Landsat Data Research Articles

Utilizing Landsat, Sentinel Image to Estimate Snow Cover Duration in Gara Mountain

The study area is considered a mountainous environment with climatic characteristics that suit the nature of this study. It is located in the northern part of the Kurdistan Region of Iraq, and it is part of a complex mountainous region according to the physiographic division of the surface of Iraq. On the other hand, it’s one of the most significant in terms of snow cover region in winter and spring season. No study has been done on this subject in the study area. In this study, Landsat data and geographic information system (GIS) techniques were used to analyze the duration of snow cover. After apply these technologies, snow cover duration maps (SCDM) for the duration (1989 – 1990) and (2022 – 2023) were extracted and analyzed. The maps show how long snow remains in each cell during these periods. This study suggested that there is an effect of terrain on snow cover duration, and this is due to the increase in air temperature.

Assessing mangrove cover change in Madagascar (1972–2019): Widespread mangrove deforestation is slowing down

Between the 1970s and early 2000s, Madagascar witnessed significant mangrove extent and condition declines due to deforestation and degradation. Data on mangrove extent prior to the early 1990s is scarce, contributing to uncertainties in long-term trend analyses. Recent estimates (since 2010) have reported widely varying mangrove extents from 2000 km² to 3400 km², indicating large uncertainties in current trends. Leveraging five decades of Landsat satellite data, this study offers a methodologically consistent approach to classify Madagascar's mangrove extent, providing a comprehensive dataset from 1972, 1989, 1999, 2009, to 2019. Our findings reveal an overall decrease of 8 % from 2935 km² in 1972–2699 km² in 2019, with loss rates decelerating from 0.4 % per year in the initial period to 0.2 % per year post-1999. This decline primarily reflects anthropogenic pressures, notably in the northern regions. Conversely, the last decade has witnessed a 5 % increase in mangrove coverage, primarily in the central and southern regions, thanks to concerted conservation and reforestation efforts, highlighting a positive shift in mangrove management and protection strategies. This study's long-term perspective is crucial for understanding the dynamics of mangrove coverage and guiding effective intervention strategies in Madagascar.

Multitemporal Analysis of the Relationship Between Vegetation Density and Built-Up Land to the Urban Heat Island Phenomenon in Cirebon City

Abstract The attractiveness of Cirebon City as a target for urbanization lies in its strategic location and significant development, particularly in tourism, the service industry, and culture. As a result, vegetation has been transformed into urban development, leading to a subsequent rise in land surface temperatures and causing the Urban Heat Island (UHI) phenomenon. UHI is prevalent in urban areas and diminishes in peripheral regions. This research aims to investigate the correlation between vegetation density and built-up areas to the UHI phenomenon in Cirebon City, utilizing Landsat-8 data for the years 2014 and 2023. Normalized Difference Built-Up Index (NDBI) is used to identify built-up areas, vegetation density is identified with the Normalized Difference Vegetation Index (NDVI) algorithm, and Land Surface Temperature (LST) data is obtained from thermal band extraction. The results of the LST analysis were used to describe the Urban UHI phenomenon. Findings throughout the study period revealed significant changes in built-up land and vegetation density, while dynamic changes occurred in LST and UHI. The results also show that there is a positive correlation between the built-up land index and UHI, which indicates a unidirectional relationship. Meanwhile, the vegetation density index to UHI is negatively correlated, which means it shows an inverse relationship.

Assessment of Continuous Growth of Glacial Lakes in the Teesta River Basin Using Semi-Automated Geospatial Approach

