urumqi, generally, Lanzhou's potential trajectory is mainly distributed in the Hexi Corridor, while Urumqi is along the northern slope of the Tianshan Mountains urban cluster distribution. The potential source area of AOD has a significant seasonal variation, and the closer to the object, the greater the contribution. Generally, Lanzhou pollution is more sensitive to the atmosphere condition and has a wider source area than Urumqi. The AOD was influenced by a combination of natural and anthropogenic factors, but natural variables seem to be more dominant. The impact of natural factors in Urumqi and Lanzhou is statistically obtained up to 75.04% and 52.01% respectively. [ABSTRACT FROM AUTHOR]

• The potential of Google Earth Engine (GEE) to detect long-term wetland change series is explored. • Comparing the response of different types of lake wetlands in the arid area to climate. • Evapotranspiration is the primary factor influencing wetland shrinkage in drylands. Climate change has caused inland wetlands shrinkage and exacerbated problems, such as sustainable development and ecological security, for years. These issues are mainly pronounced in the inland arid area. The ecological environment's deterioration is especially severe in the drylands of the interior. However, dryland wetland changes and their response to climate are poorly understood. This study uses the K-means algorithm in Google Earth Engine (GEE) to classify two typical dryland wetlands (Ebinur and Bosten Lakes) for rapidly and accurately detecting dryland wetland changes. Moreover, it explores the long-term spatial–temporal variation in wetland distribution. In addition, it investigates the response of various lakes to climate change in northern and southern Xinjiang using wavelet analysis. The study's results showed that K-means clustering in the GEE platform has a high classification accuracy (Kappa > 0.8) in wetland classification, making it a feasible approach. The terminal lake wetland types, represented by the Ebinur Lake, changed significantly between 2001 and 2021. In contrast, the inflow-outflow lake wetland types, represented by the Bosten Lake, perform more consistently. Significant spatial–temporal variation is observed at Ebinur Lake, with the lake gradually shrinking and transforming into a marsh, where the largest marsh proportion degrades into non-wetland during the year. Bosten Lake experienced frequent conversions between marsh and non-wetland throughout the year. Furthermore, the responses of various dryland lakes to climate change are consistent, and a low precedes precipitation and follows evapotranspiration. However, their sensitivity to climate response varies, with the terminal lake being most affected by climate change. Mastering the dynamic changes and climate response of dryland wetlands achieves the sustainable development goals of drylands, including carbon neutrality and peak carbon dioxide emissions. [ABSTRACT FROM AUTHOR]

The effects of 35 years of manure amendment on fungal communities were evaluated in Mollisols of northeast China. Soil samples from different fertilization regimes were collected, and quantitative PCR analysis of fungal community size and Illumina platform based analysis of the ITS gene were performed to characterize soil fungal abundance and compare community structure and diversity. The treatments were no fertilizer (CK); manure (M); nitrogen, phosphorus and potassium inorganic fertilizer (NPK); and inorganic fertilizer plus manure (MNPK) regimes. Soil fungal diversity was decreased by inorganic fertilizer. Inorganic fertilizer plus manure induced a weak increase in fungal diversity and a slight decrease in size. The predominant phyla were Ascomycota (63.77–78.70%), Zygomycota (8.33–14.80%) and Basidiomycota (4.03–13.47%). At each taxonomic level, the percentages dramatically differed, especially between MNPK and NPK. For example, the numbers of Fusarium and Gibberella with potential pathogenicity were all higher in NPK than in MNPK; the beneficial genus Podospora was the highest in MNPK and the least in NPK. Principal coordinates analysis showed that CK and M were clustered together; the incorporation of NPK with manure improved the fungal structure near to that of CK and separate from that of NPK. Redundancy analysis indicated that fungal community structure was most affected by soil available phosphorus (AP) and organic matter contents, followed by soil pH. Simpson and Shannon indices were closely correlated with soil pH, AP, total phosphorus and total nitrogen. The results indicated that manure application altered soil properties and soil fungal community structure, and manure with inorganic fertilizer counteracted some of the adverse effects of the inorganic fertilizer. [ABSTRACT FROM AUTHOR]

Aerosols significantly contribute to global and regional climate change by altering the surface solar radiation, thereby affecting plant productivity. Central Asia is a primary source of global dust aerosols. However, the mechanisms of how aerosols affect terrestrial gross primary productivity (GPP), especially in Central Asia, are not clearly understood. In this study, we investigated the spatial variation in aerosol optical depth (AOD) and GPP and the relationship between them during the growing season (April–October) from 2001 to 2018 using remote sensing data from several sources. We created a GWR-SEM model consisting of a geographically weighted model (GWR) coupled with a structural equation model (SEM) to quantify and analyze the effects of AOD on GPP. The results show that AOD decreased slightly at a rate of −0.0002 y−1 during the study period and that there was a tendency towards spatial aggregation. The extent of AOD pollution in the northwest region (around the Aral Sea) was slightly greater than that in the southeast. GPP increased significantly at a rate of 7.2965 g C m−2 y−2, especially in the northern region. There were some differences in the effects of AOD on GPP between different vegetation types; the highest AOD–GPP correlation was found in shrublands and croplands. Analysis of the GWR-SEM model suggested that AOD and two forms of radiation (surface net radiation, SNR, and photosynthetically active radiation, PAR) explained 72.4% (63.4% for 2001, 66.8% for 2018) of the spatial variation in GPP. SNR had the greatest effect on GPP, followed by AOD. Diffuse PAR had the greatest indirect effect on GPP. The findings of this study highlight the importance of aerosol pollution on spatial variation in gross primary productivity, and they provide a methodological framework for investigating the relationship between AOD and GPP in arid areas. [Display omitted] • Spatial and temporal variations of AOD and GPP over Central Asia are evaluated against MODIS products. • The correlation between AOD and GPP over Central Asia are revealed. • A GWR-SEM model is used to quantitatively evaluate aerosol effects on GPP. • The major drivers of GPP change with time and vegetation types. [ABSTRACT FROM AUTHOR]

