Environmental characterization and cartographic modeling of wild plant habitats at the northern coastal zone of Egypt

Abstract

Assessing natural vegetation through conventional methods faces considerable constraints, such as limited geographical scope, reduced precision, a lack of historical data, high expenses and time demands. The study aims to use environmental and spectral data to identify and map natural vegetation and plant species along the Mediterranean coast of Egypt. This involves employing spatial analyses and cartographic modeling techniques, marking an initial effort in this endeavor. To fulfill this aim, a total number of 70 wild plant habitats were surveyed and sampled for further laboratory identification of plant species. Multispectral and thermal bands of Landsat imagery were processed to generate land cover map, as well as calculate Normalized Difference Vegetation Index (NDVI) and Land Surface Temperature (LST) to be integrated with the naturally grown plants in a cartographic model to predict the wild plant habitats. Coastal wild plant habitats were sparsely located and associated with a vegetation density ranging from 0.096 to 0.280 with an average of 0.167. Besides, the LST of these habitats fluctuated from 30.559 to 38.652 with an average of 34.361°C. Although the wild plant habitats at the Northern coastal region of Egypt are similar in environmental and climatic conditions, there are variability in NDVI and LST of each single habitat (pure or mixed). On the other hand, the lowest LST were associated with Erodium laciniatum/Rumex pictus and Erodium laciniatum habitats reporting 30.559 and 30.741°C, respectively. However, Lotus halophilus mixed with Ifloga spicata habitats reported the highest LST (38.652°C). The high vegetation density is mainly characterized by low LST which indicates the mixed wild plant habitats. The developed cartographic model showed a narrow strip along the Mediterranean region as highly suitable habitats for wild plant growth. The model can be used to map pure and mixed habitats of various wild plants with an accuracy exceeding 90%. The model was applied for mapping Arthrocnemum macrostachyum showing that the middle coastal region is most suitable for its growth. It’s recommended to integrate remotely sensed data with spatial analyses for the environmental analyses of natural plants. Present findings support researchers and scientists interested in environmental, botanic, and medical studies.