Monitoring of mangrove forests vegetation based on optical versus microwave data: A case study western coast of Saudi Arabia

Mohammed H. Aljahdali; Baradin Adisu; Esubalew Adem; Anis Chaabani; Silvena Boteva; Lifu Zhang; Mohamed Elhag

doi:10.1515/geo-2022-0573

Article Open Access

Monitoring of mangrove forests vegetation based on optical versus microwave data: A case study western coast of Saudi Arabia

Mohammed H. Aljahdali , Baradin Adisu , Esubalew Adem , Anis Chaabani , Silvena Boteva , Lifu Zhang and Mohamed Elhag

Published/Copyright: February 15, 2024

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information

From the journal Open Geosciences Volume 16 Issue 1

Abstract

Normalized difference vegetation index (NDVI) is one of the parameters of vegetation that can be studied by remote sensing of land surface with Sentinel-2 (S-2) satellite image. The NDVI is a nondimensional index that depicts the difference in plant cover reflectivity between visible and near-infrared light and can be used to measure the density of green on a piece of land. On the other hand, the dual-pol radar vegetation index (DpRVI) is one of the indices studied using multispectral synthetic aperture radar (SAR) images. Researchers have identified that SAR images are highly sensitive to identify the buildup of biomass from leaf vegetative growth to the flowering stage. Vegetation biophysical characteristics such as the leaf area index (LAI), vegetation water content, and biomass are frequently used as essential system parameters in remote sensing data assimilation for agricultural production models. In the current study, we have used LAI as a system parameter. The findings of the study revealed that the optical data (NDVI) showed a high correlation (up to 0.712) with LAI and a low root-mean-square error (0.0296) compared to microwave data with 0.4523 root-mean-square error. The NDVI, LAI, and DpRVI mean values all decreased between 2019 and 2020. While the DpRVI continued to decline between 2020 and 2021, the NDVI and LAI saw an increase over the same period, which was likely caused by an increase in the study area’s average annual rainfall and the cautious stance of the Red Global (RSG) project on sustainability.

Keywords: DpRVI; NDVI; SAR data; Sentinel-2 vegetation monitoring

1 Introduction

Mangroves are a unique type of coastal ecosystem found in tropical and subtropical regions around the world. They are characterized by their ability to thrive in saltwater environments, where other plants would struggle to survive [1,2,3]. They are economically important natural resources that serve the coastal area inhabitants. Mangroves are also known for the diverse wildlife they support. Many species of fish, crustaceans, and mollusks breed and feed in mangrove habitats, and a variety of bird species such as herons, egrets, and kingfishers use mangroves as nesting and roosting sites. In addition, mangroves also play a crucial role in carbon sequestration and storage. They store large amounts of carbon both in their biomass and in the soil, making them one of the most efficient carbon sinks on Earth [4,5]. The presence of mangrove forests in an area can be used as the best indicator of the ecosystem of the area [6,7]. For example, the abundance of marine life like fish is strongly correlated to mangrove forests.

Unfortunately, mangrove forests are under threat from human activities such as urbanization, agriculture, aquaculture, and pollution, which have led to a decline in mangrove cover worldwide. Therefore, the conservation of mangrove ecosystems has become a priority for many countries, as well as various international organizations [6,8].

Biodiversity, ecosystem productivity, and carbon stock levels are all anticipated to shift as a result of climate change. Scientists have used multispectral satellite images of the land surface to detect and quantify such a change [9,10]. One of the most critical and demanding topics facing the remote sensing field right now is the development of appropriate approaches for the analysis of multitemporal data. Because of the dynamic nature of vegetation cover, those that include a long series of high temporal resolution images are particularly important. Only coarse (300 × 300 m) and medium (30 × 30) spatial resolutions are available for these series [11,12].

Many application platforms are utilized in earth observation. Sentinel Application Platform (SNAP) is one of the application platforms developed for earth observation. This enabled us the continuous monitoring of Earth using Synthetic Aperture Radar (SAR) since 2014 with the launch of Sentine-1 [13]. Sentinel-1 (S-1) satellite, using two satellite sectors A and B, improved the temporal revisit time to 6 days unlike Environmental Satellite (ENVISAT) and European Remote Sensing Satellite, which is 35 days [14].

The first satellite in the Copernicus Program satellite constellation operated by the European Space Agency is called S-1. S-1A and S-1B satellite constellations that make up this mission orbit in the same orbital plane. They are equipped with a C-band synthetic-aperture radar that can collect data day or night in any condition. This instrument can cover an area up to 400 km and has a spatial resolution of 5 m [15]. The constellation orbits the sun in a nearly polar (98.18°) orbit. The orbit repeats every 12 days and completes 175 orbits in that time [16,17].

S-1B was launched on April 25, 2016, and S-1A, the first satellite, was launched on April 3, 2014. Each Soyuz rocket carrying a satellite launched from the Guiana Space Centre in Kourou, French Guiana [5]. Development of S-1C and S-1D is ongoing [6], with S-1C scheduled for launch in 2023 [17].

