The 1° – 1 day accumulations are estimated, merging passive microwave rain retrievals, cloud top Infrared brightness temperatures and rain gauge data using a methodology similar to the ones developed by Huffman et al. (2001) and Kidd et al. (2003). Gridded error estimations due to sampling, calibration and algorithmic issues will be provided within the MT Level 4 rain product. The error model used was adapted from the model developed for Roca et al. (2009) to compare different rainfall estimation products with rain gauge products at meteorologically relevant scales using error estimates.
Figure 1: Space (left) and time (right) mean variogram functions calculated for the MT Level 4 product for a 5°x5° area in West Africa and during a 10-day period in 2006. The e-folding distances of the exponential models fitted are respectively 48km and 1.4h.
The rain accumulations are built with the instantaneous rain rates of the MT Level 2 Surface Rain Product (Viltard et al., 2006) derived from the MADRAS instrument on board MT and the other available conical scanning passive microwave radiometers, the Infrared 10.8μm data from the Geostationary satellites and are bias-removed with rain-gauge data. The error model performs the evaluation of the uncertainties of the 1° – 1 day accumulations and involves the estimation of the oversampling among the individual samples used to build the gridded product. Figure 1 shows an example of the space and time variogram functions that are calculated to evaluate the space and time auto-correlation of the MT-L4 rain maps and used to produce the errors.
Figure 2: Scatter plots of daily rain gauge data (x-axis) versus satellite product estimations (y-axis) over a one-degree square area in Niger. The blue bars are the kriging errors for the rain gauge and the error model estimations for the satellite products. The full-red (resp. dashed-red) lines are the regression lines with (resp. without) uncertainties taken into account.
The error that will be provided can be useful to estimate the relevance of each individual rain accumulation because they dynamically take into account the characteristics of the multiple-source samples combined into the accumulations. Therefore the MT Level 4 rain product is evaluated using the methodology developed in Roca et al. (2009) to validate rain satellite products taking into account uncertainties on both the satellite and the rain gauge estimations: Figure 2 gives an example of validation of the preliminary MT rain algorithm with three other operational rain products at the daily scale over a one-degree square area in Niger in West Africa using kriged rain-gauge data and their variance of estimation.
Huffman, G.J., R.F. Adler, M.M. Morrissey, D.T. Bolvin, S. Curtis, R. Joyce, B. McGavock, and J. Susskind, 2001: Global Precipitation at One-Degree Daily Resolution from Multisatellite Observations. J. Hydrometeor., 2, 36–50.
Kidd, C., D.R. Kniveton, M.C. Todd, and T.J. Bellerby, 2003: Satellite Rainfall Estimation Using Combined Passive Microwave and Infrared Algorithms. J. Hydrometeor., 4, 1088–1104.
Roca R., P. Chambon, I. Jobard, P-E. Kirstetter, M. Gosset, and J-C. Berges, 2009 : Comparing satellite and surface rainfall products over West Africa at meteorologically relevant scales during the AMMA campaign using error estimates. J. App. Meteor. and Climatol., early online release.
Viltard, N., C. Burlaud, and C.D. Kummerow, 2006: Rain Retrieval from TMI Brightness Temperature Measurements Using a TRMM PR–Based Database. J. Appl. Meteor. Climatol., 45, 455–466.