Google Earth Engine Platform Research Articles

Spatiotemporal changes and driving factors of ecological environmental quality in the Yongding-Luan River Basin based on RSEI

The ecological environmental quality (EEQ) of the Yongding-Luan River Basin (YLRB) is pivotal to the ecological security of the Beijing-Tianjin-Hebei (JJJ) region's core area. Evaluating the EEQ and analyzing its changes are essential for regional ecological management. However, long-term ecological changes in the YLRB remain uncovered. In this study, we constructed a seamless Remote Sensing Ecological Index (RSEI) for the YLRB from 1986 to 2022 using time-series Landsat imagery on the Google Earth Engine (GEE) platform. The Sen + Mann-Kendall method was employed to analyze the spatiotemporal trends of EEQ, and the Geodetector was used to quantitatively assess the driving factors and their interactions. The results show that: 1) The mean RSEI of the YLRB increased from 0.486 in 1986 to 0.532 in 2022, marking a 9.5% rise and indicating a fluctuating upward trend. 2) The EEQ of the YLRB experienced three distinct phases: improvement, deterioration, and re-improvement. Improvements were predominantly in the western YLRB, while deterioration was mainly in the northern Xilinguole region and the southern urban expansion areas of Beijing, Langfang, Tianjin, and Tangshan. 3) The driving factor detection indicates that land use type and annual average precipitation are the primary driving factors of RSEI change in the YLRB. Furthermore, their interaction results in a significant effect on RSEI, with a maximum of 0.691. These findings align with the historical urban expansion in the YLRB and the environmental policies implemented by the Chinese government. The ecological evolution and driving factors identified in this study offer a scientific basis for regional ecological decision-making and management.

The Dynamics of Air Pollution in the Southwestern Part of the Caspian Sea Basin (Based on the Analysis of Sentinel-5 Satellite Data Utilizing the Google Earth Engine Cloud-Computing Platform)

The Caspian region represents a complex and unique system of terrestrial, coastal, and aquatic environments, marked by an exceptional landscape and biological diversity. This diversity, however, is increasingly threatened by substantial anthropogenic pressures. One notable impact of this human influence is the rising concentration of pollutants atypical for the atmosphere. Advances in science and technology now make it possible to detect certain atmospheric pollutants using remote Earth observation techniques, specifically through data from the Sentinel-5 satellite, which provides continuous insights into atmospheric contamination. This article investigates the dynamics of atmospheric pollution in the southwestern part of the Caspian Sea basin using Sentinel-5P satellite data and the cloud-computing capabilities of the Google Earth Engine (GEE) platform. The study encompasses an analysis of concentrations of seven key pollutants: nitrogen dioxide (NO2), formaldehyde (HCHO), carbon monoxide (CO), ozone (O3), sulfur dioxide (SO2), methane (CH4), and the Aerosol Index (AI). Spatial and temporal variations in pollution fields were examined for the Caspian region and the basins of the seven rivers (key areas) flowing into the Caspian Sea: Sunzha, Sulak, Ulluchay, Karachay, Atachay, Haraz, and Gorgan. The research methodology is based on the use of data from the Sentinel-5 satellite, SRTM DEM data on absolute elevations, surface temperature data, and population density data. Data processing is performed using the Google Earth Engine cloud-computing platform and the ArcGIS software suite. The main aim of this study is to evaluate the spatiotemporal variability of pollutant concentration fields in these regions from 2018 to 2023 and to identify the primary factors influencing pollution distribution. The study’s findings reveal that the Heraz and Gorgan River basins have the highest concentrations of nitrogen dioxide and Aerosol Index levels, marking these basins as the most vulnerable to atmospheric pollution among those assessed. Additionally, the Gorgan basin exhibited elevated carbon monoxide levels, while the highest ozone concentrations were detected in the Sunzha basin. Our temporal analysis demonstrated a substantial influence of the COVID-19 pandemic on pollutant dispersion patterns. Our correlation analysis identified absolute elevation as a key factor affecting pollutant distribution, particularly for carbon monoxide, ozone, and aerosol indices. Population density showed the strongest correlation with nitrogen dioxide distribution. Other pollutants exhibited more complex distribution patterns, influenced by diverse mechanisms associated with local emission sources and atmospheric dynamics.