Global warming is one of the primary causes contributing to melting glaciers and shrinking of glaciers moth. Because of the glacier retreat, more lakes increase the risk of flooding in people’s homes and lives. Several studies on the surging glaciers have been conducted by researchers using various techniques, as well as with the aid of multiple models like the Normalized Differential Water Index (NDWI). The Number of glacial lakes is increasing in the Himalayan region due to climate change (rise of the temperature). Some glacial lakes are potentially dangerous so monitoring is very necessary. It is necessary to evaluate such vulnerable lakes. Therefore, current work is carried out to identify such glacial lakes present in the Teesta River Basin (Eastern Himalaya). Spatiotemporal Landsat data for the last four decades at intervals of ten years from 1990 to 2020 has been considered which was cloud-free and spatial resolution of 30 meters. The dataset mentioned above was used for lake identification and delineation. The findings indicate the presence of lakes with respective areas of 275 (18.90 km2), 337 (24.92 km2), 295 (22.96 km2), and 419 (31.44 km2). It has also been observed that the growth rate is increasing with approximate water spread from 1990 to 2000 (+129%), 2000 to 2010 (+106%), and 2010 to 2020 (+136%). The present study aimed to identify such glacial lakes based on their water spreading area, which is an essential step followed in the study of GLOF (Glacial Lake Outburst Flood) as it will be helpful in the identification of hazardous lakes. In that study, we found that eleven glacial lakes are in the potentially dangerous category situated in the upper Teesta Basin due to the presence of glaciers, which gives a clear reason for the time-to-time assessment of such lakes. By the conducted study it has been observed that the number of glacial lakes has increased, due to which water spread has also increased in the area. It can also be demonstrated that GIS (Geographical Information System), along with remote sensing, is one of the best tools for assessing and monitoring such change detection and differentiation of hazardous glacial lakes in the cryosphere, along with the supporting data.

The Performance of Landsat-8 and Landsat-9 Data for Water Body Extraction Based on Various Water Indices: A Comparative Analysis

The rapid and accurate extraction of water information from satellite imagery has been a crucial topic in remote sensing applications and has important value in water resources management, water environment monitoring, and disaster emergency management. Although the OLI-2 sensor onboard Landsat-9 is similar to the well-known OLI onboard Landsat-8, there were significant differences in the average absolute percentage change in the bands for water detection. Additionally, the performance of Landsat-9 in water body extraction is yet to be fully understood. Therefore, it is crucial to conduct comparative studies to evaluate the water extraction performance of Landsat-9 with Landsat-8. In this study, we analyze the performance of simultaneous Landsat-8 and Landsat-9 data for water body extraction based on eight common water indices (Normalized Difference Water Index (NDWI) and Modified Normalized Difference Water Index (MNDWI), Augmented Normalized Difference Water Index (ANDWI), Water Index 2015 (WI2015), tasseled cap wetness index (TCW), Automated Water Extraction Index for scenes with shadows (AWEIsh) and without shadows (AWEInsh) and Multi-Band Water Index (MBWI)) to extract water bodies in seven study sites worldwide. The Otsu algorithm is utilized to automatically determine the optimal segmentation threshold for water body extraction. The results showed that (1) Landsat-9 satellite data can be used for water body extraction effectively, with results consistent with those from Landsat-8. The eight selected water indices in this study are applicable to both Landsat-8 and Landsat-9 satellites. (2) The NDWI index shows a larger variability in accuracy compared to other indices when used on Landsat-8 and Landsat-9 imagery. Therefore, additional caution should be exercised when using the NDWI for water body analysis with both Landsat-8 and Landsat-9 satellites simultaneously. (3) For Landsat-8 and Landsat-9 imagery, ratio-based water indices tend to have more omission errors, while difference-based indices are more prone to commission errors. Overall, ratio-based indices exhibit greater variability in overall accuracy, whereas difference-based indices demonstrate lower sensitivity to variations in the study area, showing smaller overall accuracy fluctuations and higher robustness. This study can provide necessary references for the selection of water indices based on the newest Landsat-9 data. The results are crucial for guiding the combined use of Landsat-8 and Landsat-9 for global surface water mapping and understanding its long-term changes.

Spatiotemporal evolution and driving mechanisms of desertification on the Mongolian Plateau

Desertification poses a severe ecological and environmental challenge in the Mongolian Plateau (MP). It is difficult to quantify desertification distribution using unified indicators in the entire MP, because of its complex physical geographic conditions and various climatic zones covered. To accurately address this challenge, the spatial distribution of desertification at a 30-m resolution from 1990 to 2020 were mapped in this study. The desertification potential occurrence zone was identified by using a moisture index on the MP firstly. The feature space model and five machine learning models were constructed to make the map based on Google Earth Engine and Landsat data. The spatiotemporal distribution of desertification were further analyzed, and the dominant drivers of desertification distribution and evolution were identified using Geodetector model. The results indicate that the potential occurrence area of desertification accounted for 83.88 % of the total land area. The gradient boosted tree model for desertification assessment has the best performance with the highest overall accuracy of 88.18 %. The year 2010 marked a pivotal transition from land degradation to land restoration in the MP. Between 2010 and 2020, desertified land continued to deteriorate extensively in the southern Mongolia, while Inner Mongolia, China, essentially entered a full recovery phase. Precipitation and land use emerged as the primary drivers of the spatial distribution of desertification on the Mongolian Plateau and Mongolia, with potential evapotranspiration and precipitation influencing the distribution of desertification in Inner Mongolia, China. Land use change was the primary driver of desertification evolution on the MP and Mongolia. This study constructs an indicator system and methodology suitable for desertification monitoring on the MP, addresses the lack of refined desertification data over a long time series, and provides scientific reference for decision-making support in combating desertification in this region, and other large arid and semi-arid areas in the world.

