google earth case study

Case Studies

Global forest cover change.

A team led by University of Maryland’s Matt Hansen used Earth Engine to survey over a decade of global tree cover extent, loss, and gain. The study , published in Science, analyzed nearly all global land, excluding only Antarctica and some Arctic islands. This area comprises 128.8 million km 2 , which is the equivalent of 143 billion pixels of Landsat data at a thirty-meter spatial resolution. To conduct such extensive analysis, Earth Engine performed computations in parallel across thousands of machines, as well as automatically managed data format conversion, reprojection and resampling, and image-to-pixel metadata association. Learn more.

This is the first map of forest change that is globally consistent and locally relevant. What would have taken a single computer 15 years to perform was completed in a matter of days using Google Earth Engine computing.

Map of Life

The Map of Life team has developed an interactive map for conservators to view and analyze habitat ranges and to assess the security of individual species. Using Earth Engine to combine data from a variety of sources, Map of Life has refined their predictions for pinpointing the locations of at-risk species. Users can adjust the parameters (indicating, for instance, a species' preferred habitat), and Earth Engine updates the map on-the-fly, immediately showing the impact on the species range and the amount of protected habitat. Learn more.

Earth Engine's scalable and cloud-based technology helps us deliver vastly improved estimates about the status and trends of tens of thousands of species to users in science, conservation and policy anywhere in a visual and interactive way.

Global Forest Watch

Global Forest Watch , an initiative of the World Resources Institute , is a dynamic online forest monitoring system designed to enable better management and conservation. Global Forest Watch uses Earth Engine to measure and visualize changes to the world's forests; users can synthesize data from over the past decade or receive alerts about possible new threats in near-real-time. Launched in 2014, it’s now used by corporations, non-profits, governments, and indigenous groups for applications as diverse as protecting against illegal logging and ensuring supply chain transparency. Learn more.

Google Earth Engine has made it possible for the first time in history to rapidly and accurately process vast amounts of satellite imagery, identifying where and when tree cover change has occurred at high resolution. Global Forest Watch would not exist without it. For those who care about the future of the planet Google Earth Engine is a great blessing!

Tiger Habitat Monitoring

A team led by University of Minnesota's Anup Joshi developed a satellite-based monitoring system to track changes and prevent loss to critical endangered wild tiger habitats. Using Google Earth Engine, forest loss data generated by Dr. Matt Hansen and Google, and other data available at Global Forest Watch , the team assessed the changes to all critical tiger habitats over a 14 year period. The assessment is the first to track all 76 areas prioritized for wild tiger conservation across 13 different countries. Their analysis found that the international goal to double the wild tiger population by 2022 is achievable with effective forest protection and management. Learn more.

It took us about 1.5 years each to do the previous two range-wide tiger habitat analyses, but with Google Earth Engine we were able to get it done in less than a week.

Malaria Risk Mapping

Scientists in the Global Health Group at the University of California, San Francisco , are using Earth Engine to predict malaria outbreaks. When their tool is released, local health workers will be able to upload their own information about known cases of malaria, and the platform will combine it with real-time satellite data to predict where new cases are likely to occur. Learn more.

Here at the UCSF Global Health Group, we have been using Earth Engine as the workhorse for an online Disease Surveillance And Risk Mapping (DiSARM) platform for malaria. Earth Engine makes accessing, processing and analyzing remotely sensed data so much easier than other more manual methods and, as it is updated frequently, allows us to automate malaria risk mapping in near real time.

Collect Earth

Collect Earth , developed by the Food and Agriculture Organization (FAO) of the United Nations, is a free, open source, and user-friendly tool using Google Earth and Google Earth Engine to visualize and analyze plots of land in order to assess deforestation and other forms of land-use-change. Launched in 2014, Collect Earth is part of the Open Foris software suite , designed to help government, universities and non-profit organizations monitor land use, desertification, forest change, and land-use dynamics. Learn more.

Collect Earth is a game changer – thanks to Google Earth Engine, we will be able to monitor the world’s forests much more efficiently together with all other actors.

Global Surface Water

The European Commission's Joint Research Centre (JRC) has used Earth Engine to develop high-resolution maps of global surface water occurrence, change, seasonality, recurrence, and transitions. The study , published in Nature, analyses Landsat images collected over the past three decades to identify both permanent and seasonal water bodies. Understanding these changes is vital for ensuring the security of our global water supply for agriculture, industry, and human consumption; for assessing water-related disaster reduction and recovery; and for the study of waterborne pollution and the spread of disease. Learn more .

Earth Outreach

Get inspired by organizations who have used google mapping tools for good.

• All countries
• Google Cloud
• Google Cloud Platform
• Google Earth
• Google Earth Engine
• Google Earth Pro
• Google Maps
• Google Maps Platform
• Google My Maps
• Google Street View
• Open Data Kit
• Agriculture
• Conservation
• Cultural Mapping
• Humanitarian
• Public Health

• < Previous Event
• Next Event >

Home > WLGISDAY > gisdays2021 > lightningtalks > 33

Lightning Talks

Google Earth Engine: A Case Study in Forest Cover and Landscape Metric Calculation, Lightning Talk (7 min)

Presenter Information

Miranda Potsma , Western University Follow

London, Ontario

https://westernu.maps.arcgis.com/apps/dashboards/61234059488042ccb4ad3b9583e03dee

17-11-2021 2:00 PM

17-11-2021 3:00 PM

Description

A brief case-study in how Google Earth Engine can be useful for studying changes in forest cover over time. This presentation will be looking specifically at the JavaScript (browser-based) version of Google Earth Engine, and will be examining its strengths and weaknesses when it comes to generating forest cover layers and calculating landscape metrics.

SRT file available upon request, contact the GIS team via https://guides.lib.uwo.ca/gis/support.

Creative Commons License

Since November 30, 2021

https://ir.lib.uwo.ca/wlgisday/2021/lightningtalks/33

Advanced Search

• Notify me via email or RSS
• Expert Gallery
• Online Journals
• eBook Collections
• Reports and Working Papers
• Conferences and Symposiums
• Electronic Theses and Dissertations
• Digitized Special Collections
• All Collections
• Disciplines

Author Corner

• Submit Proposal
• GIS Days 2021

Home | About | FAQ | My Account | Accessibility Statement | Privacy | Copyright

©1878 - 2016 Western University

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 16 February 2023

Investigation of the spatial and temporal variation of soil salinity using Google Earth Engine: a case study at Werigan–Kuqa Oasis, West China

• Shilong Ma 1 , 2 , 3 ,
• Baozhong He 1 , 2 , 3 ,
• Boqiang Xie 1 , 2 , 3 ,
• Xiangyu Ge 1 , 2 , 3 &
• Lijing Han 1 , 2 , 3  

Scientific Reports volume  13 , Article number:  2754 ( 2023 ) Cite this article

2802 Accesses

6 Citations

1 Altmetric

Metrics details

• Environmental sciences
• Natural hazards

Large-scale soil salinity surveys are time-costly and labor-intensive, and it is also more difficult to investigate historical salinity, while in arid and semi-arid regions, the investigation of the spatial and temporal characteristics of salinity can provide a scientific basis for the scientific prevention of salinity, With this objective, this study uses multi-source data combined with ensemble learning and Google Earth Engine to build a monitoring model to observe the evolution of salinization in the Werigan–Kuqa River Oasis from 1996 to 2021 and to analyze the driving factors. In this experiment, three ensemble learning models, Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Light Gradient Boosting Machine (LightGBM), were established using data collected in the field for different years and some environmental variables, After the accuracy validation of the model, XGBoost had the highest accuracy of salinity prediction in this study area, with RMSE of 17.62 dS m −1 , R 2 of 0.73 and RPIQ of 2.45 in the test set. In this experiment, after Spearman correlation analysis of soil Electrical Conductivity (EC) with environmental variables, we found that the near-infrared band in the original band, the DEM in the topographic factor, the vegetation index based on remote sensing, and the salinity index soil EC had a strong correlation. The spatial distribution of salinization is generally characterized by good in the west and north and severe in the east and south. Non-salinization, light salinization, and moderate salinization gradually expanded southward and eastward from the interior of the western oasis over 25 years. Severe and very severe salinization gradually shifted from the northern edge of the oasis to the eastern and southeastern desert areas during the 25 years. The saline soils with the highest salinity class were distributed in most of the desert areas in the eastern part of the Werigan–Kuqa Oasis study area as well as in smaller areas in the west in 1996, shrinking in size and characterized by a discontinuous distribution by 2021. In terms of area change, the non-salinized area increased from 198.25 in 1996 to 1682.47 km 2 in 2021. The area of saline soil with the highest salinization level decreased from 5708.77 in 1996 to 2246.87 km 2 in 2021. overall, the overall salinization of the Werigan–Kuqa Oasis improved.

Similar content being viewed by others

A longitudinal analysis of soil salinity changes using remotely sensed imageries

Coupling of machine learning and remote sensing for soil salinity mapping in coastal area of Bangladesh

Deep learning and data fusion to estimate surface soil moisture from multi-sensor satellite images

Introduction.

Soil salinization has become one of the threats to global agricultural systems 1 , and it is expected that with climate change, the impact of salinization will be wider and the degree of harm will increase, in addition, the formation mechanism of salinization is complex 2 . For regulating salinization and preventing soil degradation, it is crucial to comprehend the characteristics of salinization's spatial and temporal distribution and its evolutionary patterns 3 .

Traditional laboratory analysis for soil salinity monitoring is time-consuming and labor-intensive, and because salinity changes widely across space and time, it is challenging to precisely characterize the geographical distribution of salinity and its evolutionary patterns 4 . Digital mapping has made a splash in the field of soil science, thanks to the advancement of computer hardware and software, as well as the creation of geographic information systems, global positioning systems, remote or proximity sensors, and digital elevation models that have generated huge volumes of data 5 , The use of remote sensing techniques to detect salinity has increased in importance with the emergence of remote sensing satellites. Microwave and multitemporal optical remote sensing are efficient methods for identifying surface salinity parameters 6 .

Various salinity indices have been constructed for modeling and prediction using the rich waveband information of optical satellites 7 , 8 . As in the instance of Khan et al. 9 who utilized salinity indices (SI) to categorize and analyze salinity-prone terrain, remote sensing-based salinity indices can instantly respond to the salinity status of the surface in places where it is barren or sparsely vegetated. Due to the influence of other elements including soil moisture, vegetation cover, and data collection time, it is extremely challenging to obtain pure saline spectral information in natural situations. Because salt-tolerant plants thrive in arid and semi-arid climates, vegetation index is employed as an Indirect indicator for salinity 10 . Many salinity prediction studies, such as Ramos, et al. 7 used the Canopy Response Salinity Index (CRSI), Enhanced Vegetation Index (EVI), and Normalized Difference Vegetation Index (NDVI) to assess salinity in the field; other indices widely used for salinity monitoring are Soil Adjust Vegetation Index (SAVI), Ratio Vegetation Index (RVI), and Divergence Vegetation Index (DVI), and Green Vegetation Index (GVI) 11 , 12 .

The formation of soil salinity is highly nonlinearly related to many environmental factors, and machine learning algorithms are popular in the field of salinity research using their efficient data mining capabilities 13 , 14 . It has been difficult to choose the optimal model for a specific area when digitally mapping soils, but machine learning has been demonstrated to perform better than conventional statistical models at accurately predicting salinity 15 , 16 . The performance of various machine learning algorithms has also been compared with linear regression models and among machine learning algorithms for salting inversion analysis, including Multi-Layer Perceptron-Artificial Neural Network (MLP-ANN), Multivariate Adaptive Regression Splines (MARS), Classification and Regression Tree (CART), support vector regression (SVR), and RF. With the maturation of the ensemble learning method, it is frequently employed in picture classification research 17 , nevertheless, it is not commonly used in soil salinity prediction studies. To assess the geographical variability of soil salinity and alkalinity in agricultural regions impacted by salinity, several researchers have employed random forests, with satisfactory results 18 . Recent studies that forecast salinity have employed XGBoost 19 , 20 , while other ensemble learning techniques, including light gradient boosting machine, have seldom ever been published in the field of salinity research LightGBM 21 . Therefore, in this study, three ensemble learning models were applied to the prediction and mapping of salinity to evaluate their potential application in salinity monitoring efforts. Long-term salinity monitoring in arid and semi-arid areas is essential because it can adequately address local human-land linkages and serve as a guide for salinity control. The enormous volume of data makes the information extraction procedure in multi-temporal remote sensing challenging. Advantageously, Google Earth Engine offers a powerful data processing platform that includes a variety of geographical data, including various types of remote sensing data 22 . The spatial and spectral resolution of multispectral remote sensing is well suited for salinity monitoring due to its large coverage and ease of acquisition 6 , 23 . In this study, Landsat5 TM and Landsat OLI satellites were selected as the remote sensing data sources for this study because of the need to predict the salinity distribution in the inversion epoch and because of the good performance of Landsat satellites in salinity monitoring 24 , 25 .

