Ecological Risk Assessment and Impact Factor Analysis of Alpine Wetland Ecosystem Based on LUCC and Boosted Regression Tree on the Zoige Plateau, China

A machine learning approach to investigate the impact of land use land cover (LULC) changes on groundwater quality, health risks and ecological risks through GIS and response surface methodology (RSM)

Groundwater resources are enormously affected by land use land cover (LULC) dynamics caused by increasing urbanisation, agricultural and household discharge as a result of global population growth. This study investigates the impact of decadal LULC changes in groundwater quality, human and ecological health from 2009 to 2021 in a diverse landscape, West Bengal, India. Using groundwater quality data from 479 wells in 2009 and 734 well in 2021, a recently proposed Water Pollution Index (WPI) was computed, and its geospatial distribution by a machine learning-based 'Empirical Bayesian Kriging' (EBK) tool manifested a decline in water quality since the number of excellent water category decreased from 30.5% to 28% and polluted water increased from 44% to 45%. ANOVA and Friedman tests revealed statistically significant differences (p < 0.0001) in year-wise water quality parameters as well as group comparisons for both years. Landsat 7 and 8 satellite images were used to classify the LULC types applying machine learning tools for both years, and were coupled with response surface methodology (RSM) for the first time, which revealed that the alteration of groundwater quality were attributed to LULC changes, e.g. WPI showed a positive correlation with built-up areas, village-vegetation cover, agricultural lands, and a negative correlation with surface water, barren lands, and forest cover. Expansion in built-up areas by 0.7%, and village-vegetation orchards by 2.3%, accompanied by a reduction in surface water coverage by 0.6%, and 2.4% in croplands caused a 1.5% drop in excellent water and 1% increase in polluted water category. However, ecological risks through the ecological risk index (ERI) exhibited a lower risk in 2021 attributed to reduced high-risk potential zones. This study highlights the potentiality in linking LULC and water quality changes using some advanced statistical tools like GIS and RSM for better management of water quality and landscape ecology.

Identification of ecological security patterns of alpine wetland grasslands based on landscape ecological risks: A study in Zoigê County

Climate change and human activities have increased ecological risks and degraded ecosystem functions in alpine wetland grassland regions, where ecological security remains largely unexplored. The construction of ecological security patterns (ESP) can help to synchronize regional ecological security and sustainable development and provide ideas to address these challenges. This article determines the current ESP of Zoigê County, China, by analyzing the spatial and temporal characteristics of landscape ecological risk (LER) and generating an ecological network by combining the InVEST model, the landscape connectivity index, and the circuit theory model. Management zoning and targeted conservation recommendations are proposed. The results indicate that the region has significant spatial heterogeneity in IER. Ecological risk exposure is increasing, with high values mainly concentrated in the central part of the region. Meanwhile, ecological protection areas were identified, which included 2578.44 km2 of ecological sources, 71 key ecological corridors, 25 potential ecological corridors, 4 river ecological corridors, 66 pinch points, and 58 barriers. This study provides a valuable reference for the ecological development of Zoigê County, as well as insights into the formation of ESP in other alpine wetland grassland regions.

Spatiotemporal changes of ecological environment quality and climate drivers in Zoige Plateau