Temporal and spatial variability of turbidity in a highly productive and turbid shallow lake (Chascomús, Argentina) using a long time-series of Landsat and Sentinel-2 data

Temporal and spatial variability of turbidity in a highly productive and turbid shallow lake (Chascomús, Argentina) using a long time-series of Landsat and Sentinel-2 data

Landscape metrics of the Brusy Commune before and after wind-storm: an assessment of the extent of changes based on Landsat-8 data

Monitoring the change in land cover in disaster-affected areas, such as forests, has become a conventional forest management practice, particularly in protected areas. Most change detection and fragmentation studies rely on single-dated satellite images even while investigating changes over a long temporal span. This study aims to move a step further to compare fragmentation before and after a derecho event that occurred in August 2017 using 23 Landsat-8 images of Brusy Commune within the Tuchola Forest Biosphere Reserve. The supervised classification was carried out in the Google Earth Engine using the machine learning algorithm of random forests within the summer months of 2017 and 2018. The high overall accuracy of 0.92 was obtained for the two images which were then analysed with landscape metrics such as mean patch size, number of patches, total edge and edge density using Patch Analyst. These landscape metrics facilitated the characterisation of landscape fragmentation at both the class and landscape levels. Shannon’s Diversity Index was employed to assess heterogeneity across the landscape. The findings indicate significant fragmentation, particularly in the forest and pasture classes, with overall low diversity. This study underscores the potential for future research to employ advanced machine learning techniques and non-parametric classifiers, such as neural networks, to enhance the prediction of fragmentation across various spatial scales.

Comparison of machine and deep learning algorithms using Google Earth Engine and Python for land classifications

Classifying land use and land cover (LULC) is essential for various environmental monitoring and geospatial analysis applications. This research focuses on land classification in District Sukkur, Pakistan, employing the comparison between machine and deep learning models. Three satellite indices, namely, NDVI, MNDWI, and NDBI, were derived from Landsat-8 data and utilized to classify four primary categories: Built-up Area, Water Bodies, Barren Land, and Vegetation. The main objective of this study is to evaluate and compare the effectiveness of comparison of machine and deep learning models. The machine learning models including Random Forest achieved an overall accuracy of 91.3% and a Kappa coefficient of 0.90. It accurately classified 2.7% of the area as Built-up Area, 1.9% as Water Bodies, 54.8% as Barren Land, and 40.4% as Vegetation. While slightly less accurate, Decision Tree model provided reliable classifications. Deep learning models showed significant accuracy, of Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN). The CNN model achieved an impressive overall accuracy of 97.3%, excelling in classifying Water Bodies with User and Producer Accuracy exceeding 99%. The RNN model, with an overall accuracy of 96.2%, demonstrated strong performance in categorizing Vegetation. These findings offer valuable insights into the potential applications of machine learning and deep learning models for perfect land classifications, with implications for environmental monitoring management and geospatial analysis. The rigorous validation and comparative analysis of these models contribute to advancing remote sensing techniques and their utilization in land classification tasks. This research presents a significant contribution to the field and underscores the importance of precise land classification in the context of sustainable land management and environmental conservation.