Black soil is common in northeast China and plays an important role in Chinese crop production. However, in the past three decades, inappropriate use of fertilizer has caused a sequence of agroecological issues. The objective of this research was to evaluate the effect of long-term fertilizer on the microbial communities in black soil. The soil was subjected to four fertilization regimes: without fertilizer (CK); manure (M); nitrogen, phosphorus and potassium inorganic fertilizer (NPK); and inorganic fertilizers with manure (MNPK). The soil pH was decreased by inorganic fertilizers and increased by manure. Quantitative PCR analysis of microbial community size and Illumina platform-based analysis of the V4 16S rRNA gene region were performed to characterize soil microbial abundance and to compare community structure and diversity. Microbial community size was enhanced by the incorporation of inorganic fertilizer and manure. Microbial diversity was decreased by inorganic fertilizer and increased by the incorporation of inorganic fertilizer and manure. The predominate phyla in all samples were Proteobacteria (29.39–33.48%), Acidobacteria (13.14–16.25%) and Actinobacteria (9.32–10.77%). The relative abundance of different classes significantly differed among the different treatments, especially MNPK and NPK. Acidobacteria and Deltaproteobacteria were relatively stable in organic fertilizer treated soil. Gammaproteobacteria, Alphaproteobacteria and Betaproteobacteria were sensitive to all the fertilization regimes. Comparatively, Spartobacteria was stable in response to fertilization practices. Principal coordinate analysis indicated that microbial communities were primarily clustered into three groups: CK and M were clustered together; MNPK was improved by manure and separated from NPK. Shannon and Simpson indexes were significantly correlated with soil pH and the concentrations of available phosphorus and total phosphorus. Redundancy analysis indicates that microbial communities were closely positively correlated with soil nitrate nitrogen concentration ( P = 0.002) and pH ( P = 0.002). These results indicate that inorganic fertilizer plus manure increased microbial size and diversity and changed microbial composition. [ABSTRACT FROM AUTHOR]

Fine particulate matter (PM 2.5) is a major source of air pollution in China. Although there have been many studies of the drivers of PM 2.5 pollution in the megacities clustered in eastern China, the behavior of PM 2.5 in the northwestern urban agglomeration is not well understood. This study used near-surface observation data for 2015–2019 obtained from the national air environmental monitoring network to examine variation in PM 2.5 in the urban agglomeration on the northern slopes of the Tianshan Mountains (UANSTM). Two-factor interaction provided new insights into the dominant factors of PM 2.5 in the study region. The annual average PM 2.5 concentrations over the study period was 54.3 μg/m3, with an exceedance rate of 23.3%. Wavelet analysis showed two dominant cycles of 320–370 d and 150–200 d with high pollution events occurring in winter. The generalized additive model (GAM) contained linear functions of pressure, non-linear functions of SO 2 , NO 2 , relative humidity, sunshine duration and temperature. The two most primary variables, NO 2 and SO 2 , represent 20.65% and 19.54% of the total deviance explained, respectively, while the meteorological factors account for 36.1% of the total deviance explained. In addition, the interaction between NO 2 and other factors had the strongest effect on PM 2.5. The deviance explained in the two factor interaction model (88.5%) was higher than that in the single factor model (78.4%). Our study emphasized that interaction between meteorological factors and pollutant emissions enhanced the impact on PM 2.5 compared with individual factors, which can provide a scientific basis for developing effective emission reduction strategies in UANSTM. [Display omitted] • Multi-scale characteristics of PM 2.5 in the UANSTM are studied by wavelet analysis. • We identify the influencing factors of PM 2.5 using the generalized additive model. • The change in PM 2.5 is influenced by multiple factors and their interactions. [ABSTRACT FROM AUTHOR]

• GMM can distinguish similar spectral features in vegetation. • After spectral feature partition, the difference modeling is superior to whole-sample modeling. • UAV hyperspectral can indirectly reveal salinization patterns by inversion of saline vegetation SPAD. The chlorophyll content of saline vegetation can indirectly reflect salinization. Rapid and non-destructive capture of chlorophyll content of saline vegetation at a regional scale is essential for saline soil improvement and sustainability of saline agriculture. However, traditional soil–plant analyzer development (SPAD) monitoring based on SPAD-502 is carried out at the leaf scale, which does not allow rapid access to SPAD information for the whole region. In this study, we proposed that post-hyperspectral classification based on spectral differences that could contribute to enhanced estimation of SPAD in saline vegetation. To test this proposal, we partitioned the hyperspectral images using a Gaussian mixture model. Then, estimation models based on spectral partition and full-sample SPAD were developed using in situ observation data and a random forest model, respectively. Finally, the unmanned aerial vehicles (UAV) hyperspectral images were used as the input data source to digitally map the SPAD of saline vegetation in the region, using the prediction model. The results indicated that there were significant intensity and shape differences in the spectral reflectance characteristics of saline vegetation under different clusters. The SPAD prediction model, based on spectral feature partition, performed significantly better than the full sample. The SPAD maps of saline vegetation before and after clustering displayed similar spatial distribution models, but the prediction uncertainty of the models, based on spectral feature partition, was relatively low. Our results confirm the effectiveness and stability of UAV hyperspectral and spectral partition-based modeling in developing SPAD spatial distribution estimation models for saline vegetation. [ABSTRACT FROM AUTHOR]

• Human activities are emerging as the primary driver of changes in land cover. • Water consumption in agriculture and ecology has opposite trends in the future. • Cultivated land expansion crowding out natural vegetation areas in the future. • Coupling GEE and PLUS to monitor and predict LULC changes. Water scarcity has emerged as a major impediment to the long-term development of inland arid region basins. Increased human activity has produced land degradation in arid areas and water stress in recent years. The objective of this research was to expose patterns in land cover changes in inland arid basin and the mechanisms driving them, as well as to optimize the socioeconomic and ecological water consumption structure of water resources. Therefore, this research used Google Earth Engine to map land cover year by year from 2000 to 2020, simulated future land cover changes and driving force analysis using a Patch-generating land use simulation model, and finally established the relationship between water consumption of different systems and different land cover types using linear equations, and scenario simulation method were used to estimates the ecological and agricultural water consumption in 2035. The results showed (1) Over the previous 20 years, there has been an increase in the amount of saline, agricultural land, and natural vegetation (grassland, scrub, and forest), as well as a considerable drop around permanent glaciers and snow. (2) Under the Markov-2035 scenario, saline land decreases to 32.92 km2, natural vegetation decreases to 381.89 km2, arable land shrinks to 135.69 km2, and the permanent glacier shrinks to 30.03 km2. Ecological water consumption would be 0.127 × 108 m3, while agricultural water consumption would be 0.394 × 108 m3. (3) Elevation influenced the variety of all land cover types. Saline land was more sensitive to temperature, railroad proximity, and pH; agricultural land was more sensitive to population density and gross regional product; and natural vegetation was more sensitive to soil organic matter, railroad proximity, and temperature. [ABSTRACT FROM AUTHOR]