In this study, four years of experimental data were aggregated to make the prediction model more stable and to produce more accurate information on the spatial distribution of salinization. The specific objectives of this study were: (1) Evaluating the predictive power of RF, XGBoost, and LightGBM in ensemble learning for soil conductivity (2) Digital mapping of salinity distribution in 1996, 2006, 2017, and 2021 based on remote sensing data using an optimal prediction model; (3) The spatial and temporal variable features of salinization in Werigan–Kuqa Oasis during the last 25 years; (4) Discuss the effects of arable land expansion and land remediation on salinity.

Materials and methods

The area of study is the Werigan–Kuqa River Oasis (also known as the Werigan–Kuqa Oasis), which is situated at an altitude of 901–1069 m above sea level in the north-central Tarim Basin of the Xinjiang Uygur Autonomous Region. It has an area of around 9769.76 km 2 . The Werigan–Kuqa Oasis features a typical warm-temperate continental dry climate due to its deep interior location and distance from the sea, with average annual precipitation and evaporation of 70 and 1100 mm, respectively, and a high evapotranspiration ratio of 16:1. The research region mostly consists of desert, agriculture, grassland, and woodlands, with salt- and drought-tolerant plants flourishing in the desert. Werigan–Kuqa Oasis is generally flat, with a high water table, a long dry season, and strong evaporation. In this context, salts can easily accumulate on the surface, so the area chosen as the study area is representative and has great significance for the improvement of the ecological environment and the development of agricultural production (Fig.  1 ).

Figure ( A ) shows the location of Xinjiang, Figure ( B ) shows the location of the study area in Xinjiang, Figure ( C ) shows the distribution of sampling sites in the study area in different years, and figure ( D ) is the elevation of the study area.

Sample collection and survey

Field sampling and surveys of the Werigan–Kuqa Oasis are conducted annually, with most of the sampling taking place in July each year. The location of sampling points as well as the number of sampling points were determined by combining existing digital soil maps (salinity maps, soil type, soil texture) and land use/cover types, while sampling strategies were changed based on field observations from the previous year to take into account changes from year to year (Fig.  1 ). The location of each sampling point is recorded using a portable GPS, and the soil samples are packed in (approximately 500 g) transparent sealed bags for the next step of laboratory analysis. In this study, 4 years of soil surface (0–10 cm) electrical conductivity (EC) data were summarized and screened. The sampling times in the field were July 2006, with 36 samples; July 2017, with 84 samples; July 2018, with 75 samples, and June 2021, with 63 samples. All samples underwent air drying, grinding, homogenization, and sieving at a 0.15 mm size. For every 20 g of soil, add 100 ml of distilled water, mix thoroughly for 30 min, and then leave for 24 h. At room temperature of 25 °C, the soil conductivity was measured using a digital multiparameter measuring system (Multi 3420 Set B, WTW GmbH, Germany) fitted with a composite electrode (TetraCon 925) 26 .

Environmental variables

The key to the selection of environmental variables is that the covariates must respond to the nature of soil formation, climate, biology and landscape type, etc. According to the SCORPAN framework (S is for soil, C is for climate, O is for organisms, R is for relief, P is for parent material, A is for age, and N is for space.) 5 , a series of environmental factors were selected, including each of the original bands of Landsat5 TM and Landsat8 OLI, various indices derived from remote sensing (vegetation index, salinity index), elevation data and their derived indices (e.g. terrain moisture index, TWI).

Remote sensing-based environment variables

In this study, the remote sensing-based index extraction was done in the Google Earth Engine cloud platform. The Landsat5 TM image of July 22, 2006, and Landsat8 OLI images of July 4, 2017, July 23, 2018, and July 15, 2021, are selected, which matched the sampling time, were selected to have less than 10% cloudiness. The remote sensing-based environmental variables include 6 raw bands, 12 vegetation indices, 9 salinity indices, 1 carbonate index, and 1 brightness index (Table 1 ).

Terrain attributes

In this study, 11 topographic indices were generated using 30 m resolution DEM data from the Geospatial Data Cloud ( http://www.gscloud.cn/ ), clipped, and stitched together using SAGA GIS software (Table 2 ). The results of Vermeulen and Van Niekerk 41 showed that the use of elevation data and its derived topographic indices as geostatistical and machine learning input variables have a great potential for salinity prediction to monitoring salt accumulation in irrigated areas.

Model framework

Random forest.

Random Forest, developed by Breiman 42 , is a popular ensemble learning algorithm based on tree-based bagging (bootstrap aggregation) 43 , which has the advantage of having nonlinear mining capabilities, data distribution that does not need to conform to any assumptions, handling both rank and continuous variables, preventing overfitting, fast training, and quantitative description of the contribution of variables. RF is a bagging improvement that enhances variable selection 44 , Instead of selecting the optimal split among all characteristics at each node, RF randomly picks a subset of features to decide the split, this makes RF more resilient to noise and less prone to overfitting. In addition, RF can handle outliers very well 45 . The number of trees and predictor variables that the random forest model allows the decision tree to grow as large as it can without being trimmed is its critical factor. The primary hyperparameters modified in this study are the number of trees in the forest and the number of features thought to divide at each leaf node 46 . In this work, we used the open-source machine learning package Scikit-learn to create an RF mode 47 .

Extreme gradient boosting

Extreme Gradient Boosting (XGBoost) is a popular boosting-based ensemble machine learning algorithm 48 , this algorithm was used in the Kaggle signal recognition competition and has attracted a lot of attention for its outstanding efficiency and high prediction accuracy 49 . Boosting, in contrast to bagging, is an iterative method that successively adds new trees to the integration, and samples erroneously predicted by the prior tree are given higher weights in the succeeding trees. Thanks to numerous significant systematic and algorithmic enhancements, the gradient boosting framework is implemented effectively and flexibly in XGBoost 49 , 50 . The number of gradients boosting trees (n_ estimators), learning rate (eta), maximum depth of the tree (max_depth), and column per level of the subsample ratio are some of the important hyperparameters that are tuned by XGBoost. To train XGBoost models, the open-source Scikit-Learn software is utilized.

Light gradient boosting machine

Light Gradient Boosting Machine (LightGBM) is a framework that implements the idea of GBDT (Gradient Boosting Decision Tree) algorithm 51 , a boosting decision tree tool open-sourced by the Microsoft DMTK team, which has fast training speed and less memory usage, which greatly speeds up the training and also has better model accuracy. LightGBM performs the following optimizations on the traditional GBDT algorithm: Gradient-based One-Side Sampling (GOSS) and Exclusive Feature Bunding (EFB) 51 . GOSS is a subsampling technique used to create training sets to build the base tree in the integration, select data with larger gradients from the sample to increase their contribution to the computed Information gain, and EFB merges certain data features to reduce the data dimensionality 52 . Generally, the prediction accuracy is significantly influenced by the hyperparameters 53 . So, before employing LightGBM, we need the first figure out how many and how widely its hyperparameters may vary. The number of Leaves, Learning Rate, and Maximum Depth is the important factors.

For this experiment, the above three models were done in the Spyder platform based on the Python 3.9.7 programming language.

Model parameter optimization

The efficacy of the model application depends on the choice of model parameters. In the fields of statistical analysis and machine learning, the K-Fold cross-validation method is frequently used to assess the generalizability of models. The grid search method is an exhaustive search method that specifies the values of the parameters, it is carried out by Scikit-GridSearchCV, learn's which arranges and combines the possible values of each parameter, lists all combinations that could exist, and performs cross-validation to optimize the estimation function's parameters in order to obtain the best learning algorithm 54 . The minimum value of Root Mean Square Error (RMSE) is used as the criterion for the selection of model parameters, In this experiment, it is assumed that the value of K is 5, as follows:

Divide the dataset into the training set, test set, and K-fold division of the training set data.

Determine the range of each parameter of the model, taking a random forest as an example, and determine the number of decision trees m as well as the depth h. The combination of parameters is the cross nodes of a two-dimensional grid with m and has horizontal and vertical axes.

Choose any K-1 data from the training set, choose a set of cross-node parameters, create one decision tree using a sample of all the K-1 data, forecast the final 1 data, and compute the average root mean square error of all trees on the final 1 training sample.

Repeat the above two steps until you have traversed K-1 copies of the data.

Iterate through the parameter combinations of all crossover nodes of the grid. 6.

Steps 3 to 5 are repeated, using cross-validation to calculate the performance of the model in the test dataset. (Table 3 ) shows the combination of model parameters optimized by grid search.

Evaluation of prediction accuracy

In this research, the coefficient of determination R 2 , the root mean square error(RMSE), and the performance to interquartile distance(RPIQ) are used to assess the performance of RF, XGBoost, and Lightgbm. The closely R 2 is to 1, the more accurate models are fitted. The closer the number is to 0, the smaller the difference between the measured value and the predicted value of the model, and the greater the model's ability to forecast the future. The value of RMSE is inversely related to the accuracy of the model. RPIQ is the interquartile range to RMSE ratio, and the interquartile range is the difference between 75 and 25% of the sample values. It is commonly accepted that RPIQ < 1.7 implies low model prediction dependability, 1.7 ≤ RPIQ ≤ 2.2 suggests somewhat balanced prediction ability, and RPIQ ≥ 2.2 indicates highly strong prediction ability. RPIQ is a more reasonable and objective measure when compared to the Ratio of Performance to Deviation (RPD), especially for soil samples with an unusual distribution 55 , 56 . Equations ( 1 )–( 3 ) show the expression of these model evaluation metrics:

where N is the number of samples, X i is the measured EC value, Y i is the calculated value, X i * is the mean measured EC value, Y i * is the estimated soil EC value, SD represents the standard deviation, and ΔQ is the interquartile distance (IQR), which is the difference between the upper quartile ( Q3 ) and the lower quartile ( Q1 ).

Soil EC prediction and mapping for different years

The flow of this experiment is shown in (Fig.  2 ). The Google Earth Engine cloud platform was used to calculate and obtain the remote sensing-based environmental variables corresponding to the sampling time to establish a soil EC prediction model. Since the sampling time is mainly concentrated in July, based on the optimal model, the spatial distribution maps of soil EC in July of each year in 1996, 2006, 2017, and 2021 are obtained (the remote sensing data of June 24 is chosen because the remote sensing Image of July 1996 Is too cloudy to meet the mapping requirements), and this step is done by using the Spyder development environment with the help of GDAL, Pandas and other libraries to complete the mapping.

Flow chart.

Soil EC descriptive statistics

In this experiment, the final data of 258 soil EC samples were obtained after the outliers were removed from the sample data. Following statistical analysis, the soil's electrical conductivity (EC) minimum, maximum, mean, standard deviation, coefficient of variation, kurtosis, and skewness were determined (Table 4 ).

Soil EC values in the Werigan–Kuqa Oasis ranged from 0.079 dS m −1 to 143.4 dS m −1 , showing that the samples had a high span. The skewness of 1.37 is much higher than 0, which indicates that the sample data do not obey a normal distribution. The standard deviation was 33.2 dS m −1 and the coefficient of variation was 1.19, which is greater than 1, thus belonging to strong variability, which is consistent with the study of Wang, et al. 40 , showing the high spatial variability of soil EC values in the Werigan–Kuqa Oasis area.

Correaltion analysis

In modeling soil salinity monitoring, not all environmental variables are involved in modeling and there are differences in their contribution to EC prediction 40 , therefore, it is necessary to screen the environmental variables. Based on the statistical analysis of the sample EC values, the skewness was 1.47 (Table 4 ), so Spearman correlation analysis was used in the analysis of the relationship between environmental variables and soil EC values. In this study, 38 environmental variables (original band, vegetation index, salinity index, topography index, etc.) were initially selected, and after Spearman correlation analysis, 31 environmental variables were selected and the remaining relevant variables were not significantly correlated (Table 5 ).

Among the raw bands of remote sensing, the correlations with soil EC were NIR (R = − 0.610), SWIR2 (R = 0.423), Red (R = 0.372), SWIR1 (R = 0.3), and Green (R = 0.246) in descending order. Salinity indices, as direct indicators in salinity monitoring 57 , showed good correlation with soil EC, and all nine selected salinity indices were significantly correlated with EC values, with correlation coefficients up to 0.531(SIA, SIB, SIT, SAIO are all salinity indices, which are different combinations of different waveforms), The correlation between vegetation index and soil EC values in descending order is, GARI (R = − 0.626), EVI (R = − 0.596), DVI (R = − 0.572), GDVI (R = − 0.541), OSAVI (R = − 0.541),RVI (R = − 0.534), NDVI (R = − 0.533), SAVI (R = − 0.550), CRSI (R = − 0.506), GRVI (− 0.469), GNDVI (R = − 0.468), it can be seen that the vegetation index is a good Indicator as an Indirect Indicator of salinity monitoring. Compared to NDVI, SAVI increases the vegetation signal and decreases the soil background, therefore, there is a strong correlation with soil EC (R = − 0.55), in addition, OSAVI has the same correlation as SAVI, but OSAVI avoids the complex calculation of soil baseline parameters. Among the topographic correlation factors, the higher correlation is with DEM (R = − 0.463), followed by CND (R = − 0.175), and finally RSP (R = − 0.174). The lower correlation between topography and Its Indices with EC Is explained by the overall flatness of the Werigan–Kuqa Oasis. Finally, the carbonate index CAEX correlated significantly (R = 0.612) with soil EC values, which were determined by the soil properties of the study area.