Analyzing vegetation health dynamics across seasons and regions through NDVI and climatic variables

This study assesses the relationships between vegetation dynamics and climatic variations in Pakistan from 2000 to 2023. Employing high-resolution Landsat data for Normalized Difference Vegetation Index (NDVI) assessments, integrated with climate variables from CHIRPS and ERA5 datasets, our approach leverages Google Earth Engine (GEE) for efficient processing. It combines statistical methodologies, including linear regression, Mann–Kendall trend tests, Sen's slope estimator, partial correlation, and cross wavelet transform analyses. The findings highlight significant spatial and temporal variations in NDVI, with an annual increase averaging 0.00197 per year (p < 0.0001). This positive trend is coupled with an increase in precipitation by 0.4801 mm/year (p = 0.0016). In contrast, our analysis recorded a slight decrease in temperature (− 0.01011 °C/year, p < 0.05) and a reduction in solar radiation (− 0.27526 W/m2/year, p < 0.05). Notably, cross-wavelet transform analysis underscored significant coherence between NDVI and climatic factors, revealing periods of synchronized fluctuations and distinct lagged relationships. This analysis particularly highlighted precipitation as a primary driver of vegetation growth, illustrating its crucial impact across various Pakistani regions. Moreover, the analysis revealed distinct seasonal patterns, indicating that vegetation health is most responsive during the monsoon season, correlating strongly with peaks in seasonal precipitation. Our investigation has revealed Pakistan's complex association between vegetation health and climatic factors, which varies across different regions. Through cross-wavelet analysis, we have identified distinct coherence and phase relationships that highlight the critical influence of climatic drivers on vegetation patterns. These insights are crucial for developing regional climate adaptation strategies and informing sustainable agricultural and environmental management practices in the face of ongoing climatic changes.

Monitoring Long-Term Land Cover Change in Central Yakutia Using Sparse Time Series Landsat Data

Recently, as global climate change and local disturbances such as wildfires continue, long- and short-term changes in the high-latitude vegetation systems have been observed in various studies. Although remote sensing technology using optical satellites has been widely used in understanding vegetation dynamics in high-latitude areas, there has been limited understanding of various landscape changes at different spatiotemporal scales, their mutual relationships, and overall long-term landscape changes. The objective of this study is to devise a change monitoring strategy that can effectively observe landscape changes at different spatiotemporal scales in the boreal ecosystems from temporally sparse time series remote sensing data. We presented a new post-classification-based change analysis scheme and applied it to time series Landsat data for the central Yakutian study area. Spectral variability between time series data has been a major problem in the analysis of changes that make it difficult to distinguish long- and short-term land cover changes from seasonal growth activities. To address this issue effectively, two ideas in the time series classification, such as the stepwise classification and the lateral stacking strategies were implemented in the classification process. The proposed classification results showed consistently higher overall accuracies of more than 90% obtained in all classes throughout the study period. The temporal classification results revealed the distinct spatial and temporal patterns of the land cover changes in central Yakutia. The spatiotemporal distribution of the short-term class illustrated that the ecosystem disturbance caused by fire could be affected by local thermal and hydrological conditions of the active layer as well as climatic conditions. On the other hand, the long-term class changes revealed land cover trajectories that could not be explained by monotonic increase or decrease. To characterize the long-term land cover change patterns, we applied a piecewise linear model with two line segments to areal class changes. During the former half of the study period, which corresponds to the 2000s, the areal expansion of lakes on the eastern Lena River terrace was the dominant feature of the land cover change. On the other hand, the land cover changes in the latter half of the study period, which corresponds to the 2010s, exhibited that lake area decreased, particularly in the thermokarst lowlands close to the Lena and Aldan rivers. In this area, significant forest decline can also be identified during the 2010s.

Broad-area-search of new construction using time series analysis of Landsat and Sentinel-2 data

New construction activities can alter surface albedo and structure, which then affect surface temperature and roughness, and hence have a significant impact on urban climate. Construction activity is also an important indicator of human development and movement and is of high interest to the intelligence community. A new approach for Broad-Area-Search of New Construction activities (BASC) by combining time series analysis and rule-based filters using Landsat data was developed and tested in five selected cities (Boston, Shanghai, São Paulo, Dubai, and Ho Chi Minh City). The algorithm transforms Landsat images into fractions of a set of four endmembers using Linear Spectral Mixture Analysis (LSMA) and then applies the Continuous Change Detection and Classification (CCDC) algorithm for change detection. A set of rule-based filters and spatial processing was then applied to narrow the search to changes related to construction activities. Overall, BASC reached a recall of 0.83, a precision of 0.58, and an F1-Score of 0.68. Among the five cities, Dubai had the highest recall of 1.0 and the highest F1-score of 0.75, while Boston had the highest precision of 0.63. BASC performed worst in Shanghai with an F1-Score of 0.6, mainly due to it having the lowest recall of 0.62, while São Paulo has the lowest precision of 0.5. Common sources of omission errors include low-density, redevelopment, and small sites, while common commission errors include roofing, land clearing, water level changes, and re-surfacing projects. For comparison, BASC using Sentinel-2 Top-of-Atmosphere (TOA) Reflectance data recorded an overall F1-Score of 0.63, but with higher recall and lower precision. Integration of Sentinel-2 Surface Reflectance and Sentinel-1 SAR data has the potential to further improve the performance of BASC. The new algorithm provided a method for routine monitoring of construction activities over large areas. The result of such monitoring can be used as a baseline to narrow down the candidate targets of construction activities, where very high-resolution imagery can then be requested to perform further examination.