Due to their superior capacity to distinguish different crop and land cover classes in varied environments, data from remote sensors with different properties (e.g., microwave and optical data) have become more and more important [18]. Such an integrative strategy has been used in earlier studies to increase classification accuracy. For instance, several research studies [19,20,21,22] used the integration of S-1 and S-2 data. Behzad and Aamir [21] made a comparative analysis using microwave and optical data and mapped wheat area in Pakistan, while Kpienbaareh et al. [20] employed S-1 and S-2 data along with crowdsourced ground truth information and a deep learning classification algorithm to identify rice and cotton fields in India. Kaplan and Avdan [22] mapped crop types in the US Midwest using Landsat data.

Various vegetation indices, other than normalized difference vegetation index (NDVI), are studied using SAR images. Radar vegetation index, radar forest degradation index, and dual pol radar vegetation index (DpRVI) are some of indices studied using multispectral SAR images. Researchers have identified that SAR images are highly sensitive to identify the buildup the biomass from leaf vegetative growth to the flowering stage [23,24].

Time series based on the NDVI, particularly, are critical for remote sensing of vegetation phenology and extracting numerical insights about vegetation change [25,26]. An unparalleled potential in vegetation phenology monitoring for agricultural management is provided by the harmonious integration of SAR and optical time series. Using the openly accessible optical S-2 and SAR S-1 time series, it is possible to observe crop phenology with great spatial precision on a weekly basis [27]. To eliminate conflicts and create a reliable long-term data collection for applications based on lengthy time series, S-1 will operate in a preprogramed operation mode [17,23,28].

The study area is a part of the Red Sea Project. Red Sea Project is a development under construction by Red Sea development company (RSG) that spans over 28,000 km² along the Red Sea’s coastlines and is developing into a sustainable luxury tourism spot on the Kingdom of Saudi Arabia’s west coast. The area is found between the Islands of Al Wajhi and Umlij cities and characterized by many habitats like coral reefs, mangroves, and seagrass. The Red Sea development company is dedicated to creating a place that will have a net-positive impact on biodiversity while promoting sustainable tourism [29]. The Regional Organization for the Conservation of the Environment of the Red Sea and Gulf of Aden (PERSGA) gives a comprehensive study of the current state of mangroves in the Red Sea and the Gulf of Aden. A majority of countries along the Red Sea coast and Gulf of Aden like Saudi Arabia, Jordan, Sudan, Somalia, Egypt, and Yemen have joined this organization to collaborate on the conservation of this coastal area [30,31,32].

The primary objective of this work is to shade a light on the best and representative method to calculate vegetation indices (DpRVI and NDVI) from optical sensor (S-2) and microwave sensor (S-1) data and to study the alteration in land use and landcover of the western Saudi Islands. Given that the study area is a part of the Red Sea mega project, it will be important for the relevant agency to respond if there are changes to the land use or land cover of the region.

2 Materials and methods

2.1 Study area

The study area is located along the Red Sea coast of Saudi Arabia, between Yanbu city to the south and Alwajhi to the north (Figure 1). The study area is part of the Red Sea project under development by the Red Sea Development company (RSG). Along the west coast of Saudi Arabia, RSG is developing a brand-new luxury regenerative tourism destination in the study area. The Red Sea, one of the last truly undiscovered wonders of the world, is encircled by the fourth-largest barrier reef system in the world and covers an area of over 28,000 km². It is home to an archipelago of over 90 uninhabited islands, pristine beaches, dormant volcanoes, sweeping desert dunes, mountain canyons, and historical cultural sites. There are several major islands (Fuawaideh, Umm Sahar, and Mount Hassan Islands) and several habitats within the study area. These islands are located close to the city of Umluj and are regarded as some of the most popular and stunning islands along the Red Sea coast. People from the cities of Umluj and Alwajh frequently visit these islands to relax away from tourist crowds of the city and hustle of the city. The area’s most common ecosystems are lagoons, salt swamps, and mangrove timberlands [3,33]. The mangrove ecosystem serves as a food source, a source of fuel, and a source of building materials for the nearby inhabitants.

Figure 1

Location map. (a) The location of the Kingdom of Saudi Arabia with respect to the world map, (b) the location of the study area with respect to Saudi Arabia, and (c) the study area.

Mangroves are important ecosystems in Umluj city, as they provide a range of ecological, economic, and social benefits. The key reason why mangroves are important is they are habitats for a variety of species, including fish, birds, and crustaceans. They serve as nurseries for many commercially important fish species, including shrimp, crab, and lobster. Mangroves act as a natural buffer against coastal erosion, storm surges, and flooding [34]. They help to stabilize the shoreline and reduce the impact of waves and currents. Moreover, mangroves have been used for traditional activities in Umluj city for centuries, such as fishing, harvesting of mangrove honey, and gathering of medicinal plants.

The study area experiences extremely low rainfall of 16 mm/year and high potential flashfloods that can go up to 120 mm. In Umluj, the winters are short, pleasant, dry, and windy, and the majority of the year is clear. The summers are lengthy, hot, unpleasant, and deserted. The average annual temperature ranges from 58 to 97°F, with occasional exceptions when it falls below 53°F or rises over 101°F. In the study area, there are two separate meteorological occurrences of monsoon seasons [33]:

Northeasterly monsoon (October and November) and
Southwesterly monsoon (April and May).

