ارزیابی پهنه‌های تالاب انزلی بر اساس سری‌های زمانی داده‌های Landsat و شاخص MNDWI

نوع مقاله : مقاله پژوهشی

نویسندگان

• مریم حقیقی خمامی ¹
• امیراسلام بنیاد ²
• محمد پناهنده ³

¹ گروه جنگلداری، دانشکده منابع طبیعی، دانشگاه گیلان، صومعه سرا، ایران

² گروه جنگلداری، دانشکده منابع طبیعی، دانشگاه گیلان.

³ گروه فرآوری پسماند، پژوهشکده محیط زیست، جهاد دانشگاهی گیلان، رشت، ایران.

چکیده

کلیدواژه‌ها

• سطح تراز دریا"
• شاخص MNDWI"
• گوگل ارث انجین"
• "
• ماشین بردار پشتیبان

20.1001.1.2008479.1402.54.1.11.1

موضوعات

• محیط زیست

عنوان مقاله [English]

Anzali Wetland Surface Area Evaluation Based on Landsat Time Series Data and NDWI Indices

نویسندگان [English]

• Maryam Haghighi Khomami ¹
• Amir Eslam Bonyad ²
• mohammad panahandeh ³

کلیدواژه‌ها [English]

• Sea Level"
• MNDWI Index"
• Google Earth Engine"
• "
• SVM

EXTENDED ABSTRACT

Introduction

Wetlands play a critical role in the environment. With the impacts of climate change as one of the most important factors affecting the balance of wetlands, these natural ecosystems have suffered severe climatic and biophysical parameters which cause undergone many changes in their functions and structures. Therefor evaluating the surface water area in wetlands is a functional requirement for studying ecological and hydrological processes. The wetland restoration and management requires sufficiently frequent and high-spatial-resolution remote sensing data to represent its dynamics. This paper reviews the current status of detecting and evaluating surface water are using remote sensing data and index in Google Earth Engine(GEE).

Materials and methods

In this study, the Anzali International Wetland in Guilan province, Iran was selected as the study area and to identify the changes of the wetland, Landsat time series images and the normalized difference waterindex in Google Earth Engine were used. this study used the available 36 scenes Landsat TM, and OLI images in this region from 1986 to 2020 and processed the data on the (GEE) platform. Support Vector Machine (SVM) algorithms were applied to produse custom maps and two classes of water and non-water area of the wetland were classifeied in ENVI 5.6. The water index was used to quantify the spatiotemporal variability of the surface water area changes over the years and MNDWI maps were computed using the appropriate GEE function. Also climate data including TRMM satellite data and PDSI index from TerraClimate data and Caspian Sea water level data were used.

Results

The results outline different aspects of the spatial and temporal distribution of the wetland water surface area. The produced maps had reasonable accuracy with the Support Vector Machine (SVM) algorithms. Overall accuracies based on the algorithms used ranged between 87% and 92% and Kappa coefficient between 88% and 94%. This level of accuracy is reasonable considering the area classified in this study. The results of comparing the MNDWI index with the Landsat classification confirm that the MNDWI index more accurately monitors the details of changes. The wetland water body loss has decreased by 20% in its area according to mNDWI index maps, so that initially (until 2000) there was an upward trend and then a downward trend. This confirms the process of drying and disappearing of the water areas. Also the water level changes of the wetland have been affected by the sea level on the trend more than climatic factors.

Conclusions

The analysis of changes in land use/land cover has always been a major theme in remote sensing techniques applied to Earth observation. The results demonstrated the capabilities of using the Google Earth Engine platform and water indices to characterize and identify the trends of the water level changes of wetlands with acceptable accuracy, and it is also applicable to wetland research in other regions. In particular, the approach proposed here, based on the use of SVM classification, appears to be quite effective and reliable (with an accuracy higher than 0.87) for the identification of changes in coastal wetland environment and MNDWI index in (GEE) was more accurate to provide water changes maps in the study period of time. The findings of this study underscore the relevance of GEE and Landsat composite data in evaluating and mapping coastal wetlands and could provide more detailed scientific guidance for wetland managers by quickly detecting wetland changes at a finer spatiotemporal resolution.