Predicting post-fire forest recovery using the 3-PG model with bi-temporal Landsat imagery in high-severity burned areas of Great Xing’an Mountain

High-severity fire altered the landscape pattern and destroyed forest resilience in the Great Xing’an Mountain. Predicting the spatiotemporal patterns of forest recovery in high-severity burned areas is critical for formulating effective post-fire management measures. Previous studies attempted to address this topic by fitting the recovery trajectory based on time-series remote sensing vegetation indices. However, these methods typically require long-term time series of satellite images for prediction and have sometimes been demonstrated as an unrealistic representation of the post-fire forest patterns due to the saturation issues of vegetation indices. In this study, we proposed a novel approach to predict post-fire forest recovery patterns by coupling a process-based Physiological Principles in Predicting Growth (3-PG) model with bi-temporal satellite imagery. First, based on tree core data, we calibrated the physiological parameters of the 3-PG model using a Model-Independent Parameter Estimation (PEST) model. Second, we used vegetation indices derived from bi-temporal Landsat data to estimate site-specific parameters of the 3-PG model and extrapolated the meteorological factors by a Mountain Microclimate Simulation Model (MTCLIM). Third, we used these variables as input parameters to drive the 3-PG model, resulting in the simulation of the post-fire 50-year regeneration dynamic of the leaf area index (LAI). The calibrated 3-PG model was validated by the independent sites, providing good estimates of DBH and biomass components with R2 of greater than 0.994 and RRMSE of less than 15.7 %. The 3-PG model predicted LAI maps were evaluated by referenced LAI retrieved from Landsat data, showing a high correlation with reference but the accuracy decreased with the prediction time increased. The model predicted forest recovery to pre-fire LAI within 32 years where the longest recovery interval was observed in the low seedling density and soil fertility. This study shows great potential for reasonable prediction and assessment of forest recovery after the fire occurrence. By combining more simulated scenarios, our approach also facilitates the exploration of the impact of artificial measures and climate on post-forest recovery.

The growth and emergence of potentially dangerous glacier lakes in Astore Basin, Western Himalaya during 1993–2021

The recent retreat of glaciers in High Mountain Asia is a major issue for downstream communities. Similarly, glaciers in the Astore Basin are melting, causing glacial lakes to expand faster, new lakes to form, and increasing the risk of glacial lakes outburst floods (GLOFs). The present study uses Landsat data from 1993 to 2021 to explore seasonal and decadal changes in glacier lakes, which are validated using an in situ differential Global Positioning System (dGPS). During the ablation period (June - October) of 2021, we observed a five-fold increase (18 to 100) in the number of glacier lakes, as well as a six-fold increase (0.62 to 3.86 km2) in lakes larger than 0.01 km2. Over the last decade, from 2011 to 2020, the number of potentially dangerous glacial lakes (PDGLs) has doubled. GLOF risk must be reduced through continual monitoring of these lakes. Prioritizing the deployment of GLOF monitoring and early warning systems, as well as sustainable water management practices, is critical for mitigation and adaptation measures in mountainous regions.