2.2 Methodological framework

Four satellite images in all were retrieved from Copernicus. Two of these images are of S-1, and the remaining images are of S-2. All images are downloaded based on the winter season, in December 2019 and December 2021. Since optical data are only used when the images are cloud free, S-2 images are retrieved first. Given that microwave data can be used in any circumstance, optical data are used as a criterion, depending on whether there are clouds in the images [35].

The NDVI and DpRVI of the two data – passive and active data – are computed after downloading and preprocessing the images in SNAP software. The two data are compared based on the biophysical parameter, leaf area index (LAI). The results of the vegetation indices computation are exported to ArcMap to mask out areas of the Sea and extract values of the parameters in the study for analysis. The short summary and workflow diagrams of the methodology are presented in Figures 2 and 3.

Figure 2

The current workflow of Sentinel-1 data.

Figure 3

The current workflow of Sentinel-2.

2.2.1 Microwave data (S-1)

2.2.1.1 TOPS Split

S-1 TOPS Split is used to filter the data so that it only contains the bursts needed for the analysis. If the region of interest spans more than one Sub-Swath after they have been debursted, they must be joined using the S-1 TOPS Merge operator, which is not the case in this instance.

2.2.1.2 Applying orbit file

The location information of the satellite at the time SAR data were being acquired is contained in orbit auxiliary data. They are added to its metadata with the Apply Orbit File operator (under the menu point Radar) and are immediately downloaded for S-1 products by SNAP. Both retrieved orbit files and precise orbit ephemerides (POE) orbit files are provided by the S-1 precise orbit determination service. With orbit state vectors spaced every 10 s, POE files have a 28-h coverage time. In the 20 days following the gathering of the data, one file is generated each day and distributed [16,36].

If precise orbits are not yet available for the product, restituted orbits can be chosen. While they may not be as accurate as the precise orbits, these restituted orbits will nevertheless be superior to the predicted orbits that are currently available for the product.

2.2.1.3 Radiometric calibration

As a calibrated measure, radiometric calibration transforms the backscatter intensity as it is received by the sensor into the normalized radar cross section. This conversion takes into consideration the global incidence angle of the image as well as other sensor-specific features. This enables the comparison of radar images obtained using various dates, sensors, or imaging geometries.

2.2.1.4 TOPS deburst

The S-1 multispectral image product is subjected to the S-1 TOPS deburst operator, under S-1 TOPS, to eliminate the seamlines between the single bursts. The output consists of the same bands as the input but with the bursts combined based on their zero Doppler times.

2.2.1.5 Multi-looking

To create ground-ranged square pixels, S-1 images are multi-looking by 4 × 1 in range and azimuth direction. Then, a 2-by-2 covariance matrix is created using these multilooking images.

2.2.1.6 Matrix generation (C2)

For a given full polarimetric SAR product, this operator generates the following polarimetric covariance or coherency matrices. Here, we have produced C2 product, which later we will use in DpRVI generation.

2.2.1.7 DpRVI generation

The dual-pol radar vegetation index (DpRVI) is then produced using the dual-pol 2 × 2 covariance matrix C2 [37].

C 2 = 1 − 4 | C 2 | ( Tr ( C 2 ) ) 2 ,

where Tr is the matrix operator.

2.2.1.8 Terrain correction

By employing a digital elevation model (DEM) to correct SAR geometric errors, terrain correction will geocode the image and create a map-projected output. Slant range or ground range geometries are transformed into a map coordinate system through the process of geocoding. When geocoding a piece of terrain, intrinsic geometric distortions such as foreshortening, layover, and shadow are corrected using a DEM.

2.2.2 Optical data (S-2)

The normalized difference vegetation index (NDVI) is one of the parameters of vegetation that can be studied by remote sensing of land surface with S-2 satellite image. The NDVI is a nondimensional index that depicts the difference in plant cover reflectivity between visible and near-infrared light and can be used to measure the density of green on a piece of land [38,39]. NDVI is obtained by dividing the difference of the red and near-infrared (NIR) bands by their sum. High NDVI indicates high green vegetation coverage, while low values represent areas of no vegetation [40,41]. The NIR and red bands differs from satellite to satellite. For instance, the NIR and red bands for Landsat are represented by bands 5 and 4, while the NIR and red bands in S-2 products are represented by bands 8 and 4, respectively, according to:

NDVI = ( NIR − RED ) / ( NIR + RED ) ,

where NIR is the near-infrared band of S-2 and RED is the red band of S-2.

Vegetation biophysical characteristics such as the LAI, vegetation water content, and biomass are frequently used as essential system parameters in remote sensing data assimilation for agricultural production models [23]. An ecosystem’s leaf area is quantified by the LAI, which is often defined as one-half of the total green leaf area per unit of horizontal ground surface area [42,43].

Satellite images contain unavoidable faults and distortions such as sensor sensitivity, atmospheric effects, positional mistakes, displacements, and more. To diminish their effects on our processing, the aforementioned errors and distortions should be addressed, and hence, to do this, we employ radiometric and geometric adjustments. These corrections may not eliminate the errors but significantly reduce their effects.

2.2.2.1 Resampling

S-2 multispectral images are a multi-size product. A product that contains bands with various sizes and/or resolutions is referred to as a multi-size product. Resampling is a methodology of changing a multi-size product to a single-size product of the same resolution, 10 m in our case, to simplify the processing workflow. This can come in handy when a multisize product’s SNAP feature is not supported.