Ecological environment is the essential material basis of human survival and connects regional economy with socially sustainable development. However, climate changes characterized by global climate warming have caused a series of ecological environmental problems in recent years. Few studies have discussed various climate factors affecting the ecological environment, and the spatial non-stationary effects of different climate factors on the ecological environment are still unclear. Dynamically monitoring ecological environment changes in fragile areas and identifying its climate-driving mechanism are essential for ecological protection and environmental repair. Taking Zoige Plateau as a case, this paper simulated the eco-environmental quality during 1987-2020 using remote sensing data, utilized Geodetector method to identify the contributions of various climate drivers to ecological environment quality, and then adopted the Geographically Weighted Regression model to explore the spatial non-stationary impacts of climate factors on ecological environment quality. The results showed that the ecological quality in the middle regions of the Zoige Plateau was slightly better than in the surrounding marginal areas. For the whole area of Zoige Plateau, the average ecological environment quality index was 54.92, 53.99, 56.17, 57.88, 63.44, 56.93, 59.43, and 59.76 in 1987, 1992, 1997, 2001, 2006, 2013, 2016 and 2020, respectively, which indicated that eco-environmental quality witnessed several fluctuations during the study period but showed a generally increasing trend. Among five climate factors, the temperature was the dominant climate factor affecting the ecological environment quality (q value: 0.11-0.19), sunshine duration (0.03-0.17), wind speed (0.03-0.11), and precipitation (0.03-0.08) were the main climate drivers, while the explanatory power of relative humidity to ecological environment quality was relatively small. Such various climate factors impacting the ecological environment quality demonstrated distinct spatial non-stationary and the range of driving impact varied with time. Temperature, sunshine duration, wind speed, and relative humidity promoted ecological environment quality in most regions (regression coefficients > 0), while precipitation mainly had a negative inhibitory impact (regression coefficients < 0). Meanwhile, the greater impacts of these five climate factors were concentrated in high-elevation regions of the south and west or the northern areas. The appropriate enhancement of climate warming and air humidity was beneficial to the improvement of the ecological environment, but the excessive precipitation would result in landslides and exhibit inhibition of vegetation growth. Therefore, selecting cold-tolerant herbs and shrubs, and strengthening climate monitoring and early warning systems (such as drought and excessive precipitation) are essential for ecological restoration.

Spatiotemporal evolution of land cover changes and landscape ecological risk assessment in the Yellow River Basin, 2015-2020

The Yellow River Basin (YRB), which has faced severe ecological issues since ancient times, is one of the largest and most difficult-to-govern basins in the world. Recently, all provincial governments within the basin have individually enacted a series of measures to protect the Yellow River; however, the lack of central governance has inhibited efforts. Since 2019, the government has comprehensively managed the YRB, improving the governance to unprecedented levels; however, evaluations of the YRB's overall ecological status remain lacking. Using high-resolution data from 2015 to 2020, this study illustrated major land cover transitions, evaluated the correlated overall ecological status of the YRB via the landscape ecological risk index, and analyzed the relationship between risk and landscape structure. The results showed that the (1) main land cover types in the YRB in 2020 are farmland (17.58%), forestland (31.96%), and grassland (41.42%), with urban land accounting for 4.21%. Some social factors were significantly related to changes in major land cover types (e.g., from 2015 to 2020, forest and urban lands have increased by 2.27% and 10.71%, grassland and farmland decreased by 2.58% and 0.63%, respectively). (2) Landscape ecological risk improved, albeit with fluctuations (high in the northwest, low in the southeast). (3) Ecological restoration and governance were imbalanced since no obvious changes were observed in the western source region of the Qinghai Province (Yellow River). (4) Finally, positive impacts of artificial re-greening showed slight lags as the detected improvements in NDVI were not recorded for approximately 2 years. These results can facilitate environmental protection and improve planning policies.

Predicting inhabitable areas for locust based on field observation and multi-environmental factors in alpine grassland—A case study in the Qilian Mountain National Park, China