Importance of selected environmental covariates

Different environmental factors have different predictive contributions to soil EC in predictive models, and not all environmental factors are significant variables in the modeling 58 , so it is necessary to rank the importance of environmental variables, and this study will rank the importance of features using each of the three models themselves and observe the differences in the contribution of variables in the three models.

Figures  3 , 4 , 5 show the results of the three models for feature selection, the degree of contribution of the variables differed, but individual variables showed high contribution in all three models, and among the vegetation indices, most of them generally contributed well, with CRSI being the most stable and showing high contribution in all three models, in agreement with Scudiero et al. 34 and Wu et al. 59 , GARI performed best among all environmental variables involved in RF. Remote sensing primitive bands are pivotal in the participation in modeling, in the study of related scholars, the relationship between each band and saline soils was analyzed in detail, the greater the salt in the soil, the higher the reflectance of all TM spectral bands 59 and the spectral reflectance of CaCO 3 , CaSO 4 ⋅ 2H 2 O, and gypsum sand were analyzed in the laboratory, they concluded that salt minerals can be detected when they are the main soil component 60 , among the primitive bands involved in modeling, the NIR band stands out, especially in the participation in the random forest modeling process, the contribution is second only to GARI. The salinity index stands out as a direct indicator in sparsely vegetated areas, and the SIA performed consistently in this study in terms of contribution across the three prediction models. The salinity index integrates most of the soil properties affected by salinity, and the salinity index is also very cost-effective for possible large-scale surveys to prevent soil salinity at the landscape scale 57 .

Characteristic importance diagram of RF.

Characteristic importance diagram of XGBoost.

Characteristic importance diagram of LightGBM.

Prediction accuracy

In this experiment, two approaches are used for model validation, the validation approach of slicing the dataset into training and test sets, and the cross-validation approach (Table 6 , Fig.  6 ), and it was found that the R 2 value of XGBoost was the highest among the three models in both the training and test sets, 0.84, 0.73, respectively, and the RMSE value was also the lowest in the training and test sets, 13.57 dS m −1 , 17.62 dS m −1 , respectively. The RPIQ value is also the highest, 3.32 in the training set and 2.45 in the test set. When RPIQ ≥ 2.2, it means that the model achieves excellent prediction, and compared with the performance of RF and LightGBM models in the test set (2.39 and 2.32, respectively), XGBoost has excellent prediction ability. Similarly, XGBoost has the lowest RMSE value of 19.9 dS m −1 for the three models after tenfold cross-validation. Therefore, XGBoost will be used as the optimal model for the digital mapping of the spatial distribution of salinity.

Measured and predicted regression analysis of the three models.

Spatial and temporal distribution characteristics and evolutionary trends of Salinization in 1996, 2006, 2017, and 2021

In the research region, all soil samples were divided into six groups by the frequently used soil salinity classification method for further analysis and visualization 61 (Table 7 ), and the spatial distribution of soil salinization in the Werigan–Kuqa Oasis on August 11, 1996, July 22, 2006, July 4, 2017, and July 15, 2021, were inverted using the selected optimal model and the corresponding optimal variables (Fig.  7 ). To further verify the accuracy of the salinity spatial distribution map after reclassification, this experiment used the 2017 and 2021 sample points as the validation set, and the accuracy was verified using the confusion matrix and kappa coefficient (Fig.  7 ), and the kappa coefficient was obtained as 0.71, which indicates that the salinity map has a high degree of consistency.

Confusion matrix verification.

According to (Fig.  8 ), the spatial distribution of salinization in the Werigan–Kuqa Oasis shows a distribution characteristic of good in the west and north and severe in the east and south. The moderate and below salinization in the Werigan–Kuqa Oasis is distributed in the west and north of the Werigan–Kuqa Oasis, an oasis area with good irrigation conditions (Fig.  1 ), where the main feature type is arable land, the terrain is relatively high, not easily waterlogged, and the vegetation cover is relatively high. With the expansion of the spatial extent of arable land, light salinization and below also show a corresponding radial change to the south, southwest, and southeast, and become more continuous spatially. By 2021, on the western and southern edges of the Werigan–Kuqa Oasis, very heavy salinization has been transformed into light salinization, in the eastern and northeastern regions, spatially discontinuous new arable land emerged, so that mild salinization also took the form of sporadic spatial distribution.

Spatial distribution of soil salinization in 1996, 2006, 2017 and 2021.

Severe and very severe salinization was mainly distributed in the northern part of the Werigan–Kuqa Oasis in 1996, and by 2006, salinization in the region improved and gradually shifted to the east and south, developing to the southeast by 2021. The development trend of severe and very severe salinization over 25 years is closely related to the low southeast and high northwest topography of the Werigan–Kuqa Oasis (Fig.  1 ).

The most pronounced spatial distribution and evolutionary characteristics of saline soils with the highest degree of salinization were mainly distributed in the southwestern edge of the Werigan–Kuqa Oasis and most of the desert areas in the east in 1996, shifting to classes such as severe and very severe in 2017, and improving significantly by 2021, especially in the eastern desert areas. Relying on years of field surveys, it was found that sparse salt vegetation grows in the eastern part of the Werigan–Kuqa Oasis, while the southeastern part of the area is sparsely forested. As a result of enhanced vegetation protection efforts in the eastern area, the vegetation cover has increased significantly and, therefore, the evaporation of surface water has decreased accordingly, reducing the rate of salt accumulation on the surface.

Change in area of salinization at different levels

As shown in (Fig.  9 ), the non-salinized area of the Werigan–Kuqa Oasis is 198.25 km 2 in 1996 and 1682.47 km 2 in 2021, an increase of 748.6%; Mild salinization was 346.78 km 2 in 1996 and increased year by year since then to 1441.29 km 2 in 2021, an increase of 315.6% compared to 1996; Moderate salinization remained stable from 1996 to 2006 and increased substantially by 2017 to 1062.26 km 2 by 2021, an increase of 134.8% compared to 1996; Heavy salinization was 431.26 km 2 in 1996 and 838.132 km 2 in 2021; Very heavy salinization remains relatively stable from 1996 to 2021, with an area of 2498.74 km 2 by 2021; The area of saline soil was 5708.77 km 2 in 1996, then declined to 5168.7 km 2 in 2006, followed by a greater decline to 794.48 km 2 in 2017 and 2246.87 km 2 in 2021, a decrease of 60.6% compared to 1996. Based on the results of the above statistical analysis: during the last 25 years, the non-salinized, lightly salinized, and moderately salinized areas increased more, the saline soil area decreased more, and the heavy and very heavy salinization changed less and remained stable, so there was an improvement of soil salinization in the Werigan–Kuqa Oasis.

Trends in the area of different levels of salinization.

Long-time series of salinity monitoring

Various multispectral sensors rely on the spectral reflectance properties of the ground for ground monitoring 62 , and the spectral reflectance varies for different levels of salinity, often with a white salt crust attached to the ground surface in highly saline areas. The higher the salinization, the higher the spectral reflectance of each band will increase accordingly 13 , therefore, it becomes possible to monitor salinization using raw bands or derived spectral indices of remote sensing. In previous studies on salinity monitoring, the choice of environmental variables varied, such as direct use of salinity indices for estimating soil salinity 63 , indirect estimation of soil salinity using vegetation indices 64 , or combining multiple environmental variables and grouping them to predict comparisons 58 .

The objective of this study is to map the spatial distribution of salinization in the Werigan–Kuqa Oasis in different years and analyze the changing trend of salinization area in different grades. Therefore, remote sensing data that can match the sampling time in different years are selected, and a stable soil EC prediction model is established based on the extraction of environmental variables from remote sensing images, which makes it possible to accomplish the goal of salinization spatial distribution mapping realistically and accurately and provide data reference for salinization management and water resources management. The earliest data collection in this study area began in 2006, so in this modeling, sample data from 2006, 2017, 2018, and 2021 were ensemble for modeling, making full use of the available laboratory data. This study utilizes the Google Earth Engine platform for fast online computational processing. Therefore, the remote sensing cloud platform presented by Google Earth Engine is an excellent option for environmental monitoring research that uses lengthy time series of remote sensing data.

Spatial and temporal evolutionary characteristics of salinization

The distribution of saline salinization in the Werigan–Kuqa Oasis shows distinct regional characteristics. In the southeast and east of the Werigan–Kuqa Oasis, which is the most affected area by salinization, salinization of very severe and higher grades is distributed, and the spatial and temporal evolution characteristics are obvious. The low elevation compared to other areas of the Werigan–Kuqa Oasis (Fig.  1 D) makes it possible to distribute high concentrations of salts in this area 40 . After years of field investigation and sampling, seasonal floods often gather in this area, and according to Ding and Yu 4 , it was found that the salts accumulated on the surface of the area do not drain outward, which makes it more difficult to manage salinization. In addition, the area is dominated by sandy soils, and during the dry season, salts are easily deposited on the surface after water evaporation 4 . During the 25 years, salinization in the eastern part of the Werigan–Kuqa Oasis has improved significantly because the local government has strengthened the vegetation protection of the desert, and built alkali drainage canals in the sparsely vegetated areas of the desert to reduce seasonal waterlogging to a certain extent, and strictly monitored overgrazing practices, so that the vegetation coverage and the area covered by the area have gradually increased, and therefore the area of very heavy salinization in the area has decreased in recent years.

In the southeast of the Werigan–Kuqa Oasis fringe area, salinization of severe and higher grades is distributed and has not improved significantly in individual areas during the last 25 years, which is since the economy of the study area is dominated by irrigated agriculture and surface irrigation is a common irrigation method, and the salts in the soil inside the Werigan–Kuqa Oasis are transported to the downstream through surface irrigation water, which deposits salts on the downstream surface and eventually intensifies the formation of salinization This is the reason why salinization is higher at the edge of the oasis than in the interior of the oasis 4 .

The salinization of moderate and lower grades is distributed in the interior of the oasis. Since the economy of the study area is based on irrigated agriculture, especially in the western and southwestern regions of the study area, which are more dependent on this economic activity, the formation of mild salinization in the region is strongly related to agricultural irrigation, while the irrigation of the regional arable land is gradually changing from the previous surface irrigation to drip irrigation, which may aggravate salinization in the region. The spatial and temporal evolution of salinization within the oasis is also more pronounced during the 25 years, due to the expansion of the arable land area, which increases significantly by 2021 compared to 1996, especially in the southwest and northeast of the study area, and therefore, the salinization grade changes accordingly, from severe and above grade to moderate and below, and to ensure healthy crop survival, before planting The land is drained of alkali to ensure healthy crop survival. In addition, the salinization of arable land areas tends to be consistent, and the area of salinization of heavy and above grades is reduced and fragmented, because the local government has been carrying out comprehensive land improvement work, leveling dry land and barren land; renovating and reinforcing branch canals and field branch; building rural field roads less than 4.5 m, serving production and travel, especially since 2018, the local government has carried out the construction of high-standard farmland, making the land more flat and contiguous, with better agricultural facilities, more fertile land and better disaster resistance. The results of the study show that human activities are the key factors affecting the aggravation and management of salinization 58 , and the key lies in whether humans destroy or protect land and water resources, and as the core area of the Belt and Road, it should focus on the protection of the ecological environment, and its starting point should be the management of salinization in arid areas. The irrational use of water resources is related to the salinity of the soil 65 , so in the future, we should discuss the planting pattern of the Werigan–Kuqa Oasis and a more economical and efficient irrigation method. It is gratifying to note that the government has in recent years become more disciplined in water resources management, such as the implementation of the river chief system, which strictly regulates the reckless diversion of rivers; the implementation of the water station chief system in irrigation areas, which provides more precise and efficient control of irrigation water resources; and the implementation of the forest chief system, which increases the protection of forest land. Through these measures, the salinization of the Werigan–Kuqa Oasis has been improved.

Conclusions

This study uses multi-year field collection data and multi-source data with the help of the ensemble learning method and Google Earth Engine cloud platform to complete the digital mapping of salinity spatial distribution in 1996, 2006, 2017, and 2021, analyze the spatial and temporal evolution characteristics and driving factors of salinity in Werigan–Kuqa Oasis, and draw the following conclusions:

Among the three ensemble learning models, RF, XGBoost, and LightGBM, XGBoost had an RMSE of 17.62 dS m −1 , R 2 of 0.73, and RPIQ of 2.45 in the test set, which had higher prediction accuracy compared with the other two models, and more accurate salinization distribution maps were obtained using XGBoost.

The salinization in the study area generally shows the distribution characteristics of good in the west and north and severe in the east and south. The moderate and below salinization is distributed in the oasis areas with good irrigation conditions and smooth drainage. And severe and above salinization is mainly distributed in the desert areas in the east and southeast.

The spatial and temporal variation of salinization in the study area has changed significantly in the last 25 years, with non-salinization and light salinization expanding in the east and southwest spatial distribution with the increase of arable land area and effective remediation planning of arable land. The distribution area of salinization of severe and above grades has shrunk more significantly.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.

Singh, A. Soil salinization management for sustainable development: a review. J. Environ. Manag. 277 , 111383 (2021).