2.2.2.2 Subsetting

On the other hand, subsetting is utilized to reduce the loaded data to the relevant area. This procedure will significantly enhance the multispectral image processing in both S-1 and S-2. Both spatial and spectral resampling are possible using the subset function. This may significantly decrease the amount of data by removing unimportant information.

2.2.2.3 Atmospheric and radiometric correction

To obtain the surface reflectance characterizing the atmosphere, atmospheric correction eliminates the effects of atmospheric scattering and absorption. Here, in our case, we have chosen to download S2 MSI L2A product from Copernicus so that it will present atmospherically corrected data. The NDVI and LAI calculations come following the steps presented in Figure 3.

3 Results and discussion

In this section, the comparison of two vegetation indices, NDVI and DpRVI, using passive (optical) data and active (microwave) data is presented. After looking at each vegetation indices separately, the best method to find the vegetation index is proposed based on the crop biophysical parameter, LAI, and the climatology of the area.

Figure 4 and Tables 1 and 2 show the results of the vegetation indices and the biophysical parameters extracted from the study area along with their statistical summaries.

Figure 4

The line charts of the studied parameters.

Table 1

Yearly mean values of the indices

Table 2

Statistical summaries of the vegetation indices

Table 2 clearly shows that the mean NDVI and LAI values increased from summer 2019 to summer 2020. In the summer of 2019, the NDVI averaged around 0.07, whereas in the summer of 2020, it was 0.08. During that time, the LAI rose from 0.364 to 0.59. The mean value of NDVI from summer 2020 to 2021 declined from 0.08 to 0.068, while the LAI value kept on increasing from 0.59 to 1.367 within the same period. The mean DpRVI, however, peaked in the summer of 2019 at a high value of 0.34 and then fell to 0.338 and 0.310 in the summers of 2020 and 2021, respectively (Table 1).

The mean values of both NDVI and LAI decreased from winter 2019 to 2020. In winter 2019, the mean value of NDVI was about 0.48, which later decreased to about 0.028, while LAI decreased from 0.738 to 0.310 within the same period. The DpRVI remained constant in both winters of 2019 and 2020 with the average value of 0.319, which later decreased to 0.311 in the winter 2021. Both LAI and NDVI rose in the winter 2021 with the average values of 0.688 and 0.055, respectively.

Between 2019 and 2020, the average values of both NDVI and LAI fell. While the NDVI dropped from 0.059 to 0.055, the LAI dropped from 0.551 to 0.450. But in 2021, they both increased to have an average value of 1.028 and 0.061, respectively. In terms of DpRVI, it showed a slight decline over the course of 3 years, with values of 0.331, 0.328, and 0.311 from 2019 to 2021, respectively.

The decrease in the mean values of NDVI and LAI from 2019 to 2020 and then their increase from 2020 to 2021 can be closely related to the climatology of the area and the Saudi Arabia’s Red Sea project (RSG), which is under development in the study area. The average annual rainfall of Saudi Arabia also decreased from 10.55 mm in 2019 to 5.27 mm in 2020 and then increased to 26.45 mm in 2021 as shown in Figure 5. The fact that the RSG required substantial work, including the excavation of some islands (such as Shuraya), and that the company intended to finish the first phase of the project in 2021 may be potentially a factor in the decline of NDVI and DpRVI from 2019 to 2020. The company’s cautious focus on sustainability may also have contributed to the increase after that [8]. The RSG asserts that they are committed to protecting the environment and the species that call it home and that they are firmly rooted in sustainability and sustainable development in their entirety. In addition, they have stated that their project is in line with 17 of the UN sustainable development goals [44].

Figure 5

The rainfall over study area: (a) monthly average precipitation, (b) daily sum and monthly average rainfall, (c) annual average precipitation.

The NDVI values ranging between 0.1 and 0.9 are observed at about 40 sampling points out of 130 total sampling points. This tells us that there is sparse to dense green vegetation in those areas. The DpRVI values range from 0.3 to 1 at about 60 sampling points showing sparse to dense vegetation in the area [45].

For comparison purpose and visualization of the data, we have normalized all the values of NDVI, DpRVI, and LAI, so that the values of all parameters lie in the same range, and their difference is more pronounced. The formula for normalization is as follows.

NDVI norm = NDVI − NDVI min NDVI max − NDVI min ,

where NDVI norm = Normalized NDVI , NDVI min = Minimum NDVI , and NDVI max = Maximum NDVI .

According to the aforementioned findings, both NDVI and DpRVI are reliable vegetation metrics. But which technique is superior? We must consider a number of independent variables to respond to this query. For instance, we can use other crop growth model like LAI and see the change with other independent variables like climatology of the area.

The correlation coefficients of NDVI and LAI were higher than the correlation coefficients of DpRVI and LAI except in the summer 2021, where the correlation coefficient of DpRVI is greater than the correlation coefficient of NDVI and LAI. The highest correlation for NDVI and LAI was 0.742 in winter 2020, while the highest correlation for DpRVI and LAI (0.489) was observed in the summer season of the same year (Figure 6).

Figure 6

The line charts of the normalized NDVI, DpRVI, and LAI.