• BEMD was applied to high-resolution SOM map in northwestern China. • Scale- and region-specific variations in SOM were unraveled. • The variance of SOM was partitioned by climate, topography and land surface. • Main factors that influencing SOM were different in different scales. Soil organic matter (SOM) significantly affects soil quality, food security, and the global carbon cycle. Variations in the SOM content are controlled by many environmental factors and differ depending on the scale and region. The Mountain–Oasis–Desert System (MODS) in arid areas is limited by climate conditions and water and soil resources. Therefore, the effects of environmental factors on the SOM content may differ. This study aimed to characterize spatial patterns of the SOM content in a MODS in northwestern China and reveal the SOM variations at different scales and regions. First, we employed the random forest model to predict the SOM content and the uncertainty at the nodes of a 90 m grid. We then extracted scale- and region-dependent variations in the SOM content using bivariate empirical mode decomposition (BEMD). Data analysis, geostatistics, and a two-dimensional fast Fourier transform were applied to analyze the SOM pattern and its correlations with environmental covariates at different scales. The results suggest that the variations in the SOM content were mainly concentrated at the large scale. The correlations between environmental factors and the SOM were scale- and region-specific, indicating complex interrelationships among topography, land use, salinization, soil erosion, and fertilization. At the intermediate-large spatial scale, topography characteristics and surface features were the major factors controlling the SOM content, whereas the climate had a larger effect on SOM than other factors at the large scale. Based on our results, BEMD has great potential to reveal SOM variations in the scale–region domain. BEMD can be used in environmental modeling in the future to gain insights into the variations in soil properties depending on environmental factors and anthropogenic activities. [ABSTRACT FROM AUTHOR]

• Successful application of the ESTARFM to study the dynamics of playas surface. • NDVI and SI are sensitive to precipitation events. • Precipitation events affect the spatiotemporal distribution of surface salinity. • Salinity spectral indices decrease after precipitation events. Playas are desert landscapes unique to arid regions that respond quickly to climate change. Changes in surface salinity in Praia are directly linked to water security and regional ecological and economic health. This study aimed to explore the sensitivity of remotely sensed spectral indices to precipitation events. An ESTARFM was utilized to fuse daily (less than 10% cloudiness) MODIS and Landsat 8 imagery from April to September 2019 in the Ebinur Lake Wetland Reserve. Salinity was tested in the laboratory after soil collections in May (spring) and August (summer). The significantly correlated normalized difference vegetation and salinity indices were compared to the precipitation data and found to fluctuate in response to precipitation. The normalized difference vegetation and salinity indices decreased following precipitation events and increased during precipitation intervals, and the precipitation amount and evaporation intensities may have influenced the magnitudes of the decrease and increase. Following precipitation, the playa surface produces new puddles and a new spatial distribution pattern of soil salinity. This study determined that the spatiotemporal fusion technique is an effective method for observing the dynamics of playa surfaces, and precipitation and evaporation affect the spatial distribution of salt on the playa surface; however, the monitoring period should be short when utilizing remote sensing to monitor playa salinity. [ABSTRACT FROM AUTHOR]

Changes in atmospheric particulate matter content directly affect the Earth-atmosphere system balance and climate change and indirectly affect the environment and ecology. Therefore, this matter is receiving increasing attention. A typical arid area with a fragile ecological environment and frequent dust activities, Xinjiang, China, was investigated using Moderate Resolution Imaging Spectroradiometer (MODIS) daily aerosol optical thickness and Ångström wavelength exponent data to obtain the temporal and spatial distribution characteristics of long-term dust aerosols over 20 years. The frequency of occurrence was calculated to accurately identify anthropogenic and natural sources, and data, such as wind speed, were superimposed to quantitatively estimate dust aerosol emissions. The factors that influenced dust aerosol emissions are discussed. The results show that the distribution of dust aerosols in South and North Xinjiang is obviously different, and the average dust aerosol optical depth (DOD) in spring is higher than that in other seasons. The sources of natural dust have little variation and are primarily located in Lop Nur, on the edges of the Taklamakan and Gurbantunggut deserts, in the central Tuha Basin, and in dried-up lakes. The sources of anthropogenic dust are mainly distributed in the farmland areas of the oasis plain and the impact fan plain. Natural dust aerosols are dominant in Xinjiang, with high natural dust emissions in spring. Changes in anthropogenic dust aerosols are caused by production activities, and the annual average anthropogenic dust emissions account for 8% of the total dust emissions. Precipitation can promote a reduction in natural dust emissions. Topography has an obvious effect on the formation and transportation of dust aerosols, which manifests as the frequent occurrence of dust in the basin. The population density data and Defense Meteorological Satellite Program/Operational Linescan System (DMSP/OLS) nighttime light data reveal, that the intensity of human activities has a great impact on anthropogenic dust aerosols. This study provides support for research on how aerosols affect climate and environmental changes in Xinjiang. · Daily dust aerosol optical depth (DOD) data was obtained from 2003 to 2019 · Anthropogenic and natural dust sources of Xinjiang, China, were identified · Dust aerosols emissions in dust sources were quantitatively estimated · The influences of precipitation, topography, and human on dust aerosols were analyzed [ABSTRACT FROM AUTHOR]

