Results Comparison of different satellite products against area-average gauge over Ethiopia and Tanzania at dekadal time scale. Ethiopia:

  Sample dekadal rainfall fields over Ethiopia for 2nd dekad of April 2009 Tanzania:                         Conclusions -  CHIRP/CHIRPS performed better than ARC and TAMSAT both over

Ethiopia and Tanzania. -  No significant differences were observed between CHIRP and

CHIRPS both over Ethiopia and Tanzania . -  CHIRP/S overestimates the areal extent of low rainfall rates. Acknowledgment This work has been funded by NASA NNH12ZDA001N-IDS: Interdisciplinary Research in Earth Science for the project on “Seasonal Prediction of Hydro-Climatic Extremes in the Greater Horn of Africa under Evolving Climate Conditions to Support Adaptation Strategies.”  

VALIDATION OF THE CHIRPS SATELLITE RAINFALL ESTIMATE OVER EAST AFRICA

Tufa Dinku

International Research Institute for Climate and Society(IRI), the Earth Institute at University

tufa@iri.columbia.edu Introduction

There are so many satellite rainfall products from many centers. The quality and temporal/spatial resolution of these products has been improving over time. One of the factors that contributed to the improvement of satellite rainfall products has been the use of passive microwave e data. However, this use of passive microwave data may have also introduced inhomogeneities in the rainfall time series due to changes in the numbers sensors over the years. In parallel to these products, there is also a need for historical rainfall time series derived from a single sensor (i.e., infrared sensors). Such products do exist (e.g. African Rainfall Climatology version 2 (ARC) and the Tropical Applications of Meteorology using Satellite data (TAMSAT)), but are limited in terms geographical coverage. The Climate Hazards Group Infrared Precipitation (CHIRP) and CHIRP combined with station data (CHIRPS) are recently produced satellite-based rainfall products with relatively high spatial and temporal resolutions, quasi-global coverage, and a relatively homogeneous time series. These two products (CHIRP and CHIRPS) were evaluated over Eastern Africa’s mountainous and coastal regions by comparing with reference rain gauge data. These products were also compared with the ARC and TAMSAT products. Data Raingauge Data Station data from a relatively dense network over Ethiopia, Kenya and Tanzania have been used for the validation. The validation period 2006 to 2010. Validation stations (O) ) and stations used in CHIRPS (*)over Ethiopia,

Kenya and Tanzania(right). The background shades represent elevation.

Satellite Data CHIRPS (http://chg.geog.ucsb.edu/data/chirps/index.html) is a global satellite rainfall product starting from 1981 and available at spatial resolution 5km. The CHIRP algorithm combines Infrared satellite rainfall estimates calibrated with TRMM data with model reanalysis products. CHIRP and CHIRPS (CHIRP merged with station) are available as 5-day and 10-day (dekadal) accumulations form ftp://chg-ftpout.geog.ucsb.edu/pub/org/chg/products/CHIRP/. TAMSAT (Tropical Application of Meteorology using Satellite and other data) from the University of and ARC (Africa Rainfall Climatology) from the National Oceanic and Atmospheric Administration (NOAA) are also used for comparison with CHIRP. Both TAMSAT and ARC are IR-only products, but ARC includes raingauge data from GTS stations. Other Data A qualitative comparison was also done with ENACTS (Enhancing National Climate Services) products. The ENACTS products are generated by merging satellite rainfall estimates (TAMSAT in this case) with about 600 raingauge stations over Tanzania and over 500 stations over Ethiopia.

8th International Precipitation Working Group(IPWG) Workshop Bologna, Italy, 3-7 Oct 2016