The root-mean-square error (RMSE) value between DpRVI and LAI is 0.452, while the RMSE value between NDVI and DpRVI is 0.0296. Since the NDVI shows a high correlation with LAI with low RMSE, this can infer that the NDVI is a better vegetation index over DpRVI.

The study of DpRVI and NDVI with LAI and the analysis of the rainfall over the study area all support the conclusion that the vegetation index from optical data is more sensitive and captured the mangrove vegetation in the region than the radar vegetation index (Figure 7). This is because the DpRVI failed to account for the situation from 2020 to 2021 as its value fell in 2021 unlike LAI and without support from the local climatology [46]. Using the same procedure, the normalization of the other parameters was done. The line charts of some parameters are shown in Figure 8.

Figure 7

Correlation coefficients of the parameters.

Figure 8

The scatter plot of NDVI and DpRVI with respect to LAI (normalized).

Climatology plays a significant role in the development and distribution of mangrove vegetation. Mangroves are found in coastal areas, where they are exposed to a range of environmental conditions such as high salinity, strong tidal currents, and fluctuating water levels [47]. These conditions are affected by the climate and can vary depending on the location and season.

Temperature is one of the most important climatic factors affecting mangroves. Mangroves require warm temperatures to grow and develop. The optimal temperature range for most mangroves is between 25 and 35°C [4,48]. Temperature variations can affect the growth, reproduction, and survival of mangroves. Cold temperatures can cause leaf drop and even death of the mangrove trees, while excessively high temperatures can cause heat stress and reduce growth rates [49].

Precipitation is another important factor influencing the development of mangroves. Mangroves require a constant supply of water to survive, but excessive rainfall can be detrimental to their growth [50]. Mangroves typically grow in areas with high rainfall, but they are also able to tolerate periods of drought. The balance between rainfall and evapotranspiration is crucial for the growth of mangroves.

Salinity is a third important factor influencing the development of mangroves. Mangroves are able to tolerate high levels of salinity, which is necessary for their survival in coastal environments. However, excessively high salinity levels can be harmful to mangroves, leading to reduced growth rates, leaf drop, and even death [51]. Changes in the sea level can also affect the salinity levels in the coastal zone, which can have implications for the distribution and growth of mangroves [52].

Mangrove vegetation monitoring using indices and remote sensing has several advantages and disadvantages. The most common advantages include quantitative measures, in which the indices provide quantitative measures of mangrove vegetation, facilitating comparisons over time and between areas [53]. Accordingly, the indices can provide consistent measures of mangrove vegetation, allowing for long-term monitoring, and accessibility, where the indices can be easily calculated from remote sensing data, making them accessible to a wide range of researchers and practitioners [54].

On the other hand, the most foreseen disadvantages are the limited scope, in which the indices may not capture the full range of ecological processes and factors affecting mangrove vegetation, limiting their usefulness [55]. The limited resolution, where the indices may not provide the spatial or temporal resolution needed to capture changes in mangrove vegetation at the appropriate scale, and the limitations in interpretation, that the indices do not provide detailed information on the underlying causes of changes in mangrove vegetation, limiting interpretation and understanding of ecosystem dynamics [56].

4 Conclusions and recommendations

The present study clearly showed that there is a shift in the vegetation along the western coast of Saudi Arabia between 2019 and 2021. The result showed that there was a decrease in the vegetation from 2019 to 2020, while there was an increase from 2020 to 2021. Of the two vegetation indices used to characterize mangrove vegetation in the study area, the NDVI showed an enhanced sensitivity since it showed a high correlation and low RMSE (0.0296) with LAI. The result also has been checked against the climatology of the study area. Since the average rainfall of the study area also decreased from 2019 to 2020 and then increased in 2021 as the average values of NDVI does, we could practically declare that the optical data accurately captured the mangrove vegetation of the research area, even though both approaches have their own pros and cons.

The comparative study of the two vegetation can be very useful provided that both the optical sensors and radar sensors work with different parameters and have their own specialty like working in different weather conditions. For example, because it is not reliant on the weather, radar vegetation index can be utilized day or night and in any weather. Therefore, to provide very good and promising results, the study of this kind needs to be based on a long time series of data.

Given that the study area is part of the RSG mega project, the management of protected areas of mangrove forests in Saudi Arabia shall involve a combination of monitoring, conservation, restoration, stakeholder engagement, regulation and enforcement, and international cooperation. These strategies are aimed at ensuring the long-term sustainability and resilience of mangrove ecosystems and the benefits they provide to the environment.

Acknowledgment

This project was funded by the Deanship of Scientific Research (DSR), King Abdulaziz University, Jeddah, under Grant No. (G: 65-150-1442). The authors, therefore, acknowledge with thanks DSR technical and financial support.

Conflict of interest: The authors declare no conflict of interest.
Data availability statement: The data used to conduct the current research are freely accessed through Copernicus Data Hub.