Aerosols are an important contributor to global atmospheric environmental changes and have critical effects on the global climate system and human health. Central Asia is one of the most important sources of dust aerosols in the world and produces a significant portion of global aerosols. Central Asia is also a scarce aerosol-data area, so it is of great significance to obtain and verify new aerosol data from this region. In this study, based on the aerosol optical depth (AOD) data from remote sensing (MYD04_L2) and ground-based observations (AERONET and Microtops II), the applicability of multiangle implementation of atmospheric correction (MAIAC) AOD in Central Asia was comprehensively analyzed, and the variation in AOD in Central Asia over the last 20 years was analyzed by the information entropy method. The results indicate that MAIAC AOD has good application prospects in Central Asia and can effectively compensate for the lack of observational data from Central Asia. MAIAC AOD exhibits excellent spatiotemporal consistency with MYD04 deep blue (DB) AOD and has a better ability than MYD04 DB AOD to describe local fine-scale features. Furthermore, MAIAC AOD demonstrates high consistency with ground-based AOD observations, showing high R (0.737) and low RMSE (0.067) values and having 65.2% of samples falling within the expected error (EE) envelope. When employing the ground-based AOD observations as a bridge, MAIAC exhibits superiority to MYD04 DB in both the richness number of valid high-quality retrievals and the retrieval accuracy of various evaluation indicators. The annual variation in AOD in Central Asia exhibits a unimodal distribution, with AOD being largest in April, followed by March and May, and comparable rangeability. Based on information entropy, interannual variation in AOD exists in most areas of Central Asia, with AOD in the Taklimakan Desert area being significantly increased and that in northern Central Asia (Kazakhstan) showing a downward trend. • MAIAC and MYD04 DB AODs exhibit good spatiotemporal consistency over Central Asia. • A good agreement between the AOD values from MAIAC AOD and AERONET was found. • Spatial and temporal variations are evaluated against MAIAC products. • The AOD annual variation in Central Asia is characterized by a unimodal distribution. [ABSTRACT FROM AUTHOR]

• Vis-NIR spectra were used to estimate the SOM in salt-affected soils. • Seven spectral resolutions (SRs) were tested from 1 nm to 100 nm. • The optimal band combination algorithm is useful for extracting spectral variables. • The soil salinity had a strong negative influence on the SOM model performance. • Vis-NIR spectra with an SR of 20 nm was recommended to estimate the SOM. Visible and near-infrared (Vis-NIR) spectroscopy is a cost-effective technique for alternative soil physical and chemical analyses for estimating soil properties. The optimal band combination algorithm is an effective method of extracting spectral variables by considering the interaction information between wavebands, but for laboratory Vis-NIR spectral data, this method is susceptible to the "dimensional curse". Here, we hypothesized that properly degrading the spectral configuration (i.e., decreasing the number of spectral bands and coarsening the spectral resolution) can improve the computational efficiency without affecting the prediction accuracy. To test this hypothesis, we constructed six degraded spectral configurations from an initial spectral database (i.e., consisting of 2001 spectral bands acquired with a portable ASD spectroradiometer) with a reduction in the number of spectral bands from 2001 to 19, a coarsened spectral resolution from 3 to 100 nm, and a spectral sampling interval equal to the spectral resolution (i.e., uniform interval sampling). In this study, the databases consisted of 255 soil samples collected from the Ebinur Lake area in Northwest China. The relationship between the soil organic matter (SOM) and the spectra was established using a partial least-squares-support vector machine (PLS-SVM) through two strategies: one is in accordance with the different salinity levels, and the other involves applying the optimal band combination algorithm from each spectral configuration. The results indicated that the soil salinity had a strong negative influence on the performance of the SOM models (R 2 cv , 0.46–0.81). However, the optimal band combination algorithm can improve the sensitivity (R 2 pre , 0.36–0.65) of spectral information and the SOM. Overall, the prediction accuracy obtained through the optimal band combination algorithm was generally superior to that from full-spectrum data. The prediction performance of the optimal band combination algorithm was accurate (R 2 pre ≥ 0.85) and stable (RPIQ pre , ~3.20), with a spectral resolution between 3 and 20 nm (i.e., the number of spectral bands decreased from 2001 to 99). Considering the accuracy and time-consuming nature of this approach, the combination of a 20 nm spectral resolution and an optimal band combination algorithm was the most effective method. In summary, this research will guide future studies in transforming hyperspectral datasets into parsimonious representations and uses the optimal band combination algorithm efficiently to determine the informative variable. Furthermore, the optimal band combination algorithm has broad application prospects in soil Vis-NIR spectroscopy and other fields of spectroscopy. [ABSTRACT FROM AUTHOR]

• The dominant scales of soil salinity variation were 21.77 km and >32 km. • GDVI was the major predictor of soil salinity on salt-affected land. • Accuracy of the predictions with wavelet reconstruction was largely improved. • The predicted performance of WRM was outperform than SMLR. The variation in soil salinity is affected by environmental factors that occur at different scales with varying intensities. It is critical to adequately consider environmental variables under scale effects for digital soil mapping which has been minimally discussed in previous studies. The objectives of this research are to analyze the scale-dependent variability in soil salinity distribution under environmental controlling factors using discrete wavelet transform (DWT) techniques and to compare the differences between the accuracy of soil salinity predictions with and without multiple scale-specific relationships. Thirteen environmental factors related to soil salinity that included influencing environmental factors and indicative environmental factors involving climate, soil, terrain, and vegetation were extracted at 500 m intervals along four transects through farmland and salt-affected land situated at the oasis and oasis-desert ecotones of Xinjiang, China. Each spatial series of soil salinity and environmental variables along the four transects was separated into seven scale components (six details components, namely D1 through D6, and one approximation component, namely A6). A Hilbert transform was used to identify the specific spatial scales of each scale component in the DWT procedure. The results indicate that 21.77 km and >32 km were the dominant scales, which explained approximately 60–80% of the spatial variation of soil salinity throughout the oases. The prediction accuracy with wavelet reconstruction that depended on all the scale components of environmental variables is significantly improved compared with the accuracy of those with the stepwise multiple linear regression method at a single sampling scale. The generalized difference vegetation index (GDVI) was the major predictor of soil salinity on salt-affected land, while evapotranspiration and the terrain ruggedness index (TRI) were the major contributing factors in farmland inside the oases. This study demonstrated that specific scale-dependent relationships can reveal the scale control of soil salinity variation and had the potential to improve the prediction accuracy of soil properties. [ABSTRACT FROM AUTHOR]