Alpine grassland is one of the most critical grassland types in the world, and it is vulnerable and sensitive to external disturbances. The development and outbreak of locust might result in the irreversible degradation. However, most locust studies have been on the tropical, temperate, and desert areas. Our knowledge of inhabitable areas in alpine grassland still needs to be explored. This study was carried out in the alpine grassland in the Qilian Mountain National Park. Environmental factors (remote sensing vegetation index, meteorology, soil, topography, and grassland types) and their impact on locust density were investigated. Finally, the inhabitable areas of locust in the study area were mapped. The results showed that: (1) six out of 26 factors [including precipitation, solar radiation (average and maximum value), normalized vegetation index (NDVI), soil, and temperature] had great influence on locust density, with a relative contribution (RC) more than 10%. (2) Among all locust density estimation models, those based on average and maximum solar radiation, maximum precipitation, maximum NDVI, average temperature, and clay content in deep soil performed better than others, with R ranging from 0.58 to 0.73 and root mean square error ranging from 21.70 to 25.82 head/m2. (3) The areas most suited for locust growth, development, and frequent outbreak were found in the south of Tianjun County, middle and northwest of Qilian County (account for 27% of the study area), while the inhabitability was weak in south of Gangcha County, northwest of Tianjun County, and most of Delingha City. Thus our study clarified the distribution region and occurrence variation of the locust and provided a scientific basis for locust prevention and control in alpine grassland in the Qilian Mountain National Park.

Ecological Risk Assessment of Geological Disasters Based on Probability-Loss Framework: A Case Study of Fujian, China

Geological disaster could pose a great threat to human development and ecosystem health. An ecological risk assessment of geological disasters is critical for ecosystem management and prevention of risks. Herein, based on the "probability-loss" theory, a framework integrating the hazard, vulnerability, and potential damage for assessing the ecological risk of geological disasters was proposed and applied to Fujian Province. In the process, a random forest (RF) model was implemented for hazard assessment by integrating multiple factors, and landscape indices were adopted to analyze vulnerability. Meanwhile, ecosystem services and spatial population data were used to characterize the potential damage. Furthermore, the factors and mechanisms that impact the hazard and influence risk were analyzed. The results demonstrate that (1) the regions exhibiting high and very high levels of geological hazard cover an area of 10.72% and 4.59%, respectively, and are predominantly concentrated in the northeast and inland regions, often distributed along river valleys. Normalized difference vegetation index (NDVI), precipitation, elevation, and slope are the most important factors for the hazard. (2) The high ecological risk of the study area shows local clustering and global dispersion. Additionally, human activities have a significant influence on ecological risk. (3) The assessment results based on the RF model have high reliability with a better performance compared with the information quantity model, especially when identifying high-level hazard areas. Our study will improve research on the ecological risk posed by geological disasters and provide effective information for ecological planning and disaster mitigation.

Spatial distribution and ecological risk assessment of heavy metals in alpine grasslands of the Zoige Basin, China

Heavy metals elements are not only related to the functions of ecosystems but also affect human health. To understand the characteristics of heavy metals in the topsoil of the Zoige alpine basin, a total of 252 grass and topsoil samples were collected in May and September 2014. The results showed that only Cd and Pb highly exceeded their background values (BV); in May and September, Cd was 2.02- and 1.55-fold higher than its BV, respectively, and Pb was 2.35- and 2.17-fold above its BV, respectively. The sources of Cd and Pb were homologous. In addition, the comprehensive potential ecological risk index was less than 150, indicating that heavy metal pollution in the study area is currently low. The spatial interpolation indicated that Cd and Pb pollution might be related to tourism and transportation, but the low biological absorption coefficient for all heavy metals showed that heavy metal absorption ability of forage was low and would not impact yak breeding. Finally, the soil was lightly contaminated by Cd and Pb due to the rapid development of the animal husbandry and tourism. The spatial variation of heavy metal in the basin is dominated by structural factors, and the random factors also have an effect on spatial distribution of As, Cd, Cu and Ni. The random factors such as overgrazing can exert an influence on physical structure and the circulation of nutrient substances of meadow soil through livestock grazing and trampling, ultimately affecting the content and distribution of soil heavy metals.