Using Google Earth Engine open-source code for land surface temperature estimation from Landsat data in Chuong My district, Hanoi city, Vietnam

Land surface temperature (LST) in association with urbanization has significantly influenced on local climates and terrestrial surface processes. By using multi-temporal Landsat data in the GEE platform, the study has found that there have been changes in land covers due to the main drivers of industrial development, urban and built-up expansion. This study estimated land surface temperature (LST) in Chuong My district from Landsat 5, 7, 8 and 9 thermal infrared sensors, using different surface emissivity sources. The Google Earth Engine (GEE), an advanced earth science data and analysis platform, offers the estimation of LST products, covering the time period from 2000 to 2024 in study site. RF algorithm for land cover classification and mapping is suggested to be applied in Chuong My district, with overall accuracy of land cover mapping in 2024 being 91.9% with Kappa coefficient of 0.89. The period 2000-2010 showed that 327.0 ha of land were converted to land for industrial development, urban and residential land, while the period 2010-2024 continued to witness a continued conversion of land to other land use purposes, with a decrease in agricultural land area (65.7 ha) and an increase in industrial land use (65.2 ha), but the rate of conversion was much lower than the previous period (2000-2024). The periods of 2000-2010 and 2010-2024 showed that there have been significant changes in land surface temperature. Most of the surface temperature in the entire region increased by over 20C, especially in the period 2003-2015 when a total area of 1424 ha had the temperature increased by above 30C. This is a consequence of the process of urbanization and industrialization, which began to take place in 2008, along with the process of changing the land cover status of the study area. Further studies are needed to better understand the thermal conductivity of surface materials and to plan how to reduce LST in such areas.

Machine learning-based crop type detection and vegetation mon-itoring using Sentinel-1 SAR and Landsat 8

This study explores the use of machine learning for crop type detection and vegetation monitoring in arid and semi-arid regions, specifically in Biskra and Khenchela, Algeria. Traditional field-based agricultural monitoring is labor-intensive and limited in spatial coverage, prompting the need for more efficient and scalable methods. The research integrates Synthetic Aperture Radar (SAR) data from Sentinel-1 and optical imagery from Landsat 8 to enhance the classification of different crop types and track vegetation dynamics. Key metrics such as radar backscatter and Normalized Difference Vegetation Index (NDVI) are employed to distinguish between irrigated and rainfed crops. The Random Forest algorithm is utilized for classification, with training data derived from field surveys, high-resolution satellite imagery, and existing land cover maps. The Google Earth Engine (GEE) platform facilitates large-scale data processing, providing real-time agricultural analysis. Over the 12-month study period (January to December 2020), the integrated approach demonstrated high accuracy, achieving over 85% precision in classifying key crops such as date palms, cereals, and vegetables. Results indicate that SAR and NDVI data provide complementary insights into vegetation health, phenology, and agricultural practices in different climatic zones. This methodology offers a promising solution for improving agricultural monitoring and resource management in arid and semi-arid regions, enhancing food security and optimizing water usage. The study’s findings also underscore the value of satellite-based remote sensing for large-scale, real-time agricultural analysis.

