Double-scale analysis on the detectability of irrigation signals from remote sensing soil moisture over an area with complex topography in central Italy
•Irrigation detection is investigated through remote sensing soil moisture.•Diagnostic indices derived by the temporal stability theory are used.•High-resolution (100 m) maps of irrigated areas are produced through the k-means algorithm.•The spatial matching between the irrigation extent and the dat...
Gespeichert in:
Veröffentlicht in: | Advances in water resources 2022-03, Vol.161, p.104130, Article 104130 |
---|---|
Hauptverfasser: | , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | •Irrigation detection is investigated through remote sensing soil moisture.•Diagnostic indices derived by the temporal stability theory are used.•High-resolution (100 m) maps of irrigated areas are produced through the k-means algorithm.•The spatial matching between the irrigation extent and the data resolution is crucial.
Despite a detailed knowledge of the spatial-temporal dynamics of irrigation being necessary to optimize the agricultural production without exacerbating the pressure exercised on the water resource, such information is still often lacking worldwide. In this study, a double-scale analysis on the detectability of the irrigation occurrence over an area in central Italy through remote sensing soil moisture is proposed; the period of interest is a 3-year time span from 2017 to 2019. The detectability of district- or sub-district-scale irrigation signals through remotely sensed soil moisture data is investigated at two different spatial resolutions: 1 km and plot scale. Three soil moisture products sampled at 1 km resolution are evaluated: a DISPATCH (DISaggregation based on Physical And Theoretical scale CHange) downscaled version of SMAP (Soil Moisture Active Passive) and two Sentinel-1-derived products, namely the 1 km version delivered by Copernicus and a plot-scale-born version developed by THEIA and aggregated at 1 km. The THEIA Sentinel-1 product aggregated at 100 m is used in the plot-scale analysis. Over the study area, the irrigation extent is determined by the fragmentation of the agricultural fields and the complex topography, making the adoption of plot-scale data necessary. Satisfactory results are obtained by comparing maps of irrigated areas at 100 m spatial resolution produced through the k-means clustering algorithm with ground-truth data, since the method fails only once out of seven in properly reproducing the irrigated or non-irrigated conditions occurred over four pilot agricultural fields.
[Display omitted] |
---|---|
ISSN: | 0309-1708 1872-9657 |
DOI: | 10.1016/j.advwatres.2022.104130 |