Search

Vegetation is usually sparse in the desert regions, but it plays important roles in stabilizing sand dunes and combating desertification. Establishing how desert vegetation responds to changes in both natural forcing and anthropogenic interference is essential for better understanding desertification processes and their future dynamics. In this study, a comprehensive analysis of NDVI time series data in combination with climate reanalysis data and topographic data was conducted to investigate the possible effects of different external factors on desert vegetation in the Tengger desert, one of the largest deserts in northern China. The results show that vegetation is mostly distributed near the desert margins and in the interdune areas with relatively low and flat topography. Multiple NDVI datasets indicate a consistently greening trend over the entire desert during the last forty years, and the greening rates are higher at sites with sand fixation and vegetation restoration practices. After a few decades of restoration practices, vegetation greenness in these sites is approaching their natural states, but they still show large interannual variability associated with precipitation fluctuations. Therefore, the greening trend of the entire desert could be related to recent climate change towards wetter, warmer and less windy conditions, while human efforts have accelerated the rate of vegetation recovery in the restoration sites. Our study implies that the present climate change has produced conditions favorable for continued vegetation increase in the drylands of northern China, and this trend has been facilitated by policy-driven restoration projects. The combined effects of climate change, topography and human interference on desert vegetation should be considered in future restoration practices.

Understanding the trend of vegetation change and its reaction to climate variation is important for revealing the mechanism of ecosystem behavior. However, current research rarely systematically analyzes the time effects of climate variation on vegetation dynamics (time-lag and time-accumulation effects), especially in arid and semi-arid mountainous terrain. The typical mountainous terrain—the Qilian Mountains was taken as the study area, and the spatiotemporal changes and vertical zonality distributions of the normalized vegetation index (NDVI) were explored. This study explored the time-lag and time-accumulation effects of the NDVI response to climate factors (precipitation, temperature), identified the main controlling factors that influence the variation of NDVI. The results show that in the growing season from 2000 to 2019, the NDVI represented an overall upward trend, especially in the northwest, and the growth rate of NDVI at low-altitude was greater. The time-accumulation effect of precipitation has an obvious effect on vegetation, especially on deserta and meadow; and the time-lag and time-accumulation effects of temperature have an obvious influence. Regarding the climate-vegetation response mechanism, this study finds that considering the optimal time effect is of great significance. In addition, compared with precipitation, the temperature has a more significant promotion effect on vegetation growth in the Qilian Mountains. The above results indicate that when the existing climate models study vegetation-climate interactions, considering the time effects of vegetation response to climate is of great significance for accurately monitoring vegetation dynamics under environmental changes.

Although correlations between soil respiration (Rs) and environmental factors in different ecosystems have been extensively investigated, results based on long-term experiments are limited. In this study, we analyzed the correlations between Rs and soil temperature (Ts), moisture (θ), the normalized difference vegetation index (NDVI), and soil organic matter (SOM) based on an 11-year field campaign in abandoned farmland (HD) and bare land (LD) areas with a Cambisol soil type in the eastern Loess Plateau of China. The results showed that the mean annual Rs and annual temperature sensitivity (Q10) values in HD were significantly larger than those in LD during the measurement period. Rs and the annual contribution ratio (Rratio) of root respiration (Rr) to Rs in HD increased significantly with vegetation recovery years. On the annual scale, Rs was controlled primarily by Ts at both sites. However, in the summer months, Rs was controlled by θ rather than Ts. A two-variable model, including Ts and θ, could more accurately predict the temporal variations in Rs at the annual and seasonal time scales than a single-variable model, and the explanatory power of the equations in HD was significantly larger than that in LD. On the interannual scale, variations in the annual Rs and Rr showed significant positive correlations with the annual θ and NDVI and SOM, and negative correlations with Ts in HD. Moreover, the annual Q10 was significantly positively related to θ in HD. In LD, the only significant relationship was between the annual Q10 and the annual θ. All differences between HD and LD in Rs, Ts, and θ and their correlations were attributed to differences in vegetation. Our findings highlight the importance of long-term observations for the accurate estimation of CO2 efflux in semiarid regions and the impact of vegetation changes on soil respiration.