References

[1] Menezes MP, Berger U, Mehlig U. Mangrove vegetation in Amazonia: a review of studies from the coast of Pará and Maranhão States, north Brazil. Acta Amazonica. 2008;38(3):403–20.10.1590/S0044-59672008000300004Search in Google Scholar

[2] Kumar A, Khan MA, Muqtadir A. Distribution of mangroves along the Red Sea coast of the Arabian Peninsula: Part-I: The northern coast of western Saudi Arabia. Earth Sci India. 2010;3.Search in Google Scholar

[3] Elhag M, Bahrawi JA. Sedimentation mapping in shallow shoreline of arid environments using active remote sensing data. Nat Hazards. 2019;99(2):879–94.10.1007/s11069-019-03780-4Search in Google Scholar

[4] Friess DA, Rogers K, Lovelock CE, Krauss KW, Hamilton SE, Lee SY, et al. The state of the world’s mangrove forests: past, present, and future. Annu Rev Environ Resour. 2019;44:89–115.10.1146/annurev-environ-101718-033302Search in Google Scholar

[5] Ewel K, Twilley R, Ong J. Different kinds of mangrove forests provide different goods and services. Glob Ecol Biogeogr Lett. 1998;7(1):83–94.10.2307/2997700Search in Google Scholar

[6] Aschbacher J, Ofren R, Delsol JP, Suselo TB, Vibulsresth S, Charrupat T. An integrated comparative approach to mangrove vegetation mapping using advanced remote sensing and GIS technologies: preliminary results. Hydrobiologia. 1995;295(1):285–94.10.1007/BF00029135Search in Google Scholar

[7] Elhag M, Boteva S. Quantitative analysis of different environmental factor impacts on land cover in nisos elafonisos, crete, greece. Int J Environ Res Public Health. 2020;17(18):6437.10.3390/ijerph17186437Search in Google Scholar PubMed PubMed Central

[8] Al-Amri N, Adss B, Reheem AA, Elhag M. Industrial contaminations and water resources degradation assessment in coastal city of Rabigh, Saudi Arabia. Arab J Geosci. 2022;15(11):1048.10.1007/s12517-022-09696-xSearch in Google Scholar

[9] Forkel M, Carvalhais N, Verbesselt J, Mahecha M, Neigh C, Reichstein M. Trend change detection in NDVI time series: Effects of inter-annual variability and methodology. Remote Sens. 2013;5(5):2113–44.10.3390/rs5052113Search in Google Scholar

[10] Pettorelli N. The normalized difference vegetation index. UK: Oxford University Press; 2013.10.1093/acprof:osobl/9780199693160.001.0001Search in Google Scholar

[11] Hussain S, Elfeki AM, Chaabani A, Yibrie EA, Elhag M. Spatio-temporal evaluation of remote sensing rainfall data of TRMM satellite over the Kingdom of Saudi Arabia. Theor Appl Climatol. 2022;150(1):363–77.10.1007/s00704-022-04148-8Search in Google Scholar

[12] Elhag M, Bahrawi JA. Deterioration of shallow costal environments using synthetic aperture radar data. Desalin Water Treat. 2020;194:333–42.10.5004/dwt.2020.25233Search in Google Scholar

[13] Adem E, Chaabani A, Elfeki A, Zhang L, Elhag M. SBAS-InSAR/GNSS surface deformation assessment in arid environments of Najran, Saudi Arabia. Earth Syst Environ. 2023;7:1–16.10.1007/s41748-023-00346-0Search in Google Scholar

[14] Delgado Blasco J, Foumelis M, Stewart C, Hooper A. Measuring urban subsidence in the Rome metropolitan area (Italy) with Sentinel-1 SNAP-StaMPS persistent scatterer interferometry. Remote Sens. 2019;11(2):129.10.3390/rs11020129Search in Google Scholar

[15] Hidayatulloh A, Chaabani A, Zhang L, Elhag M. DEM study on hydrological response in Makkah City, Saudi Arabia. Sustainability. 2022;14(20):13369.10.3390/su142013369Search in Google Scholar

[16] Alvarez-Mozos J, Villanueva J, Arias M, Gonzalez-Audicana M. Correlation between NDVI and sentinel-1 derived features for maize. In 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS. IEEE; 2021.10.1109/IGARSS47720.2021.9554099Search in Google Scholar

[17] EuropeanSpaceAgency. Sentinel Online. Retrieved 14, October 2022; https://sentinel.esa.int/web/sentinel/missions/sentinel-1.Search in Google Scholar

[18] Aldhebiani AY, Elhag M, Alshehri AA. Consideration of hyperspectral data in intraspecific variation (spectrotaxonomy) in Prosopis juliflora (Sw.) DC, Saudi Arabia. Open Geosci. 2021;13(1):280–92.10.1515/geo-2020-0231Search in Google Scholar

[19] Wang S, Azzari G, Lobell DB. Crop type mapping without field-level labels: Random forest transfer and unsupervised clustering techniques. Remote Sens Environ. 2019;222:303–17.10.1016/j.rse.2018.12.026Search in Google Scholar

[20] Kpienbaareh D, Sun X, Wang J, Luginaah I, Bezner Kerr R, Lupafya E, et al. Crop type and land cover mapping in northern Malawi using the integration of sentinel-1, sentinel-2, and planetscope satellite data. Remote Sens. 2021;13(4):700.10.3390/rs13040700Search in Google Scholar

[21] Behzad A, Aamir M. Estimation of wheat area using sentinel-1 and sentinel-2 datasets. Int J Agric Sustain Dev. 2019;1(3):81–93.10.33411/IJASD/2019010306Search in Google Scholar

[22] Kaplan G, Avdan U. Sentinel-1 and Sentinel-2 data fusion for wetlands mapping: Balikdami, Turkey. International Society for Photogrammetry and Remote Sensing; 2018.10.20944/preprints201807.0244.v1Search in Google Scholar