• Bacterial community succession was detected in two different composting process. • Investigating bacterial communities was based on metatranscriptomics. • Succession of metabolism function of bacterial community was analyzed by PICRUSt. • Bacterial community succession was tightly related to AP in a bioreactor system. In this work, Illumina MiSeq sequencing of cDNA from metatranscriptomics RNA reverse transcription were employed in combination with phylogenetic investigation of communities by reconstruction of unobserved states (PICRUSt) to estimate the dynamic variations of bacterial community structures and metabolic functions in a bioreactor and traditional composting process. Results showed that the change of bacterial α-diversity in the first three stages exhibit opposite trends in the two composting systems. The four most abundant phyla were the same in both systems (Firmicutes, Proteobacteria, Bacteroidetes and Actinobacteria), but the most abundant genera were different. The five most abundant genus-level groups in the bioreactor were Psychrobacter, Galbibacter, Pseudomonas, Staphylococcus and Flavobacterium. Within the same phase, the functional bacteria were dramatically different in the two composting processes. In the bioreactor system both bacterial community structure and metabolism function were greatly affected by available phosphorus. [ABSTRACT FROM AUTHOR]

Aerosols are an important contributor to global atmospheric changes and have critical effects on the climate system. Regionally, aerosols in central Asia comprise a significant portion of global aerosols. Based on aerosol optical depth (AOD)Level 2 daily product data and land cover type product data, the long-term AOD characteristics of six major land use/cover types and their relationships with landscape metrics are discussed. Contribution analysis is applied to quantitatively estimate the effects of land use/cover on regional AOD over central Asia. The results show that series of daily AODs over six land uses/covers display remarkable annual cyclic variations and obvious seasonal changes. The annual average AODs for barren land and cropland are highest, followed by regional AODs. There are different frequencies and times of occurrence for high AOD values of various land types. Urban areas are one of the major contributors to the regional atmosphere in winter; grasslands have a great influence on regional AOD decreases. Barren land always has a high contribution to the regional AOD. The land use types affected by anthropogenic activities were smaller contributors to regional aerosols than barren lands affected by climate factors. This paper advances the understanding of relationship between aerosols and land use/cover and facilitates land use decision making. Unlabelled Image • Daily AOD data of six land types are obtained from 2007 to 2017 • Variation characteristics of six major land use/cover types AOD are revealed • Frequencies and times of occurrence for high AOD values vary in diverse land types • Further quantitative calculations show the LUCCs contributors to the regional AOD [ABSTRACT FROM AUTHOR]

Accurate assessment of soil salinization is considered as one of the most important steps in combating global climate change, especially in arid and semi-arid regions. Multi-spectral remote sensing (RS) data including Landsat series provides the potential for frequent surveys for soil salinization at various scales and resolutions. Additionally, the recently launched Sentinel-2 satellite constellation has temporal revisiting frequency of 5 days, which has been proven to be an ideal approach to assess soil salinity. Yet, studies on detailed comparison in soil salinity tracking between Landsat-8 OLI and Sentinel-2 MSI remain limited. For this purpose, we collected a total of 64 topsoil samples in an arid desert region, the Ebinur Lake Wetland National Nature Reserve (ELWNNR) to compare the monitoring accuracy between Landsat-8 OLI and Sentinel-2 MSI. In this study, the Cubist model was trained using RS-derived covariates (spectral bands, Tasseled Cap transformation-derived wetness (TCW), and satellite salinity indices) and laboratory measured electrical conductivity of 1:5 soil:water extract (EC). The results showed that the measured soil salinity had a significant correlation with surface soil moisture (Pearson's r = 0.75). The introduction of TCW generated satisfactory estimating performance. Compared with OLI dataset, the combination of MSI dataset and Cubist model yielded overall better model performance and accuracy measures (R 2 = 0.912, RMSE = 6.462 dS m−1, NRMSE = 9.226%, RPD = 3.400 and RPIQ = 6.824, respectively). The differences between Landsat-8 OLI and Sentinel-2 MSI were distinguishable. In conclusion, MSI image with finer spatial resolution performed better than OLI. Combining RS data sets and their derived TCW within a Cubist framework yielded accurate regional salinity map. The increased temporal revisiting frequency and spectral resolution of MSI data are expected to be positive enhancements to the acquisition of high-quality soil salinity information of desert soils. Unlabelled Image • Differences between Landsat-8 OLI and Sentinel-2 MSI are distinguishable. • Satellite derived surface soil moisture is significantly correlated with soil salinity. • Cubist is a satisfactory approach for soil salinity mapping (RPIQ = 6.824). • MSI image with finer spatial resolution performs better than OLI. • We need to pay more attention to the environmental covariates. [ABSTRACT FROM AUTHOR]

• FODs were used to process and analyze soil Vis-NIR spectra. • The correlation coefficient with SOMC was best at 1.05- to 1.45-order. • MNDI was used to emphasize the SOMC information. • FOD-MNDI showed the best estimation accuracy, with an RPIQ reaching 3.40. Visible-near-infrared (Vis-NIR) spectroscopy makes it possible to estimate soil organic matter content (SOMC). Spectral pretreatment techniques have important significance in the quantitative analysis of SOMC. A total of 150 soil samples collected in northwestern China were used as data sets for calibration and validation. The SOMC values and reflectance spectra were measured in the laboratory. Fractional-order derivatives (FODs) (intervals of 0.05, range of 0–2) were used for soil spectral pretreatment, and a new three-band index (modified normalized difference index, MNDI) was constructed based on the band-optimization algorithm and the existing two-band exponential form (normalized difference index, NDI). Partial least square-support vector machine (PLS-SVM) models were calibrated using spectral parameters selected based on a single dimension (FOD), two-dimensional index (NDI) and three-dimensional index (MNDI) and subsequently applied to estimate SOMC. Three model evaluation parameters, namely, the coefficient of determination (R2), root mean squared error (RMSE), and ratio of performance to interquartile range (RPIQ), were used to evaluate the estimation accuracy of the models. The results showed that with increased derivative order, the spectral strength gradually decreased, but the spectral detail increased. Furthermore, the correlation between FOD spectra and SOMC was significantly enhanced in the visible region, with the most obvious effect in the 1.05- to 1.45-order range. The PLS-SVM modeling results showed that the sensitivity and estimation accuracy of SOMC increased with increasing spectral synergy (i.e., 1D (FOD) < 2D (NDI) < 3D (MNDI)). Among the models, MNDI exhibited the best model performance, yielding a validation R2 and RPIQ of 0.846 and 3.396, respectively. The combination of FOD and MNDI could weaken the soil noise information and improve the prediction accuracy of SOMC. Furthermore, the three-dimensional index has strong application potential for estimating other biochemical parameters of soil using Vis-NIR spectroscopy. [ABSTRACT FROM AUTHOR]