Rock fragments (RFs, material particles with diameter > 2 mm) are widely distributed in soils over globe. However, spatial pattern of RF content (RFC) and its roles on hydrological and ecological patterns have remained unclear at the global scale. In this study, based on the collected global datasets, we investigated the relationships between mean annual temperature (MAT), mean annual precipitation (MAP), elevation and RFC at 0–5 cm soil depth. In addition, we also examined the links of RFC to representative hydrological and ecological indicators (soil water content or SWC, and normalized difference vegetation index or NDVI, respectively). Results showed that through affecting the pedogenic and morphogenic processes, temperature, precipitation and elevation presented significant relationships with RFC over the globe. When MAT 20 ℃, RFC slightly increased with MAT increasing, due to that extreme high temperature induced arid condition would surpass soil production. RFC decreased with MAP increasing, because precipitation increased SWC and hydrologic flow and thus enhanced soil production. In addition, higher elevation zones usually had colder condition and greater loss of fine-textured soils due to erosion, and thus RFC increased with elevation increasing. Furthermore, a RFC threshold of about 0.2 m3 m−3 was identified. When RFC 0.2 m3 m−3, due to poor soil water and nutrient retention capacity, as well as lower MAT with higher elevation, or lower MAP, lower SWC and NDVI were observed. These findings can supplement the lack of knowledge on RFs and its effects from the global perspective.

Urmia Lake in Northern Iran is drying up, which is causing significant environmental problems in the region, including saline storms that devastate agricultural land. We developed a remote sensing-based monitoring application to detect and map the location of saline flow sources with a novel automated deep learning convolutional neural network (DL-CCN). In order to train the model, we derived a normalised difference dust index (NDDI) from MODIS satellite images and collected ground control points (GCPs). These GCPs were randomly split for training (70%) and accuracy assessment (30%). We identified the following seven predisposing factors for saline flow source modelling: normalised difference vegetation index (NDVI), humidity percentage, temperature, wind speed, geomorphology, soil and land use/cover. In order to train the DL-CNN, we used ReLu, the root mean square error function, and Stochastic Gradient Descent (SGD) for the activation, loss/cost function, and optimization, respectively. Finally, we used the frequency ratio (FR) method to identify the most effective variable for the prediction of saline storm occurrences. The results reveal a high confidence (91.86% overall accuracy and a Kappa of 90.26) for the detection of saline flow sources. According to the FR model, the NDVI (0.982), humidity percentage (0.963), and land use/cover (0.925) are the most relevant factors for detecting the occurrence of saline storms in the Urmia Lake basin. In addition, we carried out a spatial uncertainty analysis of the results based on the Dempster Shafer Theory. The results will help the local stakeholders and decision-makers to better understating the saline flow sources and their respective environmental impacts.

Vegetation in arid/semiarid areas in the northern hemisphere is highly sensitive to precipitation variability. Understanding the change of dryland vegetation and its response to spatiotemporal patterns of precipitation changes is critical to delve impacts of meteorological droughts on dryland vegetation. Based on the Normalized Difference Vegetation Index (NDVI) and precipitation data, we delineated the spatiotemporal pattern of trends of the arid/semiarid plant growth and its responses to daily precipitation changes in the northern Xinjiang. The results indicated several regular characteristics in the NDVI responses to precipitation at unequal time and spatial scales during the growing season, i.e. April-October. It identified shortening time lags of NDVI responses to precipitation over time. This study found for the first time that precipitation changes tend to impact NDVI changes given dry period length (DL) of > 18 days, and the effect of precipitation process was more significant. Three temporal distribution types of a precipitation process were classified, i.e. A-type or B-type precipitation process, which are characterized by two consecutive rainfall days with less or larger precipitation amount on the first day than the next day, and C-type precipitation process, which is characterized by three or more consecutive rainfall days. A- and B-type precipitation processes can benefit availability of moisture in the medium and early stage of the growing season respectively. Precipitation of 6.1–12.0 mm is the most favorable condition for the growth of desert vegetation. This study highlights multifarious vegetation responses to different precipitation processes and provides theoretical ground for ecological restoration and conservation in a changing environment across northern Xinjiang, China, and also in drylands in other regions of the globe.

Vegetation is essential in maintaining terrestrial ecosystem functions, and understanding why and how vegetation evolves is necessary for regional ecological management. On the Tibetan Plateau, the effects of various factors and their interactions in directing vegetation change remain unclear. Supported by the Google Earth Engine (GEE) platform, we quantified spatiotemporal vegetation variation in Gannan Prefecture on the northeastern Tibetan Plateau for 2000–2018 using trend analysis and spatial autocorrelation and explored its driving factors with a geographic detector method. The normalized difference vegetation index (NDVI) illustrated fluctuating growth from 2000 to 2018, and the overall growth rate was 0.002/year. The spatial distributions of NDVI were variable, and areas with NDVI greater than 0.8 accounted for more than 69% of Gannan Prefecture area. From 2000 to 2018, the global Moran’s index of vegetation NDVI was greater than 0.52 and showed a fluctuating upward trend, indicating that the vegetation NDVI tended to be distributed with a high spatial concentration. The local Moran’s index of NDVI was mainly characterized by “High-High” and “Low-Low” clustering types, and the former increased significantly, but the latter decreased. The annual mean temperature, soil type, and elevation were the dominant factors driving vegetation NDVI change in Gannan Prefecture, explaining more than 15% of the variability. Furthermore, the influence of the interaction between any two factors on NDVI was an almost nonlinear enhancement. These results could help us improve our understanding of the underlying mechanism of NDVI changes and provide a scientific basis for ecological protection in alpine regions.