Article   CAS   Google Scholar  

Hassani, A., Azapagic, A. & Shokri, N. Global predictions of primary soil salinization under changing climate in the twenty first century. Nat. Commun. 12 , 6663. https://doi.org/10.1038/s41467-021-26907-3 (2021).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Hassani, A., Azapagic, A. & Shokri, N. Predicting long-term dynamics of soil salinity and sodicity on a global scale. Proc. Natl. Acad. Sci. U. S. A. 117 , 33017–33027. https://doi.org/10.1073/pnas.2013771117 (2020).

Ding, J. & Yu, D. Monitoring and evaluating spatial variability of soil salinity in dry and wet seasons in the Werigan–Kuqa Oasis, China, using remote sensing and electromagnetic induction instruments. Geoderma 235–236 , 316–322. https://doi.org/10.1016/j.geoderma.2014.07.028 (2014).

Article   ADS   Google Scholar  

McBratney, A. B., Mendonça Santos, M. L. & Minasny, B. On digital soil mapping. Geoderma 117 , 3–52. https://doi.org/10.1016/s0016-7061(03)00223-4 (2003).

Metternicht, G. I. & Zinck, J. A. Remote sensing of soil salinity: potentials and constraints. Remote Sens. Environ. 85 , 1–20. https://doi.org/10.1016/s0034-4257(02)00188-8 (2003).

Ramos, T. B. et al. Soil salinity assessment using vegetation indices derived from Sentinel-2 multispectral data. Application to Lezíria Grande Portugal. Agricultural Water Management 241 , 106387. https://doi.org/10.1016/j.agwat.2020.106387 (2020).

Article   Google Scholar  

Wang, J. et al. Soil salinity mapping using machine learning algorithms with the Sentinel-2 MSI in Arid areas, China. Remote Sens. 13 (2), 305. https://doi.org/10.3390/rs13020305 (2021).

Khan, N. M., Rastoskuev, V. V., Sato, Y. & Shiozawa, S. Assessment of hydrosaline land degradation by using a simple approach of remote sensing indicators. Agric. Water Manag. 77 , 96–109 (2005).

Zhao, W., Zhou, C., Zhou, C., Ma, H. & Wang, Z. Soil salinity inversion model of oasis in arid area based on UAV multispectral remote sensing. Remote Sens. 14 , 1804 (2022).

Allbed, A. & Kumar, L. Soil salinity mapping and monitoring in arid and semi-arid regions using remote sensing technology: a review. Adv. Remote Sens. 02 , 373–385. https://doi.org/10.4236/ars.2013.24040 (2013).

Peng, J. et al. Estimating soil salinity from remote sensing and terrain data in southern Xinjiang Province China. Geoderma 337 , 1309–1319. https://doi.org/10.1016/j.geoderma.2018.08.006 (2019).

Wang, J. et al. Machine learning-based detection of soil salinity in an arid desert region, Northwest China: a comparison between Landsat-8 OLI and Sentinel-2 MSI. Sci Total Environ 707 , 136092. https://doi.org/10.1016/j.scitotenv.2019.136092 (2020).

Article   ADS   CAS   PubMed   Google Scholar  

Hoa, P. V. et al. Soil salinity mapping using SAR Sentinel-1 data and advanced machine learning algorithms: a case study at ben Tre province of the Mekong river delta (Vietnam). Remote Sens. 11 , 128 (2019).

Zhou, T., Geng, Y., Chen, J., Pan, J. & Lausch, A. High-resolution digital mapping of soil organic carbon and soil total nitrogen using DEM derivatives, Sentinel-1 and Sentinel-2 data based on machine learning algorithms. Sci. Total Environ. 729 , 138244 (2020).

Lu, H., Yang, L., Fan, Y., Qian, X. & Liu, T. Novel simulation of aqueous total nitrogen and phosphorus concentrations in Taihu Lake with machine learning. Environ. Res. 204 , 111940 (2022).

Article   CAS   PubMed   Google Scholar  

Zhang, E., Zhang, X., Jiao, L., Li, L. & Hou, B. Spectral–spatial hyperspectral image ensemble classification via joint sparse representation. Pattern Recogn. 59 , 42–54 (2016).

Nabiollahi, K. et al. Assessing agricultural salt-affected land using digital soil mapping and hybridized random forests. Geoderma 385 , 114858. https://doi.org/10.1016/j.geoderma.2020.114858 (2021).

Abedi, F. et al. Salt dome related soil salinity in southern Iran: prediction and mapping with averaging machine learning models. Land Degrad. Dev. 32 , 1540–1554. https://doi.org/10.1002/ldr.3811 (2020).

Qi, G., Chang, C., Yang, W. & Zhao, G. Soil salinity inversion in coastal cotton growing areas: a integration method of satellite-ground spectral fusion and satellite-UAV collaboration. Land Degrad. Dev. https://doi.org/10.1002/ldr.4287 (2022).

Jafarzadeh, H., Mahdianpari, M., Gill, E., Mohammadimanesh, F. & Homayouni, S. Bagging and boosting ensemble classifiers for classification of multispectral, hyperspectral and PolSAR data: a comparative evaluation. Remote Sens. 13 , 4405. https://doi.org/10.3390/rs13214405 (2021).

Ivushkin, K. et al. Global mapping of soil salinity change. Remote Sens. Environ. 231 , 111260. https://doi.org/10.1016/j.rse.2019.111260 (2019).

Moreira, L. C. J., Teixeira, A. D. S. & Galvão, L. S. Potential of multispectral and hyperspectral data to detect saline-exposed soils in Brazil. GISci. Remote Sens. 52 , 416–436. https://doi.org/10.1080/15481603.2015.1040227 (2015).

Gorji, T., Yildirim, A., Hamzehpour, N., Tanik, A. & Sertel, E. Soil salinity analysis of Urmia Lake Basin using Landsat-8 OLI and Sentinel-2A based spectral indices and electrical conductivity measurements. Ecol. Indicat. 112 , 106173. https://doi.org/10.1016/j.ecolind.2020.106173 (2020).

Masoud, A. A., Koike, K., Atwia, M. G., El-Horiny, M. M. & Gemail, K. S. Mapping soil salinity using spectral mixture analysis of landsat 8 OLI images to identify factors influencing salinization in an arid region. Int. J. Appl. Earth Observ. Geoinform. 83 , 101944. https://doi.org/10.1016/j.jag.2019.101944 (2019).

Wang, J. et al. Capability of Sentinel-2 MSI data for monitoring and mapping of soil salinity in dry and wet seasons in the Ebinur Lake region, Xinjiang. China. Geoderma 353 , 172–187. https://doi.org/10.1016/j.geoderma.2019.06.040 (2019).

Article   ADS   CAS   Google Scholar  

Carlson, T. N. & Ripley, D. A. On the relation between NDVI, fractional vegetation cover, and leaf area index. Remote Sens. Environ. 62 , 241–252 (1997).

Gitelson, A. A. & Merzlyak, M. N. Remote sensing of chlorophyll concentration in higher plant leaves. Adv. Space Res. 22 , 689–692 (1998).

Tucker, C. J. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens. Environ. 8 , 127–150 (1979).

Rondeaux, G., Steven, M. & Baret, F. Optimization of soil-adjusted vegetation indices. Remote Sens. Environ. 55 , 95–107 (1996).

Birth, G. S. & McVey, G. R. Measuring the color of growing turf with a reflectance spectrophotometer 1. Agron. J. 60 , 640–643 (1968).

Huete, A. R. A soil-adjusted vegetation index (SAVI). Remote Sens. Environ. 25 , 295–309 (1988).

Taghizadeh-Mehrjardi, R., Minasny, B., Sarmadian, F. & Malone, B. P. Digital mapping of soil salinity in Ardakan region, central Iran. Geoderma 213 , 15–28. https://doi.org/10.1016/j.geoderma.2013.07.020 (2014).

Scudiero, E., Skaggs, T. H. & Corwin, D. L. Regional-scale soil salinity assessment using Landsat ETM + canopy reflectance. Remote Sens. Environ. 169 , 335–343. https://doi.org/10.1016/j.rse.2015.08.026 (2015).

Jordan, C. F. Derivation of leaf-area index from quality of light on the forest floor. Ecology 50 , 663–666 (1969).

Jiang, Z., Huete, A. R., Didan, K. & Miura, T. Development of a two-band enhanced vegetation index without a blue band. Remote Sens. Environ. 112 , 3833–3845 (2008).

Gitelson, A. A., Kaufman, Y. J. & Merzlyak, M. N. Use of a green channel in remote sensing of global vegetation from EOS-MODIS. Remote Sens. Environ. 58 , 289–298 (1996).

Shi, C. et al. Quantitative inversion of soil salinity and analysis of its spatial pattern in agricultural area in Shihezi of Xinjiang. Geogr. Res. 33 , 2135–2144 (2015).

Google Scholar  

Allbed, A., Kumar, L. & Aldakheel, Y. Y. Assessing soil salinity using soil salinity and vegetation indices derived from IKONOS high-spatial resolution imageries: applications in a date palm dominated region. Geoderma 230–231 , 1–8. https://doi.org/10.1016/j.geoderma.2014.03.025 (2014).

Wang, F., Shi, Z., Biswas, A., Yang, S. & Ding, J. Multi-algorithm comparison for predicting soil salinity. Geoderma 365 , 114211. https://doi.org/10.1016/j.geoderma.2020.114211 (2020).

Vermeulen, D. & Van Niekerk, A. Machine learning performance for predicting soil salinity using different combinations of geomorphometric covariates. Geoderma 299 , 1–12. https://doi.org/10.1016/j.geoderma.2017.03.013 (2017).

Breiman, L. Machine learning. Mach. Learn. 45 , 5–32 (2001).

Chen, S. et al. A high-resolution map of soil pH in China made by hybrid modelling of sparse soil data and environmental covariates and its implications for pollution. Sci. Total Environ. 655 , 273–283 (2019).

Altman, N. & Krzywinski, M. Ensemble methods: bagging and random forests. Nat. Methods 14 , 933–935 (2017).

Guan, Y., Grote, K., Schott, J. & Leverett, K. Prediction of soil water content and electrical conductivity using random forest methods with UAV multispectral and ground-coupled geophysical data. Remote Sens. 14 , 1023 (2022).

Breiman, L. Random forests. Mach. Learn. 45 , 5–32 (2001).

Article   MATH   Google Scholar  

Pedregosa, F. et al. Scikit-learn: machine learning in python. J. Mach. Learn. Res. 12 , 2825–2830 (2011).

MathSciNet   MATH   Google Scholar  

Zhang, Y., Liang, S., Zhu, Z., Ma, H. & He, T. Soil moisture content retrieval from Landsat 8 data using ensemble learning. ISPRS J. Photogram. Remote. Sens. 185 , 32–47. https://doi.org/10.1016/j.isprsjprs.2022.01.005 (2022).

Chen, T. & Guestrin, C. in the 22nd ACM SIGKDD International Conference.

Friedman, J. H. Greedy function approximation: a gradient boosting machine.. Ann. Statist. https://doi.org/10.1214/aos/1013203451 (2001).

Article   MathSciNet   MATH   Google Scholar  

Ke, G. et al. Lightgbm: a highly efficient gradient boosting decision tree. Adv. Neural Inform. Process. Syst. 30 (2017).

Su, H. et al. Super-resolution of subsurface temperature field from remote sensing observations based on machine learning. Int. J. Appl. Earth Observ. Geoinform. 102 , 102440. https://doi.org/10.1016/j.jag.2021.102440 (2021).

Sun, X., Liu, M. & Sima, Z. A novel cryptocurrency price trend forecasting model based on LightGBM. Financ. Res. Lett. 32 , 101084 (2020).

Kennedy, J. & Eberhart, R. in Proceedings of ICNN'95-International Conference on Neural Networks. 1942–1948 (IEEE).

Bellon-Maurel, V. & McBratney, A. Near-infrared (NIR) and mid-infrared (MIR) spectroscopic techniques for assessing the amount of carbon stock in soils–Critical review and research perspectives. Soil Biol. Biochem. 43 , 1398–1410 (2011).

Nocita, M. et al. Prediction of soil organic carbon content by diffuse reflectance spectroscopy using a local partial least square regression approach. Soil Biol. Biochem. 68 , 337–347 (2014).

Zovko, M. et al. A geostatistical Vis-NIR spectroscopy index to assess the incipient soil salinization in the Neretva River valley Croatia. Geoderma 332 , 60–72. https://doi.org/10.1016/j.geoderma.2018.07.005 (2018).

Ge, X. et al. Updated soil salinity with fine spatial resolution and high accuracy: the synergy of Sentinel-2 MSI, environmental covariates and hybrid machine learning approaches. Catena 212 , 106054. https://doi.org/10.1016/j.catena.2022.106054 (2022).

Wu, D., Jia, K., Zhang, X., Zhang, J. & Abd El-Hamid, H. T. Remote sensing inversion for simulation of soil salinization based on hyperspectral data and ground analysis in Yinchuan China. Nat. Resour. Res. 30 , 4641–4656 (2021).