Comprehensive evaluation of the Ruoergai Prairie ecosystem upstream of the Yellow River

With complex and diverse ecosystem types and complete ecological elements such as mountains, rivers, forests, farmlands, lakes, grasslands, sand, and glaciers, the Ruoergai Prairie upstream of the Yellow River is an integral part of the Qinghai-Tibet Ecological barrier and a critical area for ensuring the ecological security of China. In the Ruoergai Prairie, climate change and human activities have led to grassland degradation, water and soil loss, and a shrinking forest area, which has highlighted the need for ecological restoration. Therefore, a comprehensive ecosystem evaluation is of great significance for ecosystem restoration. This study evaluated the ecosystem quality, ecosystem service function importance, ecological vulnerability, ecological protection importance, ecological resilience, and ecological landscape patterns of prairies, wetlands, and forests. The ecosystem quality of the study area was medium to good. The ecosystem service function of the study area with weak ecosystem resilience is important. However, the ecological landscape in the study area has been heavily degraded. Therefore, the protection and restoration of mountains, waters, forests, farmlands, lakes, grasslands, sand, and glaciers are needed.

Temporal and spatial differentiation characteristics of ecosystem service values based on the ecogeographical division of China: a case study in the Yellow River Basin, China

The identification of spatiotemporal changes in ecosystem service values (ESVs) and their drivers is the basis for ecosystem service administration and decision-making. This research focuses on the Yellow River Basin (YRB). With a multitemporal land use and land cover (LULC) dataset (1995-2018), the equivalence coefficient method with spatiotemporal dynamic correction and exploratory spatial data analysis methods were used to evaluate ESV changes due to LULC changes and their spatial characteristics. The contributions of the ESV driving factors and their mutual effects were also investigated via geographic detectors. The results revealed that (1) the land use structure of the YRB, mainly grassland and cultivated land, was stable from 1995 to 2018. However, the transition between land use types was dramatic, including urban expansion accompanied by losses of farmland, grassland, and unused land; increased forestland; and significant increases in water bodies and wetland areas. (2) During the study period, the overall ESV of the YRB increased, and hydrological regulation and climate regulation services dominated the change in the ESVs in the study area. The ESV exhibited obvious ecogeographical pattern differentiation and evident positive spatial autocorrelation. High values were concentrated in the southern part of the study area, including the southeastern part of the Qinghai-Tibet Plateau region and the central part of the East Asian monsoon region. Low values were concentrated in the northwestern arid zone, dominated by desert and grassland ecosystems. (3) Because of the fragility of the regional ecological background, the spatial differentiation of the ESVs in the YRB is dependent on natural factors; however, anthropogenic factors such as the degree of land use and the human activity intensity also lead to ESV differentiation. The synergistic effects of human activities, landscape pattern changes, and natural factors result in the spatial differentiation of the ESVs in the research region. Therefore, human activities affecting the ecological environment should be controlled, nature-based solutions should be advocated, patch diversity should be increased, landscape fragmentation should be reduced, LULC ecosystem service functions should be improved, and the relationship among economic, social, and ecological landscape resources should be coordinated.

Landscape ecological risk assessment and driving factor analysis in Dongjiang river watershed

The ecological and environmental quality of Dongjiang river watershed has great influence on Guangdong, Hong Kong and Macao. The landscape ecological risk assessment model could effectively monitor and assess environmental quality. In this study, spatial autocorrelation and geographic detector methods were used to explore the spatial characteristics of landscape ecological risk and their driving factors in the Dongjiang river watershed for four decades. The results showed that the ecological risks of Dongjiang River Source Watershed are mainly classified as low and intermediate, which are distributed in the hilly regions and the marginal mountainous regions at the junction of the Xunwu and Dingnan counties. From 1980 to 2018, the area of regions with the low ecological risk increased by 587.01 km 2. The size of regions with moderate, high and severe ecological risk decreased by 165.6 km 2, 258.82 km2 and 162.58 km2, respectively. Moreover, landscape ecological risk values exhibited an apparent spatial dependency, and high-risk areas cluster together. Among influencing factors, population density has the most significant impact on the change of landscape ecological risk in the Dongjiang river watershed, followed by elevation (DEM), human interface, vegetation index (NDVI), and urbanization level. However, the interaction of driving factors has a greater impact on the ecological risk of the Dongjiang river watershed than a single driving factor. The research provides good knowledge for environmental quality management, and the proposed methods can be used for other regions.