[23] Mandal D, Kumar V, Lopez-Sanchez JM, Bhattacharya A, McNairn H, Rao YS. Crop biophysical parameter retrieval from Sentinel-1 SAR data with a multi-target inversion of Water Cloud Model. Int J Remote Sens. 2020;41:5503–24.10.1080/01431161.2020.1734261Search in Google Scholar

[24] Kacic P, Thonfeld F, Gessner U, Kuenzer C. Forest structure characterization in Germany: novel products and analysis based on GEDI, sentinel-1 and sentinel-2 data. Remote Sens. 2023;15(8):1969.10.3390/rs15081969Search in Google Scholar

[25] Martínez B, Gilabert MA. Vegetation dynamics from NDVI time series analysis using the wavelet transform. Remote Sens Environ. 2009;113(9):1823–42.10.1016/j.rse.2009.04.016Search in Google Scholar

[26] Shahbandeh M, Elhag M. Microclimate changes and trend analysis of remotely sensed environmental parameters in West Asia Semi-arid region. Environ Dev Sustainability. 2023;1–15.10.1007/s10668-023-03201-4Search in Google Scholar

[27] Mercier A, Betbeder J, Baudry J, Le Roux V, Spicher F, Lacoux J, et al. Evaluation of Sentinel-1 & 2 time series for predicting wheat and rapeseed phenological stages. ISPRS J Photogramm Remote Sens. 2020;163:231–56.10.1016/j.isprsjprs.2020.03.009Search in Google Scholar

[28] Stendardi L, Karlsen S, Niedrist G, Gerdol R, Zebisch M, Rossi M, et al. Exploiting time series of Sentinel-1 and Sentinel-2 imagery to detect meadow phenology in mountain regions. Remote Sens. 2019;11(5):542.10.3390/rs11050542Search in Google Scholar

[29] Aljahdali MH, Elhag M. Calibration of the depth invariant algorithm to monitor the tidal action of Rabigh City at the Red Sea Coast, Saudi Arabia. Open Geosci. 2020;12(1):1666–78.10.1515/geo-2020-0217Search in Google Scholar

[30] Gladstone W, Tawfiq N, Nasr D, Andersen I, Cheung C, Drammeh H, et al. Sustainable use of renewable resources and conservation in the Red Sea and Gulf of Aden: issues, needs and strategic actions. Ocean Coast Manag. 1999;42(8):671–97.10.1016/S0964-5691(99)00040-XSearch in Google Scholar

[31] Chalastani VI, Manetos P, Al-Suwailem AM, Hale JA, Vijayan AP, Pagano J, et al. Reconciling tourism development and conservation outcomes through marine spatial planning for a Saudi Giga-Project in the Red Sea (The Red Sea Project, Vision 2030). Front Mar Sci. 2020;7:168.10.3389/fmars.2020.00168Search in Google Scholar

[32] Elhag M, Bahrawi J, Zhang L, Ejaz N. Assessment and quantification of sediment retention and dam management in arid environments using remote sensing techniques. Arab J Geosci. 2023;16(10):559.10.1007/s12517-023-11661-1Search in Google Scholar

[33] Bahrawi JA, Elhag M. Simulation of sea level rise and its impacts on the western coastal area of Saudi Arabia. Indian J Mar Sci. 2016;45:54–61.Search in Google Scholar

[34] Alharthi A, El-Sheikh MA, Elhag M, Alatar AA, Abbadi GA, Abdel-Salam EM, et al. Remote sensing of 10 years changes in the vegetation cover of the northwestern coastal land of Red Sea, Saudi Arabia. Saudi J Biol Sci. 2020;27(11):3169–79.10.1016/j.sjbs.2020.07.021Search in Google Scholar PubMed PubMed Central

[35] Bahrawi J, Al-Amri N, Elhag M. Microwave versus optical remote sensing data in urban footprint mapping of the Coastal City of Jeddah, Saudi Arabia. J Indian Soc Remote Sens. 2021;49(10):2451–66.10.1007/s12524-021-01403-9Search in Google Scholar

[36] Filgueiras R, Mantovani EC, Althoff D, Fernandes Filho EI, Cunha FF. Crop NDVI monitoring based on sentinel 1. Remote Sens. 2019;11(12):1441.10.3390/rs11121441Search in Google Scholar

[37] Mandal D, Kumar V, Ratha D, Dey S, Bhattacharya A, Lopez-Sanchez JM, et al. Dual polarimetric radar vegetation index for crop growth monitoring using sentinel-1 SAR data. Remote Sens Environ. 2020;247:111954.10.1016/j.rse.2020.111954Search in Google Scholar

[38] Rouse JW, Haas RH, Schell JA, Deering DW. Monitoring vegetation systems in the Great Plains with ERTS. NASA Spec Publ. 1974;351(1):309.Search in Google Scholar

[39] Tucker CJ. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens Environ. 1979;8(2):127–50.10.1016/0034-4257(79)90013-0Search in Google Scholar

[40] Townshend JR, Justice C. Analysis of the dynamics of African vegetation using the normalized difference vegetation index. Int J Remote Sens. 1986;7(11):1435–45.10.1080/01431168608948946Search in Google Scholar