Google Earth Engine-based Mangrove Mapping and Change Detections for Sustainable Development in Tien Yen District, Quang Ninh Province, Vietnam

Abstract. Vietnam secures a place among the top countries that possess the largest mangrove areas worldwide. The mangrove forests are mainly found in the Northern, Northeast, Central and Southern Delta, providing goods to habitants, and make significant help mitigate global climate change. Despite this, mangroves are severely threatened because of extensive deforestation in Vietnam. Recent advances have utilized remotely sensed imagery to present the spatial and temporal distribution of mangroves. Nevertheless, the approach is limited by imagery availability and computing resources, and difficult to share within the community. Therefore, a shareable Web-based tool that reinforces coastal managers to monitor the changes in mangroves is needed. Recently, Google Earth Engine (GEE) is a cloud-based geospatial analysis platform, which allows users to freely exploit the availability of satellite imagery and harness their computing capacity. This research aims to use GEE to detect mangrove changes, a case study in Tien Yen district, Quang Ninh, over a period from 2010 to 2020. Four supervised classification algorithms, including Random Forest (RF), Support Vector Machine (SVM), Naïve Bayes classifier, and Classification and Regression Trees (CART) have been implemented on GEE platform to select the best algorithm to produce spatial-temporal mangrove maps, then change detection of mangroves is performed. The results showed that RF demonstrated the highest accuracy with overall accuracy and kappa values of 91.04% and 0.75 respectively. The mangrove areas statistically reported by the GEE-based platform are in line with the governmental statistic report annually, marking the capabilities of GEE in natural resource management.

Comparative Study of Random Forest and Support Vector Machine for Land Cover Classification and Post-Wildfire Change Detection

The land use land cover (LULC) map is extensively employed for different purposes. Machine learning (ML) algorithms applied in remote sensing (RS) data have been proven effective in image classification, object detection, and semantic segmentation. Previous studies have shown that random forest (RF) and support vector machine (SVM) consistently achieve high accuracy for land classification. Considering the important role of Portugal’s Serra da Estrela Natural Park (PNSE) in biodiversity and nature conversation at an international scale, the availability of timely data on the PNSE for emergency evaluation and periodic assessment is crucial. In this study, the application of RF and SVM classifiers, and object-based (OBIA) and pixel-based (PBIA) approaches, with Sentinel-2A imagery was evaluated using Google Earth Engine (GEE) platform for the land cover classification of a burnt area in the PNSE. This aimed to detect the land cover change and closely observe the burnt area and vegetation recovery after the 2022 wildfire. The combination of RF and OBIA achieved the highest accuracy in all evaluation metrics. At the same time, a comparison with the Normalized Difference Vegetation Index (NDVI) map and Conjunctural Land Occupation Map (COSc) of 2023 year indicated that the SVM and PBIA map resembled the maps better.

An improved optical trapezoid model to identify wetlands and irrigation water ponds in the arid irrigation district using Google Earth Engine

ABSTRACT In arid regions, wetlands support ecosystem with high biodiversity and productivity. Meanwhile, irrigation during the fallow period often results in the temporary formation of water ponds in the cropland, which plays a critical role in agricultural productivity. The prompt identification of wetlands and irrigation water ponds in irrigation districts is important for the protection of local wetland ecosystem and the effective management of irrigation water. Utilizing the OPtical TRApezoid Model (OPTRAM) for soil moisture estimation, we have proposed an improved version of OPTRAM aimed to identify wetlands and irrigation water ponds in irrigation district using Sentinel-2 data through Google Earth Engine platform. Different criteria were used for the identification of wetland and irrigation water pond, given the greater diversity associated with wetlands. Parameters for the wet edge determined from OPTRAM were used to identify wetlands, while an additional threshold was added to the model and calibrated in accordance with the irrigated area to identify irrigation water ponds. The improved OPTRAM was applied in Hetao Irrigation District (HID) of Northwest China from 2016 to 2023, where autumn irrigation applied in late autumn after crop harvesting was considered. The identified distributions of wetland and autumn irrigation were validated with observations from field survey and statistical data, and were also compared with other remote sensing products. Results show that the proposed model is effective in identifying both wetlands and irrigation ponds. Considering the wetlands, the distances from identified boundaries to the actual boundaries obtained from field surveys were calculated, revealing the mean absolute error of 6.99 m, alongside a validation accuracy ratio of 0.94 for points located within wetlands. For autumn irrigation extent, the overall accuracy is 0.90 based on the observations from field survey, with mean absolute relative errors for irrigated areas across four sub-irrigation districts recorded as 20.55%, 8.10%, 12.83%, and 11.38%, respectively, when compared with statistical data. The majority of wetlands within HID are small in size and scattered, with different trends observed in the wetland areas of sub-irrigation districts and Wuliangsuhai Lake. Autumn irrigated croplands are mainly concentrated in Jiefangzha sub-irrigation district while being scattered across other sub-irrigation districts, depicting an overall decreasing trend in the autumn irrigated area. In summary, the proposed model performed well in identifying both wetlands and irrigation ponds, and can offer valuable support for the conservation of wetland ecosystem and irrigation management.