Madani, A. A. Soil salinity detection and monitoring using landsat data: a case study from Siwa Oasis, Egypt. GISci. Remote Sens. 42 , 171–181. https://doi.org/10.2747/1548-1603.42.2.171 (2013).

Richards, L. A. Diagnosis and improvement of saline and alkali soils (Scientific Publishers, 2012).

Zeraatpisheh, M., Ayoubi, S., Jafari, A., Tajik, S. & Finke, P. Digital mapping of soil properties using multiple machine learning in a semi-arid region, central Iran. Geoderma 338 , 445–452 (2019).

Han, L., Liu, D., Cheng, G., Zhang, G. & Wang, L. Spatial distribution and genesis of salt on the saline playa at Qehan Lake, Inner Mongolia, China. Catena 177 , 22–30 (2019).

Zhang, T.-T. et al. Detecting soil salinity with MODIS time series VI data. Ecol. Ind. 52 , 480–489. https://doi.org/10.1016/j.ecolind.2015.01.004 (2015).

Wichelns, D. & Qadir, M. Achieving sustainable irrigation requires effective management of salts, soil salinity, and shallow groundwater. Agric. Water Manag. 157 , 31–38 (2015).

Download references

Acknowledgements

We greatly appreciate the anonymous reviewers and editors who evaluated our article and provided insightful feedback. This study was supported by the project of Natural Science Foundation of Xinjiang Uygur Autonomous Region (2019D01C024), the Xinjiang Uygur Autonomous Region Education Department Tianchi Doctoral Research Project (tcbs201816), and the Xinjiang University Doctoral Research Initiation Grant Program (BS180239).

Author information

Authors and affiliations.

College of Geography and Remote Sensing Sciences, Xinjiang University, No. 777 Huarui Street, Xinjiang, 830017, Urumqi, China

Shilong Ma, Baozhong He, Boqiang Xie, Xiangyu Ge & Lijing Han

Xinjiang Key Laboratory of Oasis Ecology, Xinjiang University, 830017, Urumqi, China

Key Laboratory of Smart City and Environment Modelling of Higher Education Institute, Xinjiang University, 830017, Urumqi, China

You can also search for this author in PubMed   Google Scholar

Contributions

S.M.: Conceptualization, Methodology, Investigation, Software, Formal, Writing-orginal draft, Formal analysis. B.H.: Investigation, Supervision, Writing - Review & Editing, Funding acquisition. B.X.: Investigation, Software. X.G.: Investiagtion, Formal analysis. L.H.: Investigation, Software.

Corresponding author

Correspondence to Baozhong He .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Ma, S., He, B., Xie, B. et al. Investigation of the spatial and temporal variation of soil salinity using Google Earth Engine: a case study at Werigan–Kuqa Oasis, West China. Sci Rep 13 , 2754 (2023). https://doi.org/10.1038/s41598-023-27760-8

Download citation

Received : 13 November 2022

Accepted : 06 January 2023

Published : 16 February 2023

DOI : https://doi.org/10.1038/s41598-023-27760-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

ORIGINAL RESEARCH article

Quality assessment of ecological environment based on google earth engine: a case study of the zhoushan islands.

• 1 Key Laboratory of Oceanographic Big Data Mining and Application of Zhejiang Province, Zhoushan, China
• 2 School of Information Engineering, Zhejiang Ocean University, Zhoushan, China
• 3 Marine Science and Technology College, Zhejiang Ocean University, Zhoushan, China
• 4 Beijing VMinFull Limited, Beijing, China

With the development of society, the impact of human activities on the ecological environment is becoming increasingly intense, so the dynamic monitoring of the status of the ecological environment is of great significance to the management and protection of urban ecology. As an objective and rapid ecological quality monitoring and evaluation technique, the remote sensing based ecological index (RSEI) has been widely used in the field of ecological research. Free available Landsat series data has the character of a long time series and high spatial resolution provides the possibility to conduct large-scale and long-term monitoring of ecological environment quality. Compared with traditional methods, the Google Earth Engine (GEE) platform can save a lot of time and energy in the data acquisition and preprocessing steps. To monitor the quality of the ecological environment in Zhoushan from 2000 to 2020, the GEE platform was used for cloud computing to obtain the RSEI, which can reflect the quality of the ecological environment. The results show that (1) from 2000 to 2020, the average RSEI value in Zhoushan Islands decreased from 0.748 to 0.681, indicating that the overall ecological environment exhibited a degradation trend. (2) From 2000 to 2020, the change in the area of each ecological environment level indicates that the quality of the ecological environment in Zhoushan Islands exhibited a degradation trend. The proportion of the area with an excellent eco-environment grade decreased by 13.54%, and the proportion of the area with poor and fair eco-environment grades increased by 3.43%.

Introduction

The environment is not only a basic condition for human existence and development but also an important cornerstone of sustainable social and economic development ( Chen et al., 2020a , b ; Jia H. et al., 2021 ; Nourani et al., 2021 ). Under global climate change and the intensification of human activites, many ecological problems have arisen, which have a significant impact on the ecosystems on which human beings depend for survival, resulting in the continuous decline of the restoration ability of the ecosystem ( Pekel et al., 2016 ; Hussain and Khan, 2020 ; Alqadhi et al., 2021 ; Chen et al., 2021a , b ). Therefore, developing a method for the dynamic monitoring and evaluation of the quality of the ecological environment has become an important issue in ecological research.

For the past few years, scholars have been exploring quantitative methods of evaluating the regional environment. Thus far, promulgated by China’s Ministry of Environmental Protection, the eco-environment index (EI) ( Chiabai et al., 2018 ), which is based on ecological environment evaluation specification, and the remote sensing based ecological index (RSEI) proposed by Xu et al. (2018) are the most widely used to evaluate the regional environment ( Hossain and Hashim, 2019 ; Bonney and He, 2021 ; Boori et al., 2021 ; Fu et al., 2021 ; Jia M. et al., 2021 ). The RSEI model has the advantages of easy parameter acquisition and a wide evaluation range, and it makes up for the index acquisition and analysis deficiencies of the EI so it is widely used in the evaluation of the quality of the ecological environment ( Berberoglu and Akin, 2009 ; Chiabai et al., 2018 ; Xu et al., 2018 ; Eveleth et al., 2021 ; Firozjaei et al., 2021a , b ). However, when applied to a large area, complex data processing and index calculation become a problem with the RSEI that cannot be ignored ( Wen et al., 2019 ; Duan et al., 2021 ; Sun et al., 2021 ).

With the development of remote sensing technology, it is widely used in land use and cover change, forestry resource survey, and ecological environment monitoring ( Yang et al., 2018 ; Chen et al., 2022 ). The Landsat series data meet the requirements of ecological environment quality monitoring due to its long time series and high spatial resolution ( Yang et al., 2022 ). In recent years, the long time series, large spatial scale, fast, and accurate calculation of remote sensing data impose higher requirements for software and hardware ( Chen et al., 2020a , b ; Xiong et al., 2021 ). However, the Google Earth Engine (GEE) platform is a special tool for the batch processing of satellite image data, and it can quickly process a large number of images ( Yang et al., 2020 ; Fan et al., 2021 ; Firozjaei et al., 2021b ; Jia H. et al., 2021 ; Nietupski et al., 2021 ). Compared with traditional methods, it can save a lot of time and energy in the data acquisition and preprocessing steps. Thanks to the powerful computing ability and cloud storage features of the GEE platform, environmental monitoring studies based on this platform have been continuously carried out in recent years ( Fan et al., 2021 ; Jia H. et al., 2021 ; Jia M. et al., 2021 ; Murayama et al., 2021 ).

The Zhoushan Islands are a typical archipelago region in China. As the first prefecture-level city established in the form of several islands, the city of Zhoushan has a relatively special geographic location. As it is naturally formed land areas surrounded by seawater and that remain above the water surface at high tide. Islands are very vulnerable to extreme weather or natural disasters, and thus, more attention should be paid to the protection of their ecological environment ( Gernez et al., 2021 ; He et al., 2021 ; Murayama et al., 2021 ). Due to the intensification of human activities in recent years, the development of these islands has become more intense. Therefore, the monitoring of the ecological environment in the Zhoushan Islands is particularly important. Using the GEE platform, Landsat Thematic Mapper/Operational Land Imager (TM/OLI) image data were acquired in this study, and the RSEI model was used to carry out ecological environmental quality monitoring in Zhoushan from 2000 to 2020. The results of this study provide a reference for the formulation of policies and measures related to ecological restoration and protection and play an important role in the sustainable development of the ecological environment in this region.

Materials and Methods

The Zhoushan Islands as shown in Figure 1 is located along the coast of Zhejiang Province. As the gateway to the Yangtze River valley, the developed Yangtze River Delta region gives this region significant resource advantages ( Chen et al., 2021a , 2022 ). The Zhoushan Islands is the first prefecture-level city established in the form of an archipelago in China ( Chen et al., 2021c ). It consists of 1,390 islands with an area of over 500 m 2 ( Wang et al., 2021a , b ).

Figure 1. Location of the study area.

Data Source

The Landsat data were obtained from the United States Geological Survey (USGS) and were integrated on the GEE platform with a spatial resolution of 30 m. In this study, the surface reflectance (SR) datasets obtained using the Landsat 5 TM sensors in 2000, 2005, and 2010 were used. The SR datasets of the Landsat 8 OLI/Thermal Infrared Sensor sensors obtained in 2015 and 2020 were selected using the GEE platform. The obtained images are all images synthesized from the median of summer images in each year and they have been preprocessed, including radiometric correction, atmospheric correction, and geometric precision correction. Then, the Landsat cloud mask algorithm ( Xiong et al., 2021 ) was applied in the GEE platform to conduct cloud removal from the obtained data. The data used in this study are listed in Table 1 .

Table 1. Description of data used by the study.

Based on the calculation formula for each component of the Landsat image proposed by Xu et al. (2018) , in this study, remote sensing monitoring of the quality of the ecological environment was carried out using the GEE platform ( Zhao et al., 2017 ). First, the Landsat satellite remote sensing data were acquired using the GEE platform, and cloud removal was conducted. Second, the normalized difference vegetation index (NDVI), humidity component (Wet), the normalized difference building-up and soil index (NDBSI), and land surface temperature (LST) were inverted to obtain the greenness, humidity, dryness, and heat indexes, and the results were subjected to standardized processing ( Xu et al., 2018 ). Third, we conducted principal component analysis (PCA) and obtained PC1, which was used for the construction of the RSEI. Finally, the ecological environmental quality of the study area was classified, and the characteristics of the changes in the ecological quality in Zhoushan over the past 20 years were analyzed.

(1) Cloud Mask Processing

In this study, Landsat_SR image data were selected with an interval of 5 years from 2000 to 2020 in the GEE. The cloud content of the Landsat_SR image in the GEE was marked in the “CLOUD_COVER” field ( Foga et al., 2017 ). This field was used to screen all images with cloud contents of less than 50% in the study area. The mask function established according to cloud shadow and cloud attribute fields contained in the quality evaluation band “pixel_qa” in the Landsat_SR dataset image removed the cloud-containing area in each image.

(2) Calculation and Normalization of the Component Index

Xu used four important indexes of the environment as evaluation indexes of the eco-environment to construct the RSEI, namely, the greenness, humidity, heat, and dryness ( Xu et al., 2018 ). In remote sensing, they are defined as the vegetation index, soil index, moisture component, and land surface temperature, respectively.

where NDVI is the normalized difference vegetation index, Wet is the wet component of the tasseled cap transformation, LST is the land surface temperature, and NDBSI is the normalized difference built-up and soil index. f indicates that the Remote sensing based ecological index can be expressed as a function of these four indicators.

Because each indicator has a different unit and value range, the four indicators need to be normalized separately using the following equation:

where NI i is the result of the normalization of the indicators, I i is the ith pixel value; I min is the minimum value, and I max is the maximum value ( Yi et al., 2018 ).

(3) Calculation of the Remote Sensing Based Ecological Index

The principal component transformation was used to construct the RSEI. The main information for the four indicators was mainly concentrated in the first principal component (PC1), which enables the RSEI to comprehensively reflect the information about the four indicators. PCA is a multi-dimensional data compression technique. This method rotates the coordinate axis vertically and concentrates the information about multiple variables into a few feature components through linear transformation ( Yi et al., 2018 ; Turpie et al., 2021 ). This method can avoid the deviation of subjective factors during the weight assignment process, which makes the RSEI more objective and reliable.

To make a large value of PC1 represent good ecological conditions, the first principal component of the function of these four indicators can be further subtracted from 1 to obtain the initial ecological index RSEI 0 , and the formula is as follows:

where RSEI 0 is normalized to facilitate the measurement and comparison of the indicators as follows:

The obtained RSEI f value is within the range of [0–1]. The closer RSEI is to 1, the better the quality of the eco-environment of the region ( Xu et al., 2018 ; Sekovski et al., 2020 ).

Results and Analysis

Results of principal component analysis.

The first to fourth principal component analysis index values can be expressed as PC1, PC2, PC3, and PC4, respectively, as shown in Table 2 .

Table 2. The results of the principal component analysis.