[41] Zhan Y, Fan J, Meng T, Li Z, Yan Y, Huang J, et al. Analysis on vegetation cover changes and the driving factors in the mid-lower reaches of Hanjiang River Basin between 2001 and 2015. Open Geosci. 2021;13(1):675–89.10.1515/geo-2020-0259Search in Google Scholar

[42] Pasqualotto N, D’Urso G, Bolognesi SF, Belfiore OR, Van Wittenberghe S, Delegido J, et al. Retrieval of evapotranspiration from sentinel-2: Comparison of vegetation indices, semi-empirical models and SNAP biophysical processor approach. Agronomy. 2019;9(10):663.10.3390/agronomy9100663Search in Google Scholar

[43] Xie Q, Dash J, Huete A, Jiang A, Yin G, Ding Y, et al. Retrieval of crop biophysical parameters from Sentinel-2 remote sensing imagery. Int J Appl Earth Obs Geoinf. 2019;80:187–95.10.1016/j.jag.2019.04.019Search in Google Scholar

[44] Adem E, Elfeki A, Chaabani A, Alwegdani A, Hussain S, Elhag M. Impact of satellite precipitation estimation methods on the hydrological response: case study Wadi Nu man basin, Saudi Arabia. Theor Appl Climatol. https://doi.org/10.1007/s00704-024-04855-4.10.1007/s00704-024-04855-4Search in Google Scholar

[45] Aldhebiani AY, Elhag M, Hegazy AK, Galal HK, Mufareh NS. Consideration of NDVI thematic changes in density analysis and floristic composition of Wadi Yalamlam, Saudi Arabia. Geosci Instrum Methods Data Syst. 2018;7(4):297–306.10.5194/gi-7-297-2018Search in Google Scholar

[46] Bahrawi J, Alqarawy A, Chabaani A, Elfeki A, Elhag M. Spatiotemporal analysis of the annual rainfall in the Kingdom of Saudi Arabia: predictions to 2030 with different confidence levels. Theor Appl Climatol. 2021;146(3):1479–99.10.1007/s00704-021-03778-8Search in Google Scholar

[47] Elhag M, Bahrawi JA. Conservational use of remote sensing techniques for a novel rainwater harvesting in arid environment. Environ Earth Sci. 2014;72(12):4995–5005.10.1007/s12665-014-3367-6Search in Google Scholar

[48] Krauss KW, Lovelock CE, McKee KL, López-Hoffman L, Ewe SML, Sousa WP. Environmental drivers in mangrove establishment and early development: a review. Aquat Botany. 2008;89(2):105–27.10.1016/j.aquabot.2007.12.014Search in Google Scholar

[49] Elhag M, Galal HK, Alsubaie H. Understanding of morphometric features for adequate water resource management in arid environments. Geosci Instrum Methods Data Syst. 2017;6(2):293–300.10.5194/gi-6-293-2017Search in Google Scholar

[50] Bahrawi J, Ewea H, Kamis A, Elhag M. Potential flood risk due to urbanization expansion in arid environments, Saudi Arabia. Nat Hazards. 2020;104(1):795–809.10.1007/s11069-020-04190-7Search in Google Scholar

[51] Elhag M. Evaluation of different soil salinity mapping using remote sensing techniques in arid ecosystems, Saudi Arabia. J Sens. 2016;2016:2016–8.10.1155/2016/7596175Search in Google Scholar

[52] Bahrawi JA, Elhag M. Simulation of sea level rise and its impacts on the western coastal area of Saudi Arabia. Indian J Geo-Marine Sci. 2016;45:54–61.Search in Google Scholar

[53] Huxham M, Emerton L, Kairo J, Munyi F, Abdirizak H, Muriuki T, et al. Applying climate compatible development and economic valuation to coastal management: a case study of Kenya’s mangrove forests. J Environ Manag. 2015;157:168–81.10.1016/j.jenvman.2015.04.018Search in Google Scholar PubMed

[54] Ejaz N, Elhag M, Bahrawi J, Zhang L, Gabriel HF, Rahman KU. Soil erosion modelling and accumulation using RUSLE and remote sensing techniques: case study Wadi Baysh, Kingdom of Saudi Arabia. Sustainability. 2023;15(4):3218.10.3390/su15043218Search in Google Scholar

[55] Woodruff DS. Biogeography and conservation in Southeast Asia: how 2.7 million years of repeated environmental fluctuations affect today’s patterns and the future of the remaining refugial-phase biodiversity. Biodivers Conserv. 2010;19:919–41.10.1007/s10531-010-9783-3Search in Google Scholar

[56] Hoque Mozumder, MM, Shamsuzzaman MM, Rashed-Un-Nabi M, Karim E. Social-ecological dynamics of the small scale fisheries in Sundarban Mangrove Forest, Bangladesh. Aquacult Fish. 2018;3(1):38–49.10.1016/j.aaf.2017.12.002Search in Google Scholar

Received: 2023-04-11

Revised: 2023-10-26

Accepted: 2023-10-27

Published Online: 2024-02-15

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

https://doi.org/10.1515/geo-2022-0573

Keywords for this article

DpRVI; NDVI; SAR data; Sentinel-2 vegetation monitoring

Creative Commons

BY 4.0