Introduction

Floods are a common phenomenon in flat ecosystems with deficiencies in river drainage, impacting the local and regional economy. They can be identified and analyzed by remote sensing.

Objective

The aim of this study was to evaluate the performance of two normalized water indices during periods of maximum and minimum annual precipitation for the identification of flooding in the Salado River Basin, Argentina.

Materials and methods

The years with maximum and minimum annual precipitation in the period 2001-2020 were derived from satellite estimates of monthly precipitation provided by NASA through Google Earth Engine. Floods were identified using Landsat images, applying two normalized water indices (NDWI -Normalized Difference Water Index- and modified NDWI) to evaluate their performance in generating binary images that better represent the reality of the study area.

Results and discussion

Both indices showed good capability in identifying permanent or semi-permanent watercourses and water bodies; however, only the NDWI demonstrated higher effectiveness in identifying flooded areas with shallow depths (5 to 15 cm). The use of the Landsat mid-infrared band (1 566 - 1 651 µm) is less sensitive to water sediment load and can reflect subtle differences in it, providing a greater ability to delineate the water-soil boundary.

Conclusion

The use of NDWI showed a suitable behavior for the identification of flooded areas in very low slope ecosystems

        · abstract in Spanish     · text in English | Spanish     · English ( pdf ) | Spanish ( pdf )