Turini, N.; Thies, B.; Horna, N. & Bendix, J. (2021): <b>Random forest-based rainfall retrieval for Ecuador using GOES-16 and IMERG-V06 data</b>. <i>European Journal of Remote Sensing</i> <b>54</b>(1), 117-139.
Resource Description
Title:
Random forest-based rainfall retrieval for Ecuador using GOES-16 and IMERG-V06 data
A new satellite-based algorithm for rainfall retrieval in high spatio-temporal resolution fo<br/>
Ecuador is presented. The algorithm relies on the precipitation information from the Integrated<br/>
Multi-SatEllite Retrieval for the Global Precipitation Measurement (GPM) (IMERG) and infrared<br/>
(IR) data from the Geostationary Operational Environmental Satellite-16 (GOES-16). It wa<br/>
developed to (i) classify the rainfall area (ii) assign the rainfall rate. In each step, we selected<br/>
the most important predictors and hyperparameter tuning parameters monthly. Between 19<br/>
April 2017 and 30 November 2017, brightness temperature derived from the GOES-16 IR<br/>
channels and ancillary geo-information were trained with microwave-only IMERG-V06 using<br/>
random forest (RF). Validation was done against independent microwave-only IMERG-V06<br/>
information not used for training. The validation results showed the new rainfall retrieva<br/>
technique (multispectral) outperforms the IR-only IMERG rainfall product. This offers using<br/>
the multispectral IR data can improve the retrieval performance compared to single-spectrum<br/>
IR approaches. The standard verification scored a median Heidke skill score of ~0.6 for the rain<br/>
area delineation and R between ~0.5 and ~0.62 for the rainfall rate assignment, indicating<br/>
uncertainties for Andes’s high elevation. Comparison of RF rainfall rates in 2 km2 <br/>
resolution<br/>
with daily rain gauge measurements reveals the correlation of R = ~0.33.
Keywords:
| random forest | rainfall | GOES |
Literature type specific fields:
ARTICLE
Journal:
European Journal of Remote Sensing
Volume:
54
Issue:
1
Page Range:
117-139
Publisher:
Taylor & Francis
Metadata Provider:
Individual:
Jörg Bendix
Contact:
email:
bendix <at> staff.uni-marburg.de
Faculty of Geography
Deutschhausstraße 10
Philipps University of Marburg
Laboratory for Climatology and Remote Sensing
35032 Marburg
Germany
