Characteristic of Dual Polarization Sentinel-1 for Estimation of Seawater Intrusion on Kalianda Coast, South Lampung: A Preliminary Study

Salsabila, Choirunnisa
Ghazali, Mochamad Firman
Dermawan, Ananda
Zahra, Lauditta
Melliana, Ni Made Mega
Aulia, Mila

Jurnal Geografi LIngkungan Tropik (Journal of Geography of Tropical Environments)
Vol. 7, No. 1

Seawater intrusion can increase soil salinity, which occurs due to flooding, evaporation, and land cover changes in coastal areas. This research aims to map and observe the characteristics of seawater intrusion in Way Urang, Kalianda, South Lampung, using Dual-polarization of Sentinel-1 SAR imagery with VV, VH, VV+VH, and (VV+VH)/2 polarization. There were 28 samples used in this research which were divided into two types, GCP (ground control point) and ICP (independent check point). GCP samples are used to create a seawater intrusion estimation model using the regression method, while ICP samples are used to test the validation model using the RMSE method. The seawater intrusion estimation model created in this study had a good RMSE accuracy of 0.01 – 0.08. However, these results were not followed by a strong relationship between backscatter and estimated salinity value. This can be caused by high levels of soil moisture in the research area. Based on the salinity value estimation model, it shows that there is no seawater intrusion in the research area. Seawater intrusion can increase soil salinity, which occurs due to flooding, evaporation, and land cover changes in coastal areas. This research aims to map and observe the characteristics of seawater intrusion in Way Urang, Kalianda, South Lampung, using Dual-polarization of Sentinel-1 SAR imagery with VV, VH, VV+VH, and (VV+VH)/2 polarization. There were 28 samples used in this research which were divided into two types, GCP (ground control point) and ICP (independent check point). GCP samples are used to create a seawater intrusion estimation model using the regression method, while ICP samples are used to test the validation model using the RMSE method. The seawater intrusion estimation model created in this study had a good RMSE accuracy of 0.01 – 0.08. However, these results were not followed by a strong relationship between backscatter and salinity value. This can be caused by high levels of soil moisture in the research area. Based on the salinity value estimation model, it shows that there is no seawater intrusion in the research area.