• Desert-oasis thermal environmental change threats to eco-sustainability and habitat quality. • Accurate identify the "cold island zone" and "heat island zone" in arid zones. • LULC and urbanization development are closely linked to the thermal environment. • Spatial heterogeneity of natural and human factors in land use types. As a typical geographic landscape unit in the arid zone of northwest China, the distribution of the thermal environment in the substratum of the desert-oasis is of great significance to the monitoring of the ecological environment and the quality of human habitat in the region. The purpose of this research is to reveal the spatial–temporal evolution pattern of the thermal environment of the desert–oasis in the Turpan–Hami region from 2005 to 2020 and its trend changes, and to investigate the relationship between natural and human factors and the thermal environment and to conduct a long time series analysis. Firstly, the accuracy of MODIS land surface temperature (LST) data combined with site data was verified. Secondly, the mean-standard deviation method is used to identify desert heat island and oasis cold island zones. Again, the spatial–temporal distribution and change trends of thermal environment are explored by using standard deviation ellipse and spatial autocorrelation combined with land use/land cover (LULC) types. Finally, based on the multi-source remote sensing data, the natural factors and human factors are selected to explore their correlation with the thermal environment using Pearson correlation analysis. The results show that (1) the desert heat island zones are distributed in the desert areas on the periphery of urban built-up areas in Gaochang District, Shanshan County and Yizhou District of Hami Region. The oasis cold island zones are concentrated in the urban built-up areas of Gaochang District and Yizhou District, mainly because the vegetation coverage of urban built-up areas is higher than that of the peripheral desert areas. (2) The spatial development characteristics of the extremely high temperature (EHT) zone and the high temperature (HT)zone from 2005 to 2020 are "southeast-northeast-northwest" and "southeast-northwest" respectively, and the area of construction land 16a increased by 0.31%. This indicates that the spatial evolution of the thermal environment is closely related to the LULC and the degree of urbanization development. (3) From a four–period image with P < 0.001 and Z > 2.58, the thermal environment displays a high positive spatial correlation with Moran's I values of 0.45, 0.54, 0.47, and 0.45. (4) Temperature (Tem), downward longwave radiation (LWdown), and nighttime light intensity (NPP) all exhibited positive correlations with the LST and are significant in the desert region (p < 0.05); The albedo exhibited negative correlations with the LST and is significant in the grassland and woodland regions (p < 0.05). [ABSTRACT FROM AUTHOR]

Soil salinity is recognized worldwide as a major threat to agriculture, particularly in arid regions. Producers and decision-makers thus require updated and accurate maps of salinity in agronomical and environmentally relevant regions. The goals of this study were to test various regression models for estimating soil salt content based on hyperspectral data, HJ-CCD images, and Landsat OLI data using; develop optimal band Difference Index (DI), Ratio Index (RI), and Normalization Index (NDI) algorithms for monitoring soil salt content using image and spectral data; and to compare the performances of the proposed models using a Bootstrap-BP neural network model (Bootstrap-BPNN) from different data sources. The results showed that previously published optimal remote sensing parameters can be applied to estimate the soil salt content in the Ebinur Lake Wetland National Nature Reserve (ELWNNR). Optimal band combination indices based on DI, RI, and NDI were developed for different data sources. Then, the Bootstrap-BP neural network model was built using 1000 groups of Bootstrap samples of remote sensing indices (DI, RI and NDI) and soil salt content. When verifying the accuracy of hyperspectral data, the model yields an R 2 value of 0.95, a root mean square error (RMSE) of 4.38 g/kg, and a residual predictive deviation (RPD) of 3.36. The optimal model for remote sensing images was the first derivative model of Landsat OLI, which yielded R 2 value of 0.91, RMSE of 4.82 g/kg, and RPD of 3.32; these data indicated that this model has a high predictive ability. When comparing the salinization monitoring accuracy of satellite images to that of ground hyperspectral data, the accuracy of the first derivative of the Landsat OLI model was close to that of the hyperspectral parameter model. Soil salt content was inverted using the first derivative of the Landsat OLI model in the study area. [ABSTRACT FROM AUTHOR]

• The number of Karez, a water conservancy cultural heritage, has decreased sharply, facing the dilemma of no water available. • Karez system has water delivery capacity equivalent to that of the modern water conservancy projects and more conducive to water intake. • The reduction of Karez weakens ecological stability and makes the ecosystem more susceptible to external factors. • The appearance of karez increased the diversity of landscape patches, and the change of Karez was consistent with grassland. The water delivery capacity of Karez is gradually declining, and it is possible that such a traditional water supply systems might be completely replaced by modern water conservancy projects. To determine whether these ancient water conservancy projects have a future, we conducted a study in a typical research area of Xinjiang Province, China. Using remote sensing technology, in situ surveying, and the analytic hierarchy process, we evaluated Karez in terms of water delivery capacity, water intake convenience, and capability to maintaining the ecological stability of the surrounding area. The derived results are as follows. (1) During 2005–2020, the length of 14 Karez systems in the study area decreased; however, the calculated water access convenience degree (C) indicated that Karez are more convenient for water intake compared with modern transmission channels, i.e., C karez = 0.68 and C channel = 0.51. (2) The mean annual runoff of the Karez system was 69.38 × 106 m3, similar to that of the modern water transmission channels in the area (73.49 × 106 m3). (3) Change in ecological sensitivity occurred mostly in regions where Karez systems have disappeared, with increase of 70.69 % and 14.51 % in high and medium sensitivity areas, respectively, and decline of 21.53 % and 9.74 % in low-sensitivity and non-sensitivity areas. Our research shows that Karez have considerable water delivery capacity, and that their existence is beneficial to maintaining the stability of the surrounding ecology. The Karez system can be considered a template for harmonious coexistence between humans and nature. [ABSTRACT FROM AUTHOR]

