Using Sentinel-2 data for efficient monitoring and modeling of wetland protected areas

Abstract

Wetlands are ecologically vital habitats that play a crucial role in supporting biodiversity and providing essential ecosystem services. They are considered to be among the most productive ecosystems on the planet that provide numerous benefits.For the purposes of this study, Straldzha Complex Protected Area, Bulgaria was chosen as the object of investigation. Straldzha Complex Protected Area includes a reservoir and surrounding wetlands and meadows, the remains of the eastern part of the former Straldzha Plateau (the largest plateau ever in Bulgaria). The wetland is sensitive to human activities, related to the water management and unsustainable use of the former plateau as agricultural land.For the purposes of this study, data from Sentinel-2 satellite of the European Space Agency were used. The monitoring was carried out during the study period 2017 -2022. An index-based classification was used in the study, utilizing NDVI, NDWI and MSAVI2 indices for classifying the contents within the wetlands boundaries. NDGI model was applied as well, evaluating the vegetation dynamics in the marsh. The obtained results showed successful mapping and monitoring of wetlands. The wetlands are of high importance and should be protected and conserved to maintain the benefits they provide to the environment and society.The data and results of this research will be able to serve Destination Earth (DestinE), which is an ambitious initiative of the European Union to create a digital model of the Earth that will be used for monitoring the effects of natural and human activities on our planet, prediction of extreme events and adapting policies to the climate challenges. The data and models will serve the Bulgarian initiative for the construction of the Digital Twins, which is being pilot developed in the

Mapping and Classification of the Liaohe Estuary Wetland Based on the Combination of Object-Oriented and Temporal Features

“…This study extracted a total of 57 features for all image objects. These features were categorized into five groups: single-temporal spectral features (9), spectral standard deviation (9), texture (13), geometric (4), and multi-temporal time series features (22). These five feature sets are meticulously designed to encapsulate the spectral characteristics, wetness, greenness, texture, geometric shape, and temporal variations of land features.…”

“…Wang et al undertook vegetation classification within the coastal wetlands of Yancheng using the random forest algorithm with Sentinel-2 imagery [8]. Ivanova et al conducted the classification and mapping of the Straldzha Complex Protected Area utilizing Sentinel-2 satellite imagery [9]. Therefore, the present study employs Sentinel-2 images as the foundational data for classification purposes.…”