Table 2 shows the results of the principal component analysis from 2000 to 2020 in the study area. It can be seen from Table 2 that (1) The contribution rates of the first principal component from 2000 to 2020 all exceeded 70%. The contribution rates of PC1 in 2000, 2005, 2010, 2015, and 2020 were 71.81, 75.88, 77.71, 72.74, and 82.68%, respectively. (2) Compared with the other components, the first principal component contained more than 70% of the characteristic information about each indicator. Therefore, it can integrate the information about each indicator better, representing the characteristics of the regional ecological environment, and it can be used to establish the Remote sensing based ecological index.

Table 3 shows that the quality of the ecological environment in Zhoushan was generally good from 2000 to 2020, and the mean value of the Remote sensing based ecological index initially decreased and then slowly increased. The mean value of the Remote sensing based ecological index decreased from 0.748 in 2000 to 0.668 in 2010, and then, it remained basically unchanged until 2015 (0.666). Finally, it increased to 0.681 in 2020. This indicates that the eco-environmental quality in Zhoushan exhibited a slowly increasing trend after decreasing. During the study period, the standard deviation of the Remote sensing based ecological index was low, indicating a high degree of data concentration and that the research results are reliable.

Table 3. Mean values of normalized ecological environment factors in Zhoushan.

Analysis of Temporal Changes in the Quality of the Ecological Environment

To better analyze the quality of the ecological environment in Zhoushan, according to the Technical Specifications for the Assessment of the Ecological and Environmental Conditions issued in 2015 (HJ/T192-2006) ( Firozjaei et al., 2021a ), the ecological and environmental quality was divided into the following five grades with a 0.2 interval: poor [0–0.2], fair [0.2–0.4], moderate [0.4–0.6], good [0.6–0.8], and excellent [0.8–1]. The results are presented in Table 4 and Figure 2 .

Table 4. Statistics of ecological quality grade and area of Zhoushan.

Figure 2. Spatial distribution of RSEI in Zhoushan Islands during 2000–2020.

Figure 2 and Table 4 show the changes in the areas of the various eco-environmental grades in Zhoushan from 2000 to 2020. The results show that (1) from 2000 to 2020, the area with an excellent eco-environmental grade in Zhoushan decreased by 13.54%, and the area with poor and fair eco-environmental grades increased by 3.43%, so the overall quality of the eco-environment exhibited a degradation trend. (2) In 2000 and 2005, the quality of the eco-environment in Zhoushan remained basically stable, the overall quality of the eco-environment was mainly excellent, accounting for about 50% of the total area. In 2000 and 2005, the area with a good eco-environment grade accounted for 28.16 and 28.64%, respectively. The area with moderate eco-environmental quality increased from 11.97% in 2000 to 15.26% in 2005. In 2005, the area with poor and fair eco-environment grades accounted for 10.77%, an increase of 40.41 km 2 compared with 2000. (3) In 2010, the overall eco-environmental quality was predominantly excellent and good, accounting for 34.82 and 33.81% of the total area, respectively. Compared with 2005, the area with an excellent grade decreased by 132.21 km 2 . In addition, the area with poor and fair eco-environment grades accounted for 15.27%, an increase of 56.68 km 2 compared with 2005. This shows that from 2005 to 2010, the eco-environment exhibited a degradation trend. (4) In 2015, the area with poor and fair eco-environment grades decreased by 8.22%, indicating a significant increase in the eco-environmental quality compared with 2010. However, the area with an excellent eco-environment grade decreased by 107.68 km 2 from 2010 to 2015. (5) In 2020, the eco-environmental quality was predominantly excellent. Compared with 2010 and 2015, the area with an excellent eco-environment grade was significantly higher in 2020, accounting for 38.77%, while the area with a poor eco-environmental grade was significantly smaller. This shows that there was a trend of improvement in the ecological environment from 2010 to 2020.

In this study, the RSEI for Zhoushan from 2000 to 2020 was analyzed using the GEE platform. These results provide a decision-making basis for the sustainable development of the ecological environment in Zhoushan. The main conclusions are as follows.

(1) The mean value of the RSEI in Zhoushan initially decreased and then slowly increased from 2000 to 2020. The mean value of the RSEI decreased from 0.748 in 2000 to 0.668 in 2010, then increased to 0.681 in 2020.

(2) It can be concluded that the overall ecological environment in the region exhibited a degradation trend during the study period. The area with an excellent eco-environment grade decreased by 13.54%, and the area with poor and fair eco-environment grades increased by 3.43% from 2000 to 2020.

However, this study also needs to be extended in the following aspects. (1) The temporal and spatial evolution of the RSEI in the key areas of the Zhoushan Islands greatly affected by human activities should be analyzed to better reveal the changes in the RSEI in Zhoushan. (2) In order to improve the monitoring model of the quality of the ecological environment in Zhoushan, other indexes reflecting the quality of the ecological environment can be selected to add to the analysis in the future.

Data Availability Statement

The original contributions presented in this study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Author Contributions

All authors listed have made a substantial, direct, and intellectual contribution to the work, and approved it for publication.

This work was supported by the National Key Research and Development Program of China (No. 2018YFC1503204-04) and the National Natural Science Foundation of China (Grant No. 42171311), the Training Program of Excellent Master Thesis of Zhejiang Ocean University, and the Project of Beijing VMinFull Limited (VMF2021RS).

Conflict of Interest

BL was employed by Beijing VMinFull Limited.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Acknowledgments

We thank LetPub ( www.letpub.com ) for its linguistic assistance during the preparation of this manuscript.

Alqadhi, S., Mallick, J., Balha, A., Bindajam, A., Singh, C. K., and Hoa, P. V. (2021). Spatial and decadal prediction of land use/land cover using multi-layer perceptron-neural network (MLP-NN) algorithm for a semi-arid region of Asir. Saud. Arab. Earth Sci. Inform. 14, 1547–1562. doi: 10.1007/s12145-021-00633-2

CrossRef Full Text | Google Scholar

Berberoglu, S., and Akin, A. (2009). Assessing different remote sensing techniques to detect land use/cover changes in the eastern Mediterranean. Int. J. Appl. Earth Observ. Geoinform. 11, 46–53. doi: 10.1016/j.jag.2008.06.002

Bonney, M. T., and He, Y. (2021). Temporal connections between long-term Landsat time-series and tree-rings in an urban–rural temperate forest. Int. J. Appl. Earth Observ. Geoinform. 103:102523. doi: 10.1016/j.jag.2021.102523

Boori, M. S., Choudhary, K., Paringer, R., and Kupriyanov, A. (2021). Spatiotemporal ecological vulnerability analysis with statistical correlation based on satellite remote sensing in Samara, Russia. J. Environ. Manage. 285:112138. doi: 10.1016/j.jenvman.2021.112138

PubMed Abstract | CrossRef Full Text | Google Scholar

Chen, C., Chen, H., Liao, W., Sui, X., Wang, L., Chen, J., et al. (2020a). Dynamic monitoring and analysis of land-use and land-cover change using Landsat multitemporal data in the Zhoushan archipelago. China. IEEE Access 8, 210360–210369. doi: 10.1109/ACCESS.2020.3036128

Chen, C., He, X., Lu, Y., and Chu, Y. (2020b). Application of Landsat time-series data in island ecological environment monitoring: a case study of Zhoushan Islands, China. J. Coast. Res. 108, 193–199. doi: 10.2112/JCR-SI108-038.1

Chen, C., Liang, J., Xie, F., Hu, Z., Sun, W., Yang, G., et al. (2022). Temporal and spatial variation of coastline using remote sensing images for Zhoushan archipelago, China. Int. J. Appl. Earth Observ. Geoinform. 107:102711. doi: 10.1016/j.jag.2022.102711

Chen, C., Wang, L., Chen, J., Liu, Z., Liu, Y., and Chu, Y. (2021a). A seamless economical feature extraction method using Landsat time series data. Earth Sci. Inform. 14, 321–332. doi: 10.1007/s12145-020-00564-4

Chen, C., Wang, L., Zhang, Z., Lu, C., Chen, H., and Chen, J. (2021b). Construction and application of quality evaluation index system for remote-sensing image fusion. J. Appl. Remote Sensing 16:012006. doi: 10.1117/1.JRS.16.012006

Chen, H., Chen, C., Zhang, Z., Lu, C., Wang, L., He, X., et al. (2021c). Changes of the spatial and temporal characteristics of land-use landscape patterns using multi-temporal Landsat satellite data: a case study of Zhoushan Island, China. Ocean Coast. Manage. 213:105842. doi: 10.1016/j.ocecoaman.2021.105842

Chiabai, A., Quiroga, S., Martinez-Juarez, P., Higgins, S., and Taylor, T. (2018). The nexus between climate change, ecosystem services and human health: towards a conceptual framework. Sci. Total Environ. 635, 1191–1204. doi: 10.1016/j.scitotenv.2018.03.323

Duan, Y., Tian, B., Li, X., Liu, D., Sengupta, D., Wang, Y., et al. (2021). Tracking changes in aquaculture ponds on the China coast using 30 years of Landsat images. Int. J. Appl. Earth Observ. Geoinform. 102:102383. doi: 10.1016/j.jag.2021.102383

Eveleth, R., Glover, D. M., Long, M. C., Lima, I. D., Chase, A. P., and Doney, S. C. (2021). Assessing the Skill of a High-Resolution Marine Biophysical Model Using Geostatistical Analysis of Mesoscale Ocean Chlorophyll Variability From Field Observations and Remote Sensing. Front. Mar. Sci. 8:612764. doi: 10.3389/fmars.2021.612764

Fan, X., Nie, G., Xia, C., and Zhou, J. (2021). Estimation of pixel-level seismic vulnerability of the building environment based on mid-resolution optical remote sensing images. Int. J. Appl. Earth Observ. Geoinform. 101:102339. doi: 10.1016/j.jag.2021.102339

Firozjaei, M. K., Fathololoumi, S., Kiavarz, M., Biswas, A., Homaee, M., and Alavipanah, S. K. (2021a). Land Surface Ecological Status Composition Index (LSESCI): a novel remote sensing-based technique for modeling land surface ecological status. Ecol. Indicat. 123:107375. doi: 10.1016/j.ecolind.2021.107375

Firozjaei, M. K., Sedighi, A., Firozjaei, H. K., Kiavarz, M., Homaee, M., Arsanjani, J. J., et al. (2021b). A historical and future impact assessment of mining activities on surface biophysical characteristics change: a remote sensing-based approach. Ecol. Indicat. 122:107264. doi: 10.1016/j.ecolind.2020.107264

Foga, S., Scaramuzza, P. L., Guo, S., Zhu, Z., Dilley, R. D., Beckmann, T., et al. (2017). Cloud detection algorithm comparison and validation for operational Landsat data products. Remote Sensing Environ. 194, 379–390. doi: 10.1016/j.rse.2017.03.026

Fu, T., Zhang, L., Yuan, X., Chen, B., and Yan, M. (2021). Spatio-temporal patterns and sustainable development of coastal aquaculture in Hainan Island, China: 30 Years of evidence from remote sensing. Ocean Coast. Manage. 214:105897. doi: 10.1016/j.ocecoaman.2021.105897

Gernez, P., Palmer, S. C., Thomas, Y., and Forster, R. (2021). Remote sensing for aquaculture. Front. Mar. Sci. 7:1258. doi: 10.3389/fmars.2020.638156

He, X., Chen, C., Zhang, Z., Hu, H., Tan, A., and Xing, Z. (2021). Temporal and spatial characteristics of harmful algal blooms in the offshore waters, China during 1990 to 2019. J. Appl. Remote Sensing 16:012004. doi: 10.1117/1.JRS.16.012004

Hossain, M. S., and Hashim, M. (2019). Potential of Earth Observation (EO) technologies for seagrass ecosystem service assessments. Int. J. Appl. Earth Observ. Geoinform. 77, 15–29. doi: 10.1016/j.jag.2018.12.009

Hussain, D., and Khan, A. A. (2020). Machine learning techniques for monthly river flow forecasting of Hunza River. Pakistan. Earth Sci. Inform. 13, 939–949. doi: 10.1007/s12145-020-00450-z

Jia, H., Yan, C., and Xing, X. (2021). Evaluation of Eco-Environmental Quality in Qaidam Basin Based on the Ecological Index (MRSEI) and GEE. Remote Sensing 13:4543. doi: 10.3390/rs13224543

Jia, M., Wang, Z., Mao, D., Ren, C., Wang, C., and Wang, Y. (2021). Rapid, robust, and automated mapping of tidal flats in China using time series Sentinel-2 images and Google Earth Engine. Remote Sensing Environ. 255:112285. doi: 10.1016/j.rse.2021.112285

Murayama, Y., Simwanda, M., and Ranagalage, M. (2021). Spatiotemporal Analysis of Urbanization Using GIS and Remote Sensing in Developing Countries. Sustainability 13:3681. doi: 10.3390/su13073681

Nietupski, T. C., Kennedy, R. E., Temesgen, H., and Kerns, B. K. (2021). Spatiotemporal image fusion in Google Earth Engine for annual estimates of land surface phenology in a heterogenous landscape. Int. J. Appl. Earth Observ. Geoinform. 99:102323. doi: 10.1016/j.jag.2021.102323