Spatiotemporal trends and drivers of forest cover change in Metekel Zone forest areas, Northwest Ethiopia.

The spatiotemporal dynamics of forest cover are essential for understanding the patterns and processes of forest change over time and space. This research focused on the spatiotemporal trends and drivers of forest cover change in the Metekel Zone of Northwest Ethiopia. Landsat 5, Landsat 7, and Landsat 8 imagery, covering the period from 1986 to 2019, were used for land use/cover classification. Land use/cover classification was performed using random forest (RF) and support vector machine (SVM) algorithms in the Google Earth Engine (GEE) platform, with training samples obtained through visual image interpretation. Spectral indices, such as the normalized difference vegetation index, soil-adjusted vegetation index, leaf area index, and normalized difference water index, were analyzed to examine forest cover dynamics over time. In addition, key informant interviews (KIIs) and focus group discussions (FGDs) were conducted. Findings revealed that forest cover decreased significantly from 51.37% in 1986 to 17.20% in 2019, driven largely by human activities such as agricultural expansion, increased demand for firewood, and urban expansion. Findings from spectral indices further corroborated the finding that forest cover in the study region (mainly in the southwestern part) substantially decreased from 1986 to 2019. Concerning forest depletion, the lack of local community awareness has become a key challenge. This problem is attributed to communities prioritizing immediate needs such as fuel and land for agriculture over long-term forest conservation. To combat ongoing deforestation, the Metekel Zone Administration, in collaboration with the land administration office and other stakeholders, revisited and strengthened existing forest policies and control systems. It is also suggested that community awareness, chiefly among youth, should be enhanced through the strategic expansion of formal and nonformal educational initiatives, which empower the youth as agents of change and promote the dissemination of knowledge throughout the community.

Rapid lithological mapping using multi-source remote sensing data fusion and automatic sample generation strategy

ABSTRACT The advancement of remote sensing technology aids geologists in obtaining lithological maps more quickly, comprehensively, and accurately. However, key challenges in lithological mapping include the limited spectral information from individual sensors and the difficulties in visually interpreting lithological samples. In this study, we integrated 241 scenes of optical data and 106 scenes of radar data on the Google Earth Engine (GEE) platform, proposing a rapid lithological identification framework that combines an automatic lithological sample data generation strategy with multi-source data. Using various machine learning algorithms, we evaluated the classification capabilities of heterogeneous predictive factors, feature optimization algorithms, and object-based algorithms. Results indicate that: (1) Combining optical and radar data improves prediction accuracy, with terrain data further enhancing mapping capabilities; (2) Terrain factors contribute most to classification, but SWIR and TIR bands of optical data are critical for lithological identification; (3) The feature optimization algorithm reduces feature redundancy and efficiency issues from multi-source data, achieving 96.51% accuracy with the optimal feature model, an improvement of 0.1%−2.02% over original features; (4) Object-based algorithms show significant potential in mapping areas with large rock outcrops. This study offers new insights for medium- to large-scale lithological maps and provides essential data support for geological work.