Multi-scenario simulation of ecological risk assessment based on ecosystem service values in the Beijing-Tianjin-Hebei region

In this study, a framework for ecological risk assessment based on ecosystem service values and risk probability was established. Remote sensing was used to estimate the value of ecosystem services at the regional scale. Considering the natural and anthropogenic factors and using the entropy weight method to assign weights, probability index was constructed. In addition, multiple scenarios based on the ordered weighted averaging (OWA) method were simulated to reduce subjective uncertainty in the assessment. The results showed that the ecosystem service values generated by the gas regulation value accounted for the largest proportion, with a ratio of 46% in the Beijing-Tianjin-Hebei region. From 2005 to 2015, the value of ecosystem services decreased, falling by 2.5 × 10⁷ Yuan. The level of ecological risk was relatively high, with a corresponding area ratio of 32.89%. Spatially, the areas with high risk were concentrated in the southeastern areas, and areas with relatively low risk were distributed in the western and northern areas. This high risk was probably caused by urbanization which was characterized by reduction of farmland and increase in impervious surface. Multi-scenario simulation showed that the areas of unstable ecological risk zones covered 30% and were mainly concentrated in the surroundings of developing cities. In areas of unstable risk distribution, the relationship between development and protection should be considered. This framework increases the reliability and practicability of ecological risk assessment results and has potential application value for regional risk control in the context of urbanization.

Ecological Risk Assessment of Transboundary Region Based on Land-Cover Change: A Case Study of Gandaki River Basin, Himalayas

Land-cover change is a major cause of global ecosystem degradation, a severe threat to sustainable development and human welfare. In mountainous regions that cross national political boundaries, sensitive and fragile ecosystems are under complex disturbance pressures. Land-cover change may further exacerbate ecological risks in these regions. However, few studies have assessed the ecological risks in transboundary areas. This study focused on the Gandaki Basin (GRB), a typical transboundary region in the Himalayas. Based on the dynamic change in land cover, the landscape ecological risk index (ERI) model was constructed to assess the ecological risk in the GRB, revealing the evolution characteristics and spatial correlation of such a risk during the period 1990–2020. The results showed that all land cover types in the GRB have changed over the last 30 years. The interconversion of cropland and forestland was a distinctive feature in all periods. Overall, the medium and medium to low ecological risk level areas account for approximately 65% of the study area. The areas of high ecological risk were mainly distributed in the high elevation mountains of the northern Himalayas, while the low risk areas were located in the other mountains and hills of Nepal. In addition, the ecological risk in the Gandaki basin has shown a fluctuating trend of increasing over the past 30 years. However, there were different phases, with the order of ecological risk being 2020 > 2000 > 2010 > 1990. Ecological risks displayed positive spatial correlation and aggregation characteristics across periods. The high–high risk clusters were primarily located in the high and medium high ecological risk areas, while the low–low risk clusters were similar to low risk levels region. The findings provided the reference for ecosystem conservation and landscape management in transboundary areas.

Changes in Landscape Pattern and an Ecological Risk Assessment of the Changshagongma Wetland Nature Reserve