Nourani, V., Foroumandi, E., Sharghi, E., and Dąbrowska, D. (2021). Ecological-environmental quality estimation using remote sensing and combined artificial intelligence techniques. J. Hydroinform. 23, 47–65. doi: 10.2166/hydro.2020.048

Pekel, J. F., Cottam, A., Gorelick, N., and Belward, A. S. (2016). High-resolution mapping of global surface water and its long-term changes. Nature 540, 418–422. doi: 10.1038/nature20584

Sekovski, I., Del Río, L., and Armaroli, C. (2020). Development of a coastal vulnerability index using analytical hierarchy process and application to Ravenna province (Italy). Ocean Coast. Manage. 183:104982. doi: 10.1016/j.ocecoaman.2019.104982

Sun, W., Liu, K., Ren, G., Liu, W., Yang, G., Meng, X., et al. (2021). A simple and effective spectral-spatial method for mapping large-scale coastal wetlands using China ZY1-02D satellite hyperspectral images. Int. J. Appl. Earth Observ. Geoinform. 104:102572. doi: 10.1016/j.jag.2021.102572

Turpie, K. R., Ackleson, S. G., Byrd, K. B., and Moisan, T. A. (2021). Science and Applications of Coastal Remote Sensing. Front. Mar. Sci. 8:641029. doi: 10.3389/fmars.2021.641029

Wang, L., Chen, C., Xie, F., Hu, Z. J., Zhang, Z., Chen, H. X., et al. (2021a). Estimation of the value of regional ecosystem services of an archipelago using satellite remote sensing technology: a case study of Zhoushan Archipelago, China. Int. J. Appl. Earth Observ. Geoinform. 105:102616. doi: 10.1016/j.jag.2021.102616

Wang, L., Chen, C., Zhang, Z., Gan, W., Yu, J., and Chen, H. (2021b). Approach for estimation of ecosystem services value using multitemporal remote sensing images. J. Appl. Remote Sensing 16:012010. doi: 10.1117/1.JRS.16.012010

Wen, X., Ming, Y., Gao, Y., and Hu, X. (2019). Dynamic monitoring and analysis of ecological quality of pingtan comprehensive experimental zone, a new type of sea island city, based on RSEI. Sustainability 12:21. doi: 10.3390/su12010021

Xiong, Y., Xu, W., Lu, N., Huang, S., Wu, C., Wang, L., et al. (2021). Assessment of spatial–temporal changes of ecological environment quality based on RSEI and GEE: a case study in Erhai Lake Basin. Yunnan province, China . Ecol. Indicat. 125:107518. doi: 10.1016/j.ecolind.2021.107518

Xu, H., Wang, M., Shi, T., Guan, H., Fang, C., and Lin, Z. (2018). Prediction of ecological effects of potential population and impervious surface increases using a remote sensing based ecological index (RSEI). Ecol. Indicat. 93, 730–740. doi: 10.1016/j.ecolind.2018.05.055

Yang, X., Qin, Q., Grussenmeyer, P., and Koehl, M. (2018). Urban surface water body detection with suppressed built-up noise based on water indices from Sentinel-2 MSI imagery. Remote Sensing Environ. 219, 259–270. doi: 10.1016/j.rse.2018.09.016

Yang, X., Qin, Q., Yésou, H., Ledauphin, T., Koehl, M., Grussenmeyer, P., et al. (2020). Monthly estimation of the surface water extent in France at a 10-m resolution using Sentinel-2 data. Remote Sensing Environ. 244:111803. doi: 10.1016/j.rse.2020.111803

Yang, X., Zhu, Z., Qiu, S., Kroeger, K., Zhu, Z., and Covington, S. (2022). Detection and characterization of coastal tidal wetland change in the northeastern US using Landsat time series. Remote Sensing Environ. 276:113047. doi: 10.1016/j.rse.2022.113047

Yi, L., Chen, J., Jin, Z., Quan, Y., Han, P., Guan, S., et al. (2018). Impacts of human activities on coastal ecological environment during the rapid urbanization process in Shenzhen. Chin. Ocean Coast. Manage. 154, 121–132. doi: 10.1016/j.ocecoaman.2018.01.005

Zhao, X., Wang, P., Chen, C., Jiang, T., Yu, Z., and Guo, B. (2017). Waterbody information extraction from remote-sensing images after disasters based on spectral information and characteristic knowledge. Int. J. Remote Sensing 38, 1404–1422. doi: 10.1080/01431161.2016.1278284

Keywords : ecological environmental quality, RSEI, Google Earth Engine, Zhoushan Islands, cloud computing

Citation: Liu Z, Wang L and Li B (2022) Quality Assessment of Ecological Environment Based on Google Earth Engine: A Case Study of the Zhoushan Islands. Front. Ecol. Evol. 10:918756. doi: 10.3389/fevo.2022.918756

Received: 13 April 2022; Accepted: 20 May 2022; Published: 09 June 2022.

Reviewed by:

Copyright © 2022 Liu, Wang and Li. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Liyan Wang, [email protected]

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Case Studies with Google Earth

“Some people look for a beautiful place. Others make a place beautiful.”

– Hazrat Inayat Khan

Geography Case Studies with Google Earth

Getting the ‘death by powerpoint’ feeling? Create a detailed sense of place with your case study by taking a virtual tour using Google Earth .

This page was created in the hope that Geography teachers can have access to a plethora of tours, ready to use when teaching case studies.

Please consider helping this resource grow by clicking the ‘Contact & Contribute’ button and sharing a Google Earth tour of your own.

Flying into case study locations… pic.twitter.com/ujy59kt5Ck — Andy Funnell (@GeogAndy) February 20, 2021

“Oven Ready” Case Studies

Clicking on each of the links below will take you to the case study in Google Earth. You don’t need to have a Google account to view, but you will need to sign into an account to be able to edit, copy & share your own version.

When the case study has loaded, click on the three dots (‘more actions’) and then either ‘ copy project ‘ or ‘ export as KML file .’

Want all of them? No problem! – Click here to get access to a folder of downloadable KML files.

Keep checking this list as more will be added:

• Thanks to Adam Corbridge ( @MrCorbridgeGeog )
• 12 Rivers / 12 Deltas – 12 Major Deltas and their Locations
• Thanks to Alasdair Monteith ( al_monteith )
• Bangladesh – Causes of LEDC/LIC Flooding
• Belfast – Urban Transect (Virtual Fieldwork)
• Belfast – Segregation & Multiculturalism
• Thanks to Miss Beggs ( @beggs_miss )
• Thanks to Jonny Rock ( @jonnyrock4 )
• British Isles – A Quick Tour
• Thanks to Nick Mahy ( @MahyNick )
• Chernobyl – The Geography of the disaster
• Thanks to Alistair Hamill ( @lcgeography )
• Thanks to M. Cuthbertson
• Thanks to Luke Tayler ( @geotayler )
• Alistair Hamill ( @lcgeography )
• Holderness Coast – From Flamborough Head to Spurn Point
• Thanks to Robert Frost
• Nick Mahy ( @MahyNick )
• Thanks to Anna Pennington ( @GeogAnna )
• Jonny Rock ( @jonnyrock4 )
• Mississippi River, USA – Management: Hard & Soft Engineering
• Thanks to Ross Wales ( @RossWales7 )
• Rio de Janeiro – Favelas / Shanty Towns / Slums
• Somerset Levels – Causes of MEDC/HIC Flooding
• Thanks to Luke McHale ( @mrlukemchale )
• Thanks to Alice McCaughern ( @TeachWithAlice )
• Thanks to Tom Hanson ( @hano_teach ) & Sam Hunak ( @gbrSam )
• Thanks to Teresa Hughes
• Thanks to Rachael Norris ( @geo_getters84 )
• Alice McCaughern ( @TeachWithAlice )
• Tokyo – Urban Heat Islands ( Google Earth Voyager )
• Top 10: Tallest Mountains on Earth – Can you Name Them?
• Top 21: Most Visited Landmarks on Earth – Can you Name Them?

34 year timelapse of an oxbow lake forming on the Ganges River. Link to @googleearth Engine Timelapse: https://t.co/UwPOPz3J1K #geographyteacher pic.twitter.com/uBZW3YXKTM — Andy Funnell (@GeogAndy) March 22, 2021

This mini-site has been designed with the vision of establishing a bank of Google Earth tours to assist in giving students a sense of place when they learn their case studies in Geography.

More Background

Get In Touch

[email protected] Twitter: @GeogAndy

Share a Case Study of Your Own

Share this:

• Already have a WordPress.com account? Log in now.
• Subscribe Subscribed
• Copy shortlink
• Report this content
• View post in Reader
• Manage subscriptions
• Collapse this bar

We apologize for the inconvenience...

To ensure we keep this website safe, please can you confirm you are a human by ticking the box below.

If you are unable to complete the above request please contact us using the below link, providing a screenshot of your experience.

https://ioppublishing.org/contacts/

Quality Assessment of Ecological Environment Based on Google Earth Engine: A Case Study of the Zhoushan Islands

• This person is not on ResearchGate, or hasn't claimed this research yet.

• Beijing VMinfull Limited

Abstract and Figures

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations

• Xiangwen Cai
• Xianglin Tang
• Xinqin Yong
• Shenghong Mao

• Kristofor R. Brye
• Lawson Connor
• Chunming Hu
• Xuefeng Peng

• Peihao Peng

• Zhanzhong Tang
• Linlin Jiang

• Jintao Liang
• Zhisong Liu
• Jianyu Chen
• Naichang Zhang
• Yongxiang Cao
• Xingcheng Wang

• Jiaqi Xiong
• REMOTE SENS ENVIRON

• INT J APPL EARTH OBS

• Guangbo Ren

• OCEAN COAST MANAGE

• J APPL REMOTE SENS
• Huixin Chen
• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

IEEE Account

• Change Username/Password
• Update Address

Purchase Details

• Payment Options
• Order History
• View Purchased Documents

Profile Information

• Communications Preferences
• Profession and Education
• Technical Interests
• US & Canada: +1 800 678 4333
• Worldwide: +1 732 981 0060
• Contact & Support
• About IEEE Xplore
• Accessibility
• Terms of Use
• Nondiscrimination Policy
• Privacy & Opting Out of Cookies

A not-for-profit organization, IEEE is the world's largest technical professional organization dedicated to advancing technology for the benefit of humanity. © Copyright 2024 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.

Information

• Author Services

Initiatives

You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.

All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .

Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.

Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.

Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.

Original Submission Date Received: .

• Active Journals
• Find a Journal
• Proceedings Series
• For Authors
• For Reviewers
• For Editors
• For Librarians
• For Publishers
• For Societies
• For Conference Organizers
• Open Access Policy
• Institutional Open Access Program
• Special Issues Guidelines
• Editorial Process
• Research and Publication Ethics
• Article Processing Charges
• Testimonials
• Preprints.org
• SciProfiles
• Encyclopedia

Article Menu

• Subscribe SciFeed
• Recommended Articles
• Google Scholar
• on Google Scholar
• Table of Contents

Find support for a specific problem in the support section of our website.

Please let us know what you think of our products and services.

Visit our dedicated information section to learn more about MDPI.

JSmol Viewer

Simulation of land use/land cover dynamics using google earth data and qgis: a case study on outer ring road, southern india.

1. Introduction

2. study site and experimental design, 2.1. segment classification, 2.2. preparation of time-series (land use/land cover) maps, 2.3. change detection analysis, 2.4. prediction of future land use/landover maps, 3. results and discussion, 3.1. chikkarayapuram segment, 3.2. nazarathpettai segment, 3.3. meppur segment, 3.4. perungalathur segment, 3.5. analysis of predicted output, 3.5.1. chikkarayapuram segment, 3.5.2. nazarathpettai segment, 3.5.3. meppur segment, 3.5.4. perungalathur segment, 4. conclusions.

• This work demonstrates the ability of remote sensing and GIS in capturing spatial-temporal capacity of datasets to analyze and predict the growth aspects and its impacts.
• Most of the landscape of the outer ring road has undergone a transition due to the anthropogenic and developmental activities over the past decade. It is well known that the stretch is important for residential and migratory purpose because it connects the major national highways around Chennai metropolitan region.
• From the above analysis, it is observed that due to the existing urban sprawl, the number of land use and land cover classes which currently exist will be decreased from seven classes to two or three classes. This may increase the future demand of natural resources, hence a consistent and proactive decision system needs to be created for managing the resource distribution.
• Based on Figure 16 , it could be inferred that the industrial and residential classes have been predicted to increase in comparison with the other classes in the four regions of interest. Further, the Perungalathur region has shown a contrasting variation where industrial zones increase with decrease in water body class. Besides, from Figure 17 , the spread of area among the study sites have been depicted.
• The primary socio-economic function of each site is reported by the categorical regionalized variable termed as ‘land use’ where the function is inferred from the pattern of land use. However, uncertainty in land use data arise in case of unreliable positional and categorical data. Effective utilization of metadata will reduce the issues surrounding the accuracy of the prepared land use data [ 37 , 38 ].
• An enhanced impact analysis system, which has been set up in recent times for bringing a balance in the urban growth and natural landscape of the region, needs to be monitored in a consistent manner.
• Besides improving the accuracy of mapping and avoiding the limitations caused due to the positional accuracy of existing images, change detection studies are planned to be carried out using Google Earth Engine (GEE) which caters to historical data of high resolution satellite imagery. Instead of digitization, automated algorithms of object-oriented models are being utilized in the GEE interface for future analysis.