Sand and Dust Storm Risk Assessment in Arid Central Asia: Implications for the Environment, Society, and Agriculture

AbstractThis study aimed to assess sand and dust storm (SDS) risks in arid Central Asia during 2001–2021 from a multisectoral (environment, society, and agriculture) and comprehensive perspective on the Google Earth Engine (GEE) platform. The results show that the areas with moderate or greater SDS risk accounted for 18.75% of the total area of arid Central Asia. The high SDS risk areas are mainly concentrated in the oases around the desert and are most widely distributed in spring and summer. The SDS risk in the oasis area of southern Xinjiang increased significantly, while the SDS risk in the northeastern Aral Sea region and the Kazakh hilly region decreased significantly over the 21 years. Khwarazm of Uzbekistan, located in the Amu Darya River Delta, is the administrative district with the highest comprehensive risk of sandstorms, and the Balkan State of Turkmenistan and Kashi City and Zepu County in China are the administrative districts with the highest multisectoral risk of sandstorms. The results of this study provide a complete picture of SDS risks in the arid Central Asia region and will provide some guidance to policymakers and local authorities in SDS risk mitigation.

Volcán de Fuego (Guatemala) monitoring with the Normalized Hotspot Indices (NHI) tool

Abstract. Volcán de Fuego is a stratovolcano located in Guatemala among the most active in the world. In this work, we investigate its eruptive activity from space by means of the Normalized Hotspot Indices (NHI) algorithm. The latter runs operationally under the Google Earth Engine (GEE) platform providing information on active volcanoes at global scale by means of daytime Sentinel 2 MSI (Multispectral Instrument) and Landsat 8/9 OLI/OLI2 (Operational Land Imager) data. In study, we present the results of a time series analysis performed by investigating 40 years of Landsat observations through the NHI algorithm. Results show that during the major periods of thermal activity (e.g., 2000−2003; 2012−2013; 2015−2018) the Volcán de Fuego generated extended lava flows, which were well identified and mapped by satellite. In addition, periods of reduced thermal activity (e.g., 2022−May 2024) marked by NHI were in good agreement with information provided by independent sources. The accurate localization and mapping of high-temperature features, and the characterization of different eruptive phases, demonstrate that the NHI algorithm, through the GEE-App available online (https://sites.google.com/view/nhi-tool/nhi-tool-for-volcanoes), may support scientists also in the monitoring of frequently active volcanoes such as the Volcán de Fuego, providing information about changes of thermal volcanic activity that could precede future and more dangerous eruptions.

A Novel Phenology-Based Index for Plastic-Mulched Farmland Extraction and Its Application in a Typical Agricultural Region of China Using Sentinel-2 Imagery and Google Earth Engine

Plastic-mulching technology has a crucial role to play in modern agriculture by optimizing crop growth environments and enhancing yields. Accurately detecting and mapping the distribution of plastic-mulched farmlands (PMFs) is essential for improving both agricultural management and production efficiency. By analyzing the temporal spectral characteristics of PMFs and crop phenological information, we developed a phenology-based plastic-mulched farmland index (PPMFI). This index, when combined with Sentinel-2 imagery and an automated high-precision extraction process via the Google Earth Engine platform, effectively distinguishes PMFs from other land cover types, especially in complex agricultural landscapes. Validation across areas varying in their background complexity and PMF coverage demonstrated that the proposed PPMFI consistently achieves an overall accuracy rate that exceeds 90%, showcasing its robust performance and significantly outperforming other comparative extraction methods. Applying the PPMFI to the Yudong agricultural region of Henan Province, China, further confirmed its capability for large-scale PMF monitoring, thereby offering critical technical support for sustainable agricultural management and environmental protection.