The Changshagongma wetlands is the Chinese National Nature Reserve were listed as a Ramsar Wetland of International Importance in 2018. Here, we examined four periods (1992, 2002, 2013, and 2020) of remote sensing image data to analyze the changes in wetland landscape patterns and the ecological risk in Changshagongma Wetland Nature Reserve over the past 30 years. The results showed that wetlands account for approximately 30% of the study area, and swamp meadows were the main type of wetland, accounting for approximately 95% of the total wetland area. In terms of landscape patterns, wetland fragmentation declined, wetland patch shapes became less complicated, and spatial connectivity increased. The landscape fragmentation of non-wetland alpine meadows was reduced. The patches of sandy grasslands tended to be regular, and their spatial connectivity was reduced. The wetland regions of high ecological risk are concentrated in the central and southern parts of the Changshagongma Wetland Nature Reserve. Low-risk regions are mainly concentrated in the contiguous swamp meadows in the northwest and wetlands in the southwest. From 1992 to 2020, the level of ecological risk of the Changshagongma Wetland Nature Reserve showed a “∧”-shaped trend, with the highest risk in 2002 and the lowest risk in 2020. Among the selected indicators, climate conditions constituted the main factor affecting the ecological risk of the Changshagongma Wetland Nature Reserve, followed by topographical conditions, and human activities were the least influential. Over the past 30 years, the temperature and precipitation in the study area increased significantly. The climate in the study area can be roughly divided into two periods bounding 2002, and the climate has been changing from cold and dry to warm and wet. The ecological environment of the study area is affected by natural and human activities. Cold and dry climatic conditions and uncontrolled grazing accelerate the destruction of the wetland ecological environment, and warm and wet climatic conditions and ecological conservation policies are conducive to the ecological restoration of wetlands. In general, the wetland landscape structure in the study area has become less complex, landscape heterogeneity has decreased, and ecological quality has improved.

Identification of Priority Conservation Areas for Natural Heritage Sites Integrating Landscape Ecological Risks and Ecosystem Services: A Case Study in the Bogda, China

The conservation of World Natural Heritage Sites has become a global concern. The identification of priority conservation areas can preserve the value of heritage sites while promoting sustainable development, which is important for balancing the conservation and development of heritage sites. This paper proposes an integrated framework for the identification of priority conservation areas for natural heritage sites based on landscape ecological risks (LERs) and ecosystem services (ESs), taking the Bogda heritage site in Xinjiang, China as a case study. The innovative approach combined the natural and cultural elements of natural heritage sites and included the following steps: (1) the LER index, Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST) model and questionnaire method were adopted to assess the LERs and ESs of Bogda heritage sites during 1990–2018; (2) ordered weighted averaging (OWA) was used to identify conservation priorities by weighing LERs and ESs; and (3) the optimal priority conservation area was determined by comparing the conservation efficiencies under different scenarios. The results revealed that the LER, carbon storage (CS), habitat quality (HQ), aesthetic value (AV), and recreational value (RV) showed significant spatiotemporal variation. The most suitable priority conservation area was located at the central forestlands and high-coverage grasslands, with conservation efficiencies of 1.16, 2.91, 1.96, 1.03, and 1.21 for LER, CS, HQ, AV, and RV, respectively. Our study demonstrated that integrating LERs and ESs is a comprehensive and effective approach to identifying conservation priorities for heritage sites. The results can provide decision support for the conservation of the Bogda heritage site and a methodological reference for identifying conservation priorities for natural heritage sites. Furthermore, this study is also an effective application of LERs and ESs in identifying priority conservation areas.

Coupling Coordination Analysis and Prediction of Landscape Ecological Risks and Ecosystem Services in the Min River Basin