In the past 50 years, the Kaidu–Kongqi River Basin (KKRB) in the arid region of northwest China (ARNC) has experienced drastic climate variability. Meanwhile, from 2000 to 2013, the growths of population and socioeconomic development in the area were drastic, along with much more intensive water management activities. These factors may have caused considerable land use/cover change (LUCC) in the area. Based on the land use/cover classification data derived from the Landsat TM imageries in 1990, 2000, and 2010, as well as the governmental socioeconomic statistics and field observation data, this study investigated the LUCC in the KKRB during 1990–2010. The findings include: (1) The LUCC in the Kaidu–Kongqi River Basin was considerable during the study period, and this change was largely limited to the grassland and cultivated land. The natural grassland in the area decreased with a rate of 118.1 km 2 /y, whereas the cultivated land increased with a rate of 79.2 km 2 /y. The rapid expansion of cultivated land was mainly sourced from reclamations of wasteland and natural grassland. (2) The LUCC has been resulted from the interaction of natural environmental changes and human activities. The changing runoff affected by climate change has played a fundamental role in land use conversion. The human activity intensity index value rose from 0.75 for 1990–2000 to 0.88 for 2000–2010. [ABSTRACT FROM AUTHOR]

Riparian vegetation in arid regions is subject to frequent water environmental fluctuations and plays an important role in maintaining the riparian ecosystem. In recent decades, Populus euphratica (P. euphratica) has been threatened by decreasing groundwater levels, and large areas of P. euphratica seedlings are withered. To better understand the adaptive mechanism underlying hydrological process alterations, we designed an experiment to identify the water sources of P. euphratica seedlings using δ 18 O values (plant stem xylem, soil water and groundwater) under different groundwater scenarios by simulating the wild habitats conditions in the Tarim River bank of Central Asia. Seedlings were grown in lysimeters (2.0 m tall × 0.4 m inner diameter) under varying hydrological conditions. P. euphratica seedlings generally took up water from the shallow soil layer (0–80 cm). Under the T. ramosissima disturbance, the water source depths of the P. euphratica seedlings moved down as the groundwater depth increased. With increasing groundwater depth, the proportions of groundwater used by P. euphratica seedling monocultures (only P. euphratica seedlings) and P. euphratica seedling mixtures ( P. euphratica and T. ramosissima seedlings) were reduced. Under shallow groundwater treatment (W 1 : 25 cm), the plant height growth of the P. euphratica seedlings slowed, and their biomass accumulation decreased. The aboveground and total biomasses of the non-coexistence seedlings under W 2 (75 cm) treatment were maximized, while the maximum value for coexistence seedlings occurred under W 3 (125 cm) treatment. Thus, we suggest that shallow groundwater depth is not beneficial to P. euphratica seedling growth, and appropriately decreasing the groundwater depth may promote their growth. Such information will be valuable to provide the configuration of water resources for a typical river basin in an arid region of Central Asia based on the ecological water requirements of desert riparian vegetation. [ABSTRACT FROM AUTHOR]

Black carbon (BC), which consists of the strongest light-absorbing particles (LAPs) in snow, has been regarded as a potential factor accelerating regional climate change and the melting of snow cover globally. In this study, we used remote sensing (Moderate-resolution Imaging Spectroradiometer, MODIS) observations combined with a snow albedo model (Snow, Ice, and Aerosol Radiation, SNICAR) and a radiative transfer model (Santa Barbara DISORT Atmospheric Radiative Transfer, SBDART) to retrieve the radiative forcing (RF) by BC in snow (R MODIS BC) across Xinjiang, China, for the first time. The observations in January–February show that the concentrations of BC (equivalent BC) in snow ranged from 44.08 to 1949.9 ng g−1, with an average of 536.71 ng g−1. The lowest concentrations of BC were on the border of the Altay region (AR), with a median concentration in snow of 98.5 ng g−1. South of this area in the industrial region (Tianshan Mountain North Slope Economic Development Belt, TMNSEDB), the median concentration of BC in snow was 913.2 ng g−1 R MODIS BC presents distinct spatial variability, with the minimum (3.01W m−2) in the AR and the maximum (40.2W m−2) near industrial areas in TMNSEDB. The regional mean R MODIS BC was 20.43 ± 7.3 W m−2 in Xinjiang, and the average values of the impurity index (I LAPs) and SGS in the region were 0.273 and 241.38 μm, respectively. Moreover, based on the multiple linear regressions, the BC emission intensity values were significantly correlated with I LAPs and RF, and the correlation reached 0.681 and 0.661, respectively; thus, the BC emission could explain above 75% of the spatial variance of BC contents in TMNSEDB, confirming the reasonability of the spatial patterns of retrieved RF MODIS BC in Xinjiang. Additionally, we found that the distribution of R MODIS BC in northern Xinjiang is dominated by I LAPs and BC emissions. We validated R MODIS BC using in situ RF estimates (R site estimate), and the error was 24.05 W m−2; furthermore, the biases in R MODIS BC were negatively correlated with the BC concentrations and ranged from 24.3% to 326% in Xinjiang. • MODIS, observations data and SNICAR is proposed to retrieve RF by BC across Xinjiang, first time. • MODIS derived albedo data is an effective method to RF by BC. • I LAPs and BC emission were very similar to RF in distribution of Xinjiang. [ABSTRACT FROM AUTHOR]