Upscaling and downscaling approaches for early season rice yield prediction using Sentinel-2 and machine learning for precision nitrogen fertilisation

Early season yield prediction could support rice farmers in adopting precision agriculture for nitrogen fertilisation management. Remote sensing and machine learning (ML) can be used to predict and map crop yield during phenological stages relevant to nitrogen application, like tillering in rice, at both within-field and field scales. This study evaluated the transferability of ML models in early season yield prediction through upscaling and downscaling approaches. The effects of two prediction times (tillering and ripening stages) and training/testing set sizes on ML models performance were evaluated over five rice growing seasons (from 2018 to 2022) in northern Italy, using whole-field-average yields and yield maps. Vegetation indices from Sentinel-2 imagery using the Google Earth Engine platform fed five ML algorithms (Cubist-CUB, Gaussian Process Regression-GPR, Neural Network-NNET, Random Forest-RF, and Support Vector Machines-SVM). ML algorithms were trained with yield maps and tested with whole-field-average yields to obtain a downscaling approach, while the opposite was done to obtain an upscaling approach. The downscaling approach showed higher accuracy than upscaling approach. Ripening stage predictions were more accurate than tillering stages, although the downscaling approach showed small differences between tillering and ripening stages. The highest tillering stage accuracy was achieved by SVM for both downscaling and upscaling approaches with 20 % and 27.8 % of Normalized Root Mean Square Error (NRMSE), and 0.99 and 0.99 of Simple Additive Weighting (SAW) score, respectively. Set size and data distribution effected ML models accuracy, with the highest performance achieved by RF and GPR with 0.80 and 1.00 of SAW score for the downscaling and upscaling approaches, respectively. This study demonstrated how ML models and downscaling approach could support rice farmers to calculate the nitrogen dose using the predicted yield at the tillering stage, enabling them to apply a site-specific nitrogen fertilisation based on the within-field yield prediction variability.

Understanding tea plantation growth stages and cultural operations by synergy of optical and Synthetic Aperture Radar (SAR) indices

ABSTRACT Tea, as a key cash crop, demands year-round monitoring including the rainy months. Our study in Dehradun, Uttarakhand, India, leverages Synthetic Aperture Radar (SAR) data for its cloud-penetrating capabilities. Radar Vegetation Index (RVI) from Sentinel-1 and the Normalized Difference Vegetation Index (NDVI) from Sentinel-2 on the Google Earth Engine (GEE) platform was used. Results indicate that RVI can detect stages of tea plantation growth, even during the rainy season, while NDVI provides detailed insights into cultural operations, except rainy seasons. This method supports effective tea plantation monitoring and supply chain management across seasons.

Mapping aquaculture in inland continental areas of Brazil using machine learning on the Google Earth Engine

Aquaculture has played a significant and growing role in the provision of food, food security, and job creation, while simultaneously establishing itself as an important agricultural activity subject to increasingly stringent market practices. The generation of spatial information on aquaculture activity in Brazil can support more assertive planning of new enterprises as well as enable the monitoring of this activity over time, fulfilling an agenda of sustainable aquaculture growth in the country. The aim of this article was to map aquaculture areas in the state of Paraná (Brazil) by developing an automated methodology for the extraction of aquaculture ponds. The methodology was implemented on the Google Earth Engine (GEE) platform and utilized satellite images with a spatial resolution of 4.77 m from Norway’s International Climate and Forest Initiative (NICFI) Program – Planet. The process involved pixel-oriented classification of satellite images and the utilization of the Random Forest (RF) algorithm for classification, based on 1200 training samples. The accuracy of the mapping and the algorithm was evaluated using Producer Accuracy (PA), User Accuracy (UA), and Overall Accuracy (OA). The mapping achieved an OA of 0.87, with a PA of 0.89 and a UA of 0.96 for the aquaculture class, and a PA of 0.66 and a UA of 0.41 for the non-aquaculture class. 42,369 aquaculture ponds were identified in the study area, covering a total area of 11,515 ha. The ponds had sizes smaller than 0.7 ha and non-circular shapes (compactness 0.70). Around 40% of the identified aquaculture tanks were concentrated in just two out of the ten analyzed mesoregions, accounting for 40% of the water surface area used for aquaculture in Paraná. Considering that the results obtained show potential to fill the gap in geospatial information about state aquaculture, our methodology would provide a broad understanding of the state’s aquaculture framework, definition of regions with capacity for the expansion of aquaculture ponds, accuracy in estimating the state’s productive space, monitoring the improper expansion of aquaculture areas, as well as establishing locations where the pressure on water resources by aquaculture prevents the expansion of this activity. Therefore, this study could benefit public and private organizations, and other stakeholders, by systematically monitoring aquaculture production in the country’s leading fish farming state, contributing to assertive decisions in regions where aquaculture production needs to expand sustainably.