Assessing the trends and drivers of agricultural land use change in the Ramsar County, northern Iran

Agricultural land use change (ALUC) is a primary driver of global environmental change, with implications for food security and the environment. Ramsar County in the north of Iran has continuously undergone many changes due to the tourism characteristics of the region. The present study demonstrates the ALUC trends from 1992 to 2022 and the main drivers affecting ALUC in Ramsar County, northern Iran, using household surveys and Landsat data. Six land use classifications have been made for the region by image processing. Images have been categorized using the maximum likelihood classifier. Likewise, utilizing a questionnaire from 200 randomly selected farmers, in-person interviews were conducted. The kappa coefficient and overall accuracy for 1992, and 2022 were 85.91%, 88.86%, and 88.38%, 90.83%, respectively. It was observed that from 1992 to 2022, orchard (12%), open land (4.8%), and built-up areas (160.5%), increased respectively. In contrast, forest (−12.1%), water (−82.6%), and agri-land (−14.8%), declined respectively. Additionally, the analysis of the household survey indicates the high cost of production, the lack of a suitable market and price fluctuations of products, the aging of farmers, the increase in the population of non-native residents, weak policies to support farmers, lack of coordination between relevant institutions, the beautiful nature and tourism of the region, and climate changes and drought in other parts of the country were the most important drivers affecting ALUC in the region. Overall, this study can provide significant insights to decision-makers at the beginning of strategic planning to control the ALUC in the region.

A multi-source change detection algorithm supporting user customization and near real-time deforestation detections

The abundance of free and accessible satellite data has revolutionized our ability to study deforestation with remote sensing. Recent advances have enabled us to monitor deforestation in near real-time (NRT), and a number of operational NRT alert systems using both optical and synthetic aperture radar (SAR) data have been developed. Yet despite their success, there are three primary issues with current systems. First, alerts often require multiple observations to confirm deforestation, which does not help enforcement if deforestation events are short-lived. Second, the methods can be computationally intensive and often focus on presenting binary detection alerts. Third, results are often reported as static accuracies or temporal lags, with relatively little discussion about how either could be improved for different use-cases. We therefore propose a novel, multisource NRT monitoring algorithm that addresses these issues by quickly achieving high-accuracy detections, using simple data harmonization to combine data sources while yielding probabilities of deforestation, and can be customized to meet management goals where users can explicitly change the algorithm to prioritize latency or accuracy in their detections based on their chosen data inputs. To demonstrate the algorithm's potential, we used combined Landsat-8, Sentinel-2, and Sentinel-1 SAR data in northern Myanmar to monitor areas with contrasting forest types. When testing all possible parameterizations, we achieved comparable results to previous studies, with median false positive rates and false negative rates under 10% and 25%, respectively, and median F1 scores consistently above 75% across the sensor combinations. Median post-data acquisition detection times were faster than other studies and ranged from 2 to 8 days, though the majority were between 2 and 5 days. In attempting to optimize the algorithm for our demonstrated region, we discovered there was no optimal parameterization that simultaneously provided the best accuracy and detection. In addition, we found that including Sentinel-1 data did not improve latency or accuracy, most likely due to relatively high levels of noise in our monsoonal study system that experienced high seasonal precipitation variation. We finally chose an example parameterization to create daily landscape maps of deforestation at 10 m resolution in two regions within the study area (∼60,000 ha). We assert this new, customizable method with the quantification of accuracy and latency trade-offs represents a clear advance for NRT change detection and can improve current operational systems.

Spatiotemporal coupling analysis of surface water and groundwater in Tibet based on multi-source sensors

Abstract. Lakes and groundwater are two crucial components of the global terrestrial water cycle, collectively forming a vital network for Earth's water resources. However, in the Tibetan region, the spatiotemporal relationship between lakes and groundwater and their impact on the local hydrological cycle remain inadequately understood. Satellite remote sensing serves as an effective observational tool, enabling comprehensive investigation and analysis of surface lakes and groundwater in Tibet with high spatial resolution. Therefore, this study integrates Landsat and Cryosat-2 satellite data to examine the spatiotemporal patterns of surface lake extents and water levels in Tibet. Additionally, combining Gravity Recovery and Climate Experiment (GRACE) satellite observations with the Global Land Data Assimilation System (GLDAS) model data, we quantitatively analyze the spatiotemporal variations in groundwater storage and its correlation with lake water. The results indicate that: 1) Rivers and lakes in Tibet are mainly located in the central and northwest regions, displaying noticeable intra-annual variations; 2) Substantial lagged relationships exist between groundwater storage and lake water levels and areas, revealing that lakes contribute significantly to groundwater replenishment, especially in the Ngari prefecture and Lhokha prefecture. This study comprehensively utilizes multi-source remote sensing data to dynamically monitor surface and groundwater in Tibet, providing robust support for a better understanding of the interaction between groundwater and surface water.