Watershed landscape ecological security and ecosystem service functions are the material basis and environmental guarantee for promoting socioeconomic development. Analyzing the spatiotemporal characteristics of landscape ecological risks (LERs) and ecosystem services (ESs) and exploring the coupling coordination relationship between the two are of great significance for promoting the construction of ecological civilization and achieving sustainable development in the watershed. With the Min River Basin as the study area, the landscape ecological risk assessment, Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST), and Carnegie Ames–Stanford Approach (CASA) models were used to evaluate the LERs and ESs based on the shared socioeconomic pathways (SSPs), and the patch-generating land use simulation (PLUS) model was used to predict the land use distribution of the Min River Basin in 2030. On this basis, the coupling coordination degree model was used to explore the coupling coordination relationship between the LERs and ESs. The results show that, from 2000 to 2020, the LER of the Min River Basin gradually decreased, and the overall spatial distribution pattern was “high in the north and low in the south”. The ES of the Min River Basin initially decreased and then increased, showing a spatial distribution pattern of “low in the south and high in the north”. Among the SSPs in 2030, the LER is the largest under the SSP3 scenario and the smallest under the SSP4 scenario. The ES improvement is the most significant under the SSP1 scenario and the lowest under the SSP3 scenario. From 2000 to 2030, the coupling coordination degree of the Min River Basin first decreased and then increased, showing a spatial distribution pattern of “high in the south and low in the north”. Among the five SSPs, the coupling coordination degree was the highest under SSP1. The spatial distribution of urban area is the main driving factor affecting the coupling coordination relationship between the LER and ES, and the development of social and economy is the beginning of landscape pattern optimization.

Mapping of Kobresia pygmaea Community Based on Umanned Aerial Vehicle Technology and Gaofen Remote Sensing Data in Alpine Meadow Grassland: A Case Study in Eastern of Qinghai–Tibetan Plateau

The Kobresia pygmaea (KP) community is a key succession stage of alpine meadow degradation on the Qinghai–Tibet Plateau (QTP). However, most of the grassland classification and mapping studies have been performed at the grassland type level. The spatial distribution and impact factors of KP on the QTP are still unclear. In this study, field measurements of the grassland vegetation community in the eastern part of the QTP (Counties of Zeku, Henan and Maqu) from 2015 to 2019 were acquired using unmanned aerial vehicle (UAV) technology. The machine learning algorithms for grassland vegetation community classification were constructed by combining Gaofen satellite images and topographic indices. Then, the spatial distribution of KP community was mapped. The results showed that: (1) For all field observed sites, the alpine meadow vegetation communities demonstrated a considerable spatial heterogeneity. The traditional classification methods can hardly distinguish those communities due to the high similarity of their spectral characteristics. (2) The random forest method based on the combination of satellite vegetation indices, texture feature and topographic indices exhibited the best performance in three counties, with overall accuracy and Kappa coefficient ranged from 74.06% to 83.92% and 0.65 to 0.80, respectively. (3) As a whole, the area of KP community reached 1434.07 km2, and accounted for 7.20% of the study area. We concluded that the combination of satellite remote sensing, UAV surveying and machine learning can be used for KP classification and mapping at community level.

Land Use/Land Cover Changes and Their Driving Factors in the Northeastern Tibetan Plateau Based on Geographical Detectors and Google Earth Engine: A Case Study in Gannan Prefecture

As an important production base for livestock and a unique ecological zone in China, the northeast Tibetan Plateau has experienced dramatic land use/land cover (LULC) changes with increasing human activities and continuous climate change. However, extensive cloud cover limits the ability of optical remote sensing satellites to monitor accurately LULC changes in this area. To overcome this problem in LULC mapping in the Ganan Prefecture, 2000–2018, we used the dense time stacking of multi-temporal Landsat images and random forest algorithm based on the Google Earth Engine (GEE) platform. The dynamic trends of LULC changes were analyzed, and geographical detectors quantitatively evaluated the key driving factors of these changes. The results showed that (1) the overall classification accuracy varied between 89.14% and 91.41%, and the kappa values were greater than 86.55%, indicating that the classification results were reliably accurate. (2) The major LULC types in the study area were grassland and forest, and their area accounted for 50% and 25%, respectively. During the study period, the grassland area decreased, while the area of forest land and construction land increased to varying degrees. The land-use intensity presents multi-level intensity, and it was higher in the northeast than that in the southwest. (3) Elevation and population density were the major driving factors of LULC changes, and economic development has also significantly affected LULC. These findings revealed the main factors driving LULC changes in Gannan Prefecture and provided a reference for assisting in the development of sustainable land management and ecological protection policy decisions.