Canopy Height Mapper: a Google Earth Engine application for predicting global canopy heights combining GEDI with multi-source data.

Spatially and temporally discontinuous canopy height footprints collected by NASA’s GEDI (Global Ecosystem Dynamics Investigation) mission are accessible on the Google Earth Engine (GEE) cloud computing platform. This study introduces an open-source, user-friendly, code-free GEE web application called Canopy Height Mapper (CH-GEE), available at https://ee-calvites1990.projects.earthengine.app/view/ch-gee, which automatically generates (10 m) high-resolution canopy height maps for a specific area by integrating GEDI with multi-source remote sensing data: Copernicus and topographical data from the GEE data catalogue. CH-GEE generates local-to-country scale calibrated canopy height maps worldwide using machine learning algorithms and leveraging the GEE platform's big data and cloud computing capabilities. CH-GEE allows customization of geographic area, algorithms and time windows for GEDI and predictors. Canopy heights generated by CH-GEE were validated using the Italian National Forest Inventory across 110,000 km2 at multiple scales (Country-based R-squared = 0.89, RMSE = 17%). CH-GEE's accuracy and scalability make it suitable for forest monitoring.

Variations in air pollution before, during and after the COVID-19 lockdown in Peruvian cities.

The high concentrations of air pollutants in Peru remain a persistent problem, significantly impacting public health. Understanding the extent to which the COVID-19 lockdown affected these contaminants is crucial. To determine variations in NO2, O3, CO, and SO2 concentrations in 10 Peruvian cities before, during, and after lockdown. A comparative ecological study was conducted in urban areas of 10 major Peruvian cities using the Google Earth Engine (GEE) platform. Data on atmospheric pollutant concentrations were extracted from the Sentinel-5P/TROPOMI satellite images for the period between March 16 and June 30, across the years 2019, 2020, 2021, and 2022, for comparative analysis. The Wilcoxon test was used to evaluate changes between the study periods. We included 10 urban cities located across three geographic regions of Peru. Most urban cities experienced a decrease in NO2 concentrations and an increase in O3 and CO levels during the lockdown, while SO2 concentrations remained relatively constant. The lockdown has caused variations in NO2, O3 and CO concentrations. Future studies with accurate data on air pollutant concentrations are needed to ensure targeted and effective interventions.

Climate change and vegetation greening jointly drive the spatial pattern of net radiation variability in northern China

ABSTRACT The spatial and temporal variations of net surface radiation (Rn) are critical for comprehending ecological environments. Nonetheless, the intricate interplay among Rn dynamics, vegetation growth, climate, and natural factors remains inadequately elucidated. In this study, we estimated net surface radiation based on Landsat data and ERA5 meteorological data in the Google Earth Engine (GEE) platform, which closely matched the observable spatial distribution (R 2 = 0.96), with an average growth rate of 0.15 MJ m−2 mth−1. Trend analyses and spatial autocorrelation were used to explore the spatial and temporal changes in net radiation from 2000 to 2020, the global Moran's index for net radiation was found to exceed 0.76, with fluctuating increases, showing a highly positive spatial distribution of Rn. Local Moran's I predominantly fell into two categories: ‘High-High’ and ‘Low-Low’, with the first increasing in range and the latter decreasing. Combining GeoDetector and PLS-SEM analyses, temperature and vegetation emerge as predominant drivers of net surface radiation variation within the study area, each contributing more than 17% to Rn change. Furthermore, interactions between any two factors typically exhibits nonlinear enhancement. PLS-SEM underscores the influence of vegetation and climate on Rn, with other factors indirectly affecting net radiation changes by influencing vegetation growth.