Mapping the dynamics of intensive forage acreage during 2008–2022 in Google Earth Engine using time series Landsat images and a phenology-based algorithm

The increasing demand for livestock feed in China has led to a remarkable increase in forage acreage, particularly in forage fields under intensive agricultural management. However, there are no maps currently available to demonstrate and document the spatial and temporal patterns of the continuous expansion of intensive forage. Here we proposed a pixel- and phenology-based approach to map intensive forage by utilizing time series Landsat images and the Google Earth Engine platform. Given the challenge of accurately identifying intensive forage fields in diverse regions using fixed multi-temporal Landsat data, along with the distinct phenological characteristics of intensive forage fields that undergo multiple annual harvests, we proposed an innovative data mining approach based on dynamic monthly composite images (DMCI). This approach aims to enhance the algorithm’s efficiency and precision in identifying intensive forage. As a pilot study, we analyzed all available Landsat images (1732 scenes) from 2008 to 2022 in a county located in the farming-pastoral ecotone of northern China and tracked the historical dynamics of intensive forage expansion over five epochs at three-year intervals. The accuracy assessment of the intensive forage maps for the five epochs using validation sample points showed that the overall accuracy and Mathews correlation coefficient ranged from 97.1 % to 98.9 % and 0.93 to 0.98 respectively. The intensive forage acreage in the study area exhibited a dramatic expansion from 2008 to 2022, particularly after the 2012s. This study demonstrates the potential of our DMCI- and phenology-based algorithm and time series Landsat images for tracking the dynamics of intensive forage acreage at 30-m resolution in temperate single-cropping regions.

Deep Siamese Network for annual change detection in Beijing using Landsat satellite data

The detection of spatiotemporal changes in land use/land cover (LULC-SC) plays a paramount role in the analysis of smart cities, it can describe complex urban distribution, functions, and patterns. Chinas capital city has been developing rapidly in the past two decades, however, there are only few long-term studies on an annual scale of LULC-SC. To fill this research gap, we propose a remote sensing parallel framework for the detection of LULC-SC based on the combination of the Deep Siamese Network and long time series, which focuses on the spatial semantic information at the object level. A Landsat time series from 2002 to 2022 serves as input satellite data. First, we focus on building graph constructions at the object level and then use an autonomously constructed deeper-feature graph convolutional network (DF-GCN) to mine deeper features, spatial semantic, and relationships at the object level. Finally, the Siamese Network recognizes the changes in the spatial semantic tensors of long time series of Landsat images and quickly maps LULC-SC. The results prove that the proposed spatiotemporal change detection framework is effective in LULC-SC in Beijing. Compared with other networks, the optimal accuracy of semantic mining based on DF-GCN can reach about 90%. Over the past two decades, the LULC-SC of Beijing has changed in a complex way, with urbanization occurring primarily through the replacement of farmland. Consequently, the proposed framework can generate accurate LULC change maps at high temporal frequencies, which can contribute to a better comprehension of sustainable urban development and planning.

A longitudinal analysis of soil salinity changes using remotely sensed imageries

Soil salinization threatens agricultural productivity, leading to desertification and land degradation. Given the challenges of conducting labor-intensive and expensive field studies and laboratory analyses on a large scale, recent efforts have focused on leveraging remote sensing techniques to study soil salinity. This study assesses the importance of soil salinity indices’ derived from remotely sensed imagery. Indices derived from Landsat 8 (L8) and Sentinel 2 (S2) imagery are used in Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Decision Tree (DT), and Support Vector Machine (SVR) are associated with the electrical (EC) conductivity of 280 soil samples across 24,000 hectares in Northeast Iran. The results indicated that the DT is the best-performing method (RMSE = 12.25, MAE = 2.15, R2 = 0.85 using L8 data and RMSE = 10.9, MAE = 2.12, and R2 = 0.86 using S2 data). Also, the results showed that Multi-resolution Valley Bottom Flatness (MrVBF), moisture index, Topographic Wetness Index (TWI), and Topographic Position Indicator (TPI) are the most important salinity indices. Subsequently, a time series analysis indicated a reduction in salinity and sodium levels in regions with installed drainage networks, underscoring the effectiveness of the drainage system. These findings can assist decision-making about land use and conservation efforts, particularly in regions with high soil salinity.