Tarim River Basin is experiencing heavy soil degeneration in a long term because of the extreme natural conditions, added with improper human activities such as reclamation and rejected field repeatedly, which hindered the soil health. One of the mainly form is soil salinization. Spatial distribution and variation of soil salinity is essential both for agricultural resource management and local economic development. However, knowledge of the spatial distribution of soil salinization in this region has not been updated since 1980s while land use and climate have undergone major changed. Electromagnetic induction (EMI) has been successfully used to directly measurement the spatial distribution of targeting soil property at field- scale, and apparent electrical conductivity (ECa, mS m−1) has become a surrogate of soil salinity (EC, dS m−1) studied by many researchers at local scale. However, the effectiveness of this equipment has not been verified in the typical soil salinization areas in southern Xinjiang, especially on a large scale. This study was aimed to test the performance of ECa jointed with Random Forest (RF) for soil salinity regional–scale mapping at a typical arid area, taking Tarim River Basin as an example. The result showed that ECa together with environmental derivative variables and with RF were suited for regional–scale soil salinity mapping. Predicted accuracy of EC was higher at surface (0–20 cm, R2 = 0.65, RMSE = 5.59) and deeper soil depth (60–80 cm, R2 = 0.63, RMSE = 2.00, and 80–100 cm, R2 = 0.61, RMSE = 1.73), lower at transitional zone (20–40 cm, R2 = 0.55, RMSE = 2.66, and 40–60 cm, R2 = 0.51, RMSE = 2.49). When ECa is involved in modeling, the prediction accuracy of multiple depths of EC is improved by 13.33%–61.54%, of which the most obvious depths are 60–80 cm and 0–20 cm. The results of variable importance show that SoilGrids were also favored the power EC model. Hence, we strongly recommended to joint EMI reads with remote sensing imagery for soil salinity monitoring at large scale in southern Xinjiang. These EC and ECa map can provide a data source for environmental modeling, a benchmark against which to evaluate and monitor water and salt dynamics, and a guide for the design of future soil surveys. Unlabelled Image • The CV coefficients of the four ECa coil configurations exceeded 100%. • All calibration and validation RF-ECa models were highly significant (P < 0.001). • Temperature at nights showed the highest impact on ECa distribution. • SoilGrids and WorldClim were important datasets for predicting ECa. • Spatial patterns of ECa were roughly similar to the ECe values in the HWSD database. [ABSTRACT FROM AUTHOR]

• Cubist model performed better than the RF for soil salinity modeling in arid area. • Soil related indices with time-effect fusion have the most important effects on soil salinity predicting. • Our results were broadly consistent with the Harmonized World Soil Database (HWSD) on the whole. • The overall salinity trend is mitigated by 23.58% on average level currently. Precise and spatially explicit regional estimates of soil salinity are necessary to efficiently management and utilise limited land and water resources. Despite advances achieved in remote sensing over the past century, knowledge about the distribution and severity of soil salinization in economically important areas, such as oasis agroecosystems and desert-oasis ecotones (OADoE), is currently limited. An example of an area is southern Xinjiang, where the OADoE has a high anthropogenic influence. This study was conducted with the aim of mapping soil salinity in typical OADoE using remote sensing and machine learning techniques (Cubist and Random Forest, RF). A range of covariates was obtained from the multi-temporal Landsat-8 operational land imager (OLI) satellite for the period from 2013 to 2018. The values of coefficients of determination (R2), Lin's concordance correlation coefficient, root mean square error, and relative root mean squared error values, were 0.78, 0.87, 9.59, and 0.76, respectively, for the Cubist and 0.78, 0.86, 9.79, and 0.78, respectively, for RF models. The slope of the linear fitting equation was higher for the Cubist model (0.75) than for RF (0.69). The explanatory power of Cubist and RF for soil salinity variation were 33.22% and 31.41% in the agroecosystem, and 72.25% and 71.66% in desert-oasis ecotone, respectively. For the agroecosystem, the range of the predicted values for 89.13% (Cubist) and 84.78% (RF) of sample was controlled within the same observational range at an interval of 0–5 dS m−1. Compared to single-year data (from 2013 to 2018), the ability to account for model spatial variability in soil salinity based on multi-year Landsat images was increased by 16%–35%. According to the variable importance evaluation, soil-related indices are the most important predictor variables, followed by vegetation, topography, landform, and land use, with relative importance values of 60%, 21%, 16%, and 3%, respectively. The predicted map was also broadly consistent with those obtained for Xinjiang in the Harmonized World Soil Database (HWSD) from the second national soil survey of China conducted from 1984 to 1997. The results also showed that the average value of the study area is 8.10 dS m−1 based on the Cubist-based map whereas that of the HWSD is 10.60 dS m−1, this implied that the overall salinity level has reduced by 23.58%. The methodological framework presented covers all prediction process steps and has considerable potential to be used in future soil salinity mapping at large scales for other similar region as OADoEs. The map derived from the Cubist/RF model revealed more detailed variation information about spatial distribution of the soil salinity compared to HWSD, and can further assist with decision-making when planning and utilising on existing soil and water resources in OADoEs. [ABSTRACT FROM AUTHOR]

Excessive farming has transformed large proportions of Xinjiang's terrestrial surface, which leads to widespread loss and degradation of ecosystems and biodiversity. Mapping croplands at regional scale and in a rapid and less costly approach is becoming increasingly needed. This study explored an integrated approach based on a combined use of multiple remotely sensed data to map croplands in the oasis of northwest China. The Terra MODIS land surface temperature, Enhanced vegetation index, TRMM precipitation and SRTM digital elevation model data products were examined and combined. The Random Forest models were established to estimate fractional croplands at the regional scale. The influences of basin, images and percentage of croplands were all examined on the accuracy of cropland estimation. Moreover, time series analysis of croplands from 2000 to 2017 was conducted across the basin scale. The results are: ① the integrated approach for cropland mapping at regional scale is promising with the explained variances of fifteen trial models ranging from 58%–93% and two basin scale models ranging from 77.21%–86.10%. ② The prediction accuracy of croplands in Tarim Basin was much higher than that in Junggar Basin indicated by explained variances. ③ Prediction errors increase and then decrease with the increase of percentage of croplands. ④ EVI and DEM were identified to be the most important variable in Junggar and Tarim basin. ⑤ During the past 18 years, the area of croplands has been expanding at approximately 641.3km2/year in Junggar Basin and 271.3 km2/year in Tarim Basin respectively with no sign of cease. The fastest expansion period and the most dramatic changing area were identified. This study is especially valuable for time series analysis of croplands and its correlation with land degradation in arid region. • Approach for large scale and time series prediction of arable land in arid area is raised. • Prediction errors increase and then decrease with the increase of percentage of croplands. • Annual expansion of cropland in Junggar Basin is over two times that of Tarim Basin. • The fastest expansion period and the most dramatic changing area were identified. [ABSTRACT FROM AUTHOR]