Synthesizing Landsat images using time series model-fitting methods for China’s coastal areas against sparse and irregular observations

ABSTRACT Long historical records and free accessibility have made Landsat data valuable for time-series analysis. However, Landsat time-series analysis is restricted for coastal areas due to the lack of sufficient numbers of clear images. The generation of synthetic Landsat images using model-fitting methods is accepted as an effective means of overcoming this problem. But for coastal areas, available observations are typically sparse and irregular, leading to ill-fitted models that distort synthetic Landsat images. In this study, we propose a linear harmonic model with different orders and implement it on an annual basis to generate synthetic Landsat images based on the Google Earth Engine (GEE) platform. First, we incorporated the available observations from adjacent years and filtered them by season to address the problems of data sparsity and irregularity. Then, we combined the threshold segmentation and ridge regression approaches to address the fake cloud problem, which sometimes contaminates synthetic Landsat images in summer. All cloud-free real Landsat images from five typical sites along China’s coasts during 1986–2020 were used to evaluate the accuracy and robustness of the synthetic Landsat images generated using our proposed methods. The principal results are as follows: (1) the R2 value of the linear harmonic model averaged across different Landsat bands and land cover types was 0.640, and the model was especially successful in simulating the surface reflectance in near NIR bands in forest and grassland areas; (2) 81.1% of the synthetic Landsat images covered by fake clouds were effectively restored, and there was the particular need to remove fake clouds for the synthetic Landsat images at low latitudes; (3) The mean absolute error for our synthetic Landsat images was 0.015, with the rate of 0.124; this was achieved under the clear-sky probability of only 36% and the average number of annual observations below 9, indicating a good performance. Compared with the continuous change detection and classification (CCDC) and seasonal median composite (SMC) methods, our proposed method offers advantages in both the accuracy and integrity of synthetic Landsat images. Our proposed method for synthesizing images also has potential for application to global coastal areas and other satellite datasets.

Examining thedrivers of forest cover change and deforestation susceptibility in Northeast India using multicriteria decision-makingmodels.

The increasing rates of forest cover change and heightened vulnerability to deforestation present significant environmental challenges in Northeast India. This study investigates the dynamics of forest cover changeand susceptibility to deforestation in this region from 2001 to 2021, utilizing data from the Hansen Global Forest Change (HGFC) productonthe Google Earth Engine (GEE) platform. A suite of multicriteria decision-making (MCDM) models-including VlseKriterijumska optimizacija I Kompromisno Resenje (VIKOR), Simple Additive Weighting (SAW), Evaluation Based on Distance from Average Solution (EDAS), and Weighted Aggregates Sum Product Assessment (WASPAS)-was employed to assess changes in forest cover and deforestation susceptibility across varied zones. Multicollinearity tests confirmed the relevance of the factors influencing deforestation. Statistical validations, such as the Wilcoxon Signed Ranks Test, underscored the models' robustness, revealing statistically significant outcomes. Additionally, Receiver Operating Characteristic (ROC) curve and Area Under the Curve (AUC) analysis demonstrated the superior fit of the VIKOR model (AUC = 0.938) compared to SAW (AUC = 0.901), EDAS (AUC = 0.895), and WASPAS (AUC = 0.864) in predicting current deforestation susceptibility. Validation affirmed the reliability of all MCDM methods, with VIKOR displaying high sensitivity (True Positive Rate, TPR = 0.878) and optimal AUC (0.938). Correlation analyses among the models identified significant inter-relationships, notably a positive correlation between EDAS and SAW, and a negative correlation between VIKOR and SAW. The regions of Assam, Nagaland, Mizoram, and Arunachal Pradesh were identified as experiencing significant forest cover loss, indicating a pronounced susceptibility to future deforestation. These findings underscore the need for immediate intervention to address this critical environmental issue.