The salinized wasteland (SW) is a buffer zone to accumulate salt introduced from irrigation water in salinized irrigation districts. It is crucial to identify the locations and sizes of SW to achieve effective land management and reclamation, as well as agricultural sustainable development. However, the SW is less explored due to the lack of low-cost and effective techniques. In this study, a decision tree-based remote sensing model was proposed to distinguished SW from other land types using the high-resolution Gaofen-1 (GF-1) images in the Hetao Irrigation District of China during 2017–2020. Meanwhile, multiple validation approaches including points (field and random points) and area (small and medium-sized SWs, i.e., SWA, SWB) matching validations were adopted. To establish the model of SW, four spectral indices including Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), Normalized Difference Water Index (NDWI), and Salinity Index (SI), as well as a series of multi-period and multi-year remote sensing images were used. The threshold values of the four spectral indices were determined based on the probability density curve. The SW remote sensing model was highly accurate and reliable. The average producer accuracy (PA) and average user accuracy (UA) were respectively 88.0% and 81.6% for field investigation (field points matching validation), and 85.9% and 80.1% for Google Earth image investigation (random points matching validation). Besides, the average overlapping rate and misclassification rate were respectively 95.6% and 2.3% for small-sized typical SWs (area validation), and 92.9% and 4.7% for medium-sized typical SWs during 2017–2020. The model indicated that the SW area in the region was respectively 9969, 9694, 9660, and 9608 ha from 2017 to 2020. The area of small and medium-sized SWs decreased with time, while that of large-sized SWs was the opposite.

In the digital soil mapping (DSM) framework, machine learning models quantify the relationship between soil observations and environmental covariates. Generally, the most commonly used covariates (MCC; e.g., topographic attributes and single-time remote sensing data, and legacy maps) were employed in DSM studies. Additionally, remote sensing time-series (RST) data can provide useful information for soil mapping. Therefore, the main aims of the study are to compare the MCC, the monthly Sentinel-2 time-series of vegetation indices dataset, and the combination of datasets (MCC + RST) for soil organic carbon (SOC) prediction in an arid agroecosystem in Iran. We used different machine learning algorithms, including random forest (RF), Cubist, support vector machine (SVM), and partial least square regression (PLSR). A total of 237 soil samples at 0–20 cm depths were collected. The 5-fold cross-validation technique was used to evaluate the modeling performance, and 50 bootstrap models were applied to quantify the prediction uncertainty. The results showed that the Cubist model performed the best with the MCC dataset (R2 = 0.35, RMSE = 0.26%) and the combined dataset of MCC and RST (R2 = 0.33, RMSE = 0.27%), while the RF model showed better results for the RST dataset (R2 = 0.10, RMSE = 0.31%). Soil properties could explain the SOC variation in MCC and combined datasets (66.35% and 50.82%, respectively), while NDVI was the most controlling factor in the RST (50.22%). Accordingly, results showed that time-series vegetation indices did not have enough potential to increase SOC prediction accuracy. However, the combination of MCC and RST datasets produced SOC spatial maps with lower uncertainty. Therefore, future studies are required to explicitly explain the efficiency of time-series remotely-sensed data and their interrelationship with environmental covariates to predict SOC in arid regions with low SOC content.

Land use and land cover change (LUCC) will directly affect the types, structures and functions of ecosystems, and then have important impact on vegetation net primary productivity (NPP). The evaluation of the spatio-temporal dynamic changes of LUCC and regional NPP, and assessing the impact of LUCC on NPP can facilitate the corresponding management of different environmental types of regions and provide scientific guidance for scientific development of ecological resources, and better restore and control the ecological environment, thus promoting rapid economic development. Using MODIS normalized difference vegetation index (NDVI) data and meteorological data from 2001 to 2018, combined with Carnegie-Ames-Stanford model (CASA), the regional NPP was estimated, and the influence of LUCC on NPP in Yangtze River basin from 2001 to 2018 and its causes were analyzed. The results showed that the area with a land cover and land use change account for 10.98% of the total study area in the past 18 years, and the most important change is the transformation from grassland to forest and grassland to crop/natural vegetation mosaic (NVM). Forests, wetlands, crop/natural vegetation mosaic, low vegetation cover (LVC) land, urban, and water bodies are growing. The forest land area increased the most and that of grassland decreased the most. Total NPP has increased by 53887.51GgC in the past 18 years, which was due to the combination effect of LUCC change and climate change.