EGU2020-4951
https://doi.org/10.5194/egusphere-egu2020-4951
EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Irrigation and precipitation consistency with SMOS, SMAP, ESA-CCI, Copernicus, Neural Network SSM, AMSR-2 remotely sensed soil moisture

Chiara Corbari1, nicola paciolla1, Ahmad Al Bitar2, Yann Kerr2, and Marco Mancini1
Chiara Corbari et al.
  • 1Politecnico di Milano, SIA - edificio 4A, DICA, Milano, Italy (chiara.corbari@polimi.it)
  • 2Centre d'Etudes Spatiales de la BIOsphère, CESBIO, France

Numerous surface soil moisture (SSM) products are available from remote sensing, ranging different spatial and temporal resolutions. Varying techniques are employed to retrieve SSM and different spatial scales highlight different distributions. Notwithstanding this variety between the available data, all of them should be coherent with the recorded rainfall and irrigation.

In this work we have crossed recorded precipitations with a number of SSM products deriving from remote sensing: Soil Moisture Ocean Salinity (SMOS) mission, Soil Moisture Active Passive (SMAP) mission, European Space Agency Climate Change Initiative (ESA-CCI) products, Copernicus Global Land Operations product, a Neural Network SSM retrieval algorithm and AMSR-2 data.

All the dataset products have been compared with recorded precipitation from on-ground stations over two agricultural sites in Italy: one in the north, near Lake Garda (Chiese Irrigation Consortium) and the other in the south-east in the Apulia region (Capitanata Irrigation Consortium).

In both cases, a first SSM-rain comparison through well-established indexes (Pearson and Spearman correlations) has not yielded encouraging results.

Then, a methodology has been developed to determine whether the variation of SSM is consistent with the presence/absence of precipitation. An Agreement Index (AI) has been derived as a way to measure the coherency between SSM and precipitation. Any time a measure of SSM is available, a positive or negative value for the AI is recorded, according to the rainfall registered since the previous measurement. During the irrigation season (March through September), the presence of this artificial input of water into the system is also taken into account. For every year, the proportion between “coherent” SSM-rainfall pairings (positive AIs) and “non-coherent” pairings (negative AIs) has been computed.

This method is applied to all SSM products in the dataset, and results are compared. When aggregating the results for all the pixels within the irrigation consortia, all seem to align to a similar proportion between “coherent” and “non-coherent” SSM-rainfall pairings, notwithstanding the wide variety of data types, spatial resolutions and retrieval methods. However, even if the overall performances of the products are similar, each shows different spatial distributions, as each product is influenced differently by the physical features of the different areas.

How to cite: Corbari, C., paciolla, N., Al Bitar, A., Kerr, Y., and Mancini, M.: Irrigation and precipitation consistency with SMOS, SMAP, ESA-CCI, Copernicus, Neural Network SSM, AMSR-2 remotely sensed soil moisture , EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-4951, https://doi.org/10.5194/egusphere-egu2020-4951, 2020

Displays

Display file