Author Contributions

Institutional review board statement, informed consent statement, data availability statement, conflicts of interest, abbreviations.

• Turner, M.G. Spatial and Temporal Analysis of Landscape Patterns. Landsc. Ecol. 1990 , 4 , 21–30. [ Google Scholar ] [ CrossRef ]
• Aithal, D.B.; Ramachandra, T. V Visualization of Urban Growth Pattern in Chennai Using Geoinformatics and Spatial Metrics. J. Indian Soc. Remote Sens. 2016 , 44 , 617–633. [ Google Scholar ] [ CrossRef ]
• Hegazy, I.R.; Kaloop, M.R. Monitoring Urban Growth and Land Use Change Detection with GIS and Remote Sensing Techniques in Daqahlia Governorate Egypt. Int. J. Sustain. Built Environ. 2015 , 4 , 117–124. [ Google Scholar ] [ CrossRef ] [ Green Version ]
• Singh, P.; Kikon, N.; Verma, P. Impact of Land Use Change and Urbanization on Urban Heat Island in Lucknow City, Central India. A Remote Sensing Based Estimate. Sustain. Cities Soc. 2017 , 32 , 100–114. [ Google Scholar ] [ CrossRef ]
• Rahaman, S.; Jahangir, S.; Haque, M.S.; Chen, R.; Kumar, P. Spatio-Temporal Changes of Green Spaces and Their Impact on Urban Environment of Mumbai, India. Environ. Dev. Sustain. 2021 , 23 , 6481–6501. [ Google Scholar ] [ CrossRef ]
• Vani, M.; Prasad, P.R.C. Assessment of Spatio-Temporal Changes in Land Use and Land Cover, Urban Sprawl, and Land Surface Temperature in and around Vijayawada City, India. Environ. Dev. Sustain. 2020 , 22 , 3079–3095. [ Google Scholar ] [ CrossRef ]
• Ohri, A.; Yadav, P. Urban Sprawl Mapping and Land Use Change Detection Using Remote Sensing and GIS. Int. J. Remote Sens. GIS 2012 , 1 , 12–25. [ Google Scholar ]
• Jhawar, M.; Tyagi, N.; Dasgupta, V. Urban Planning Using Remote Sensing. Int. J. Innov. Res. Sci. Eng. Technol. 2013 , 1 , 42–57. [ Google Scholar ]
• Mundia, C.N.; Aniya, M. Analysis of Land Use/Cover Changes and Urban Expansion of Nairobi City Using Remote Sensing and GIS. Int. J. Remote Sens. 2005 , 26 , 2831–2849. [ Google Scholar ] [ CrossRef ]
• Kalyani, P.; Govindarajulu, P. A Multi-Scale Urban Analysis Using Remote Sensing and GIS. Geoinformatica 2013 , 3 , 373–379. [ Google Scholar ]
• Alves, D.S.; Skole, D.L. Characterizing Land Cover Dynamics Using Multi-Temporal Imagery. Int. J. Remote Sens. 1996 , 17 , 835–839. [ Google Scholar ] [ CrossRef ] [ Green Version ]
• Chen, J. Improving Urban Planning by Integrated Utilization of Remote Sensing and GIS’s. Int. Arch. Photogramm. Remote Sens. 1993 , 29 , 598. [ Google Scholar ]
• Belal, A.A.; Moghanm, F.S. Detecting Urban Growth Using Remote Sensing and GIS Techniques in Al Gharbiya Governorate, Egypt. Egypt. J. Remote Sens. Space Sci. 2011 , 14 , 73–79. [ Google Scholar ] [ CrossRef ] [ Green Version ]
• Rai, P.; V.K.Kumra, P. Role of Geoinformatics in Urban Planning. J. Sci. Res. 2011 , 55 , 11–24. [ Google Scholar ]
• Naik, M.G. Urban Sprawl of Bhubaneswar City Using GIS Applications and Entropy. UG. Dissertation/Thesis, National Institute of Technology, Rourkela, India, 2013. [ Google Scholar ]
• Sarif, M.O.; Gupta, R.D. Spatiotemporal Mapping of Land Use/Land Cover Dynamics Using Remote Sensing and GIS Approach: A Case Study of Prayagraj City, India (1988–2018). Environ. Dev. Sustain. 2022 , 24 , 888–920. [ Google Scholar ] [ CrossRef ]
• MohanRajan, S.N.; Loganathan, A.; Manoharan, P. Survey on Land Use/Land Cover (LU/LC) Change Analysis in Remote Sensing and GIS Environment: Techniques and Challenges. Environ. Sci. Pollut. Res. 2020 , 27 , 29900–29926. [ Google Scholar ] [ CrossRef ]
• Khan, A.; Sudheer, M. Machine Learning-Based Monitoring and Modeling for Spatio-Temporal Urban Growth of Islamabad. Egypt. J. Remote Sens. Space Sci. 2022 , 25 , 541–550. [ Google Scholar ] [ CrossRef ]
• Rimal, B.; Zhang, L.; Keshtkar, H.; Haack, B.N.; Rijal, S.; Zhang, P. Land Use/Land Cover Dynamics and Modeling of Urban Land Expansion by the Integration of Cellular Automata and Markov Chain. ISPRS Int. J. Geo-Inf. 2018 , 7 , 154. [ Google Scholar ] [ CrossRef ] [ Green Version ]
• Taylor, J.R.; Lovell, S.T. Mapping Public and Private Spaces of Urban Agriculture in Chicago through the Analysis of High-Resolution Aerial Images in Google Earth. Landsc. Urban Plan. 2012 , 108 , 57–70. [ Google Scholar ] [ CrossRef ]
• Malarvizhi, K.; Kumar, S.V.; Porchelvan, P. Use of High Resolution Google Earth Satellite Imagery in Landuse Map Preparation for Urban Related Applications. Procedia Technol. 2016 , 24 , 1835–1842. [ Google Scholar ] [ CrossRef ]
• Hu, Q.; Wu, W.; Xia, T.; Yu, Q.; Yang, P.; Li, Z.; Song, Q. Exploring the Use of Google Earth Imagery and Object-Based Methods in Land Use/Cover Mapping. Remote Sens. 2013 , 5 , 6026–6042. [ Google Scholar ] [ CrossRef ] [ Green Version ]
• Sathunuri, K.; Kumar, R.; Gogoi, K. Land Use/Land Cover Dynamics During 2001 And 2021 Using Google Earth Engine and GIS in Ramagundam Coal Mining Area, A Part of Pranhita Godavari Valley, Southern India. J. Sci. Res. 2022 , 66 , 63–68. [ Google Scholar ] [ CrossRef ]
• Muhammad, R.; Zhang, W.; Abbas, Z.; Guo, F.; Gwiazdzinski, L. Spatiotemporal Change Analysis and Prediction of Future Land Use and Land Cover Changes Using QGIS MOLUSCE Plugin and Remote Sensing Big Data: A Case Study of Linyi, China. Land 2022 , 11 , 419. [ Google Scholar ] [ CrossRef ]
• Kamaraj, M.; Rangarajan, S. Predicting the Future Land Use and Land Cover Changes for Bhavani Basin, Tamil Nadu, India, Using QGIS MOLUSCE Plugin. Environ. Sci. Pollut. Res. 2022 , 29 , 86337–86348. [ Google Scholar ] [ CrossRef ]
• Ning, J.; Liu, J.; Kuang, W.; Xu, X.; Zhang, S.; Yan, C.; Li, R.; Wu, S.; Hu, Y.; Du, G.; et al. Spatiotemporal Patterns and Characteristics of Land-Use Change in China during 2010–2015. J. Geogr. Sci. 2018 , 28 , 547–562. [ Google Scholar ] [ CrossRef ]
• Naikoo, M.W.; Rihan, M.; Ishtiaque, M. Shahfahad Analyses of Land Use Land Cover (LULC) Change and Built-up Expansion in the Suburb of a Metropolitan City: Spatio-Temporal Analysis of Delhi NCR Using Landsat Datasets. J. Urban Manag. 2020 , 9 , 347–359. [ Google Scholar ] [ CrossRef ]
• Appiah, D.O.; Schröder, D.; Forkuo, E.K.; Bugri, J.T. Application of Geo-Information Techniques in Land Use and Land Cover Change Analysis in a Peri-Urban District of Ghana. ISPRS Int. J. Geo-Inf. 2015 , 4 , 1265–1289. [ Google Scholar ] [ CrossRef ] [ Green Version ]
• Chatterjee, A.; Chattopadhyay, R.N. Metropolitan Regions of India: Rapid Growth, Changing Landuse and Challenges. In Satellite Towns in Neo-metropolitan Development in India ; Springer: Berlin/Heidelberg, Germany, 2020; pp. 87–100. [ Google Scholar ]
• Sekar, S.P.; Kanchanamala, S. An Analysis of Growth Dynamics in Chennai Metropolitan Area. Inst. Town Planners 2011 , 8 , 31–57. [ Google Scholar ]
• Rajendran, V.; Kaneda, T. A Simulation of Land Use/Cover Change for Urbanization on Chennai Metropolitan Area, India. In Proceedings of the REAL CORP 2014—PLAN IT SMART! Clever Solutions for Smart Cities. 19th International Conference on Urban Planning, Regional Development and Information Society, Vienna, Austria, 21–23 May 2014; pp. 25–34. [ Google Scholar ]
• Chanu, C.S.; Elango, L.; Shankar, G.R. A Geospatial Approach for Assessing the Relation between Changing Land Use/Land Cover and Environmental Parameters Including Land Surface Temperature of Chennai Metropolitan City, India. Arab. J. Geosci. 2021 , 14 , 132. [ Google Scholar ] [ CrossRef ]
• Lilly Rose, A.; Devadas, M.D. Analysis of Land Surface Temperature and Land Use/Land Cover Types Using Remote Sensing Imagery-a Case in Chennai City, India. In Proceedings of the 7th International Conference on Urban Climate, Yokohama, Japan, 29 June–3 July 2009; Volume 29. [ Google Scholar ]
• Wibowo, A.; Osman Salleh, K.; Frans, F.T.; Semedi, J.M. Spatial Temporal Land Use Change Detection Using Google Earth Data. IOP Conf. Ser. Earth Environ. Sci. 2016 , 47 , 12031. [ Google Scholar ] [ CrossRef ]
• Zurqani, H.A.; Post, C.J.; Mikhailova, E.A.; Schlautman, M.A.; Sharp, J.L. Geospatial Analysis of Land Use Change in the Savannah River Basin Using Google Earth Engine. Int. J. Appl. Earth Obs. Geoinf. 2018 , 69 , 175–185. [ Google Scholar ] [ CrossRef ]
• Floreano, I.X.; de Moraes, L.A.F. Land Use/Land Cover (LULC) Analysis (2009–2019) with Google Earth Engine and 2030 Prediction Using Markov-CA in the Rondônia State, Brazil. Environ. Monit. Assess. 2021 , 193 , 239. [ Google Scholar ] [ CrossRef ]
• Castilla, G.; Hay, G. Uncertainties in Land Use Data. Hydrol. Earth Syst. Sci. 2007 , 11 , 1857–1868. [ Google Scholar ] [ CrossRef ] [ Green Version ]
• Congalton, R.; Gu, J.; Yadav, K.; Ozdogan, M. Global Land Cover Mapping: A Review and Uncertainty Analysis. Remote Sens. 2014 , 6 , 12070. [ Google Scholar ] [ CrossRef ]

Click here to enlarge figure

Share and Cite

Padma, S.; Vidhya Lakshmi, S.; Prakash, R.; Srividhya, S.; Sivakumar, A.A.; Divyah, N.; Canales, C.; Saavedra Flores, E.I. Simulation of Land Use/Land Cover Dynamics Using Google Earth Data and QGIS: A Case Study on Outer Ring Road, Southern India. Sustainability 2022 , 14 , 16373. https://doi.org/10.3390/su142416373

Padma S, Vidhya Lakshmi S, Prakash R, Srividhya S, Sivakumar AA, Divyah N, Canales C, Saavedra Flores EI. Simulation of Land Use/Land Cover Dynamics Using Google Earth Data and QGIS: A Case Study on Outer Ring Road, Southern India. Sustainability . 2022; 14(24):16373. https://doi.org/10.3390/su142416373

Padma, SrinivasaPerumal, Sivakumar Vidhya Lakshmi, Ramaiah Prakash, Sundaresan Srividhya, Aburpa Avanachari Sivakumar, Nagarajan Divyah, Cristian Canales, and Erick I. Saavedra Flores. 2022. "Simulation of Land Use/Land Cover Dynamics Using Google Earth Data and QGIS: A Case Study on Outer Ring Road, Southern India" Sustainability 14, no. 24: 16373. https://doi.org/10.3390/su142416373

Article Metrics

Article access statistics, further information, mdpi initiatives, follow mdpi.

Subscribe to receive issue release notifications and newsletters from MDPI journals