Algorithm Overview
The purpose of Algorithm 3B-42 is to produce Tropical Rainfall Measuring Mission (TRMM) merged high quality (HQ)/infrared (IR) precipitation and root-mean-square (RMS) precipitation-error estimates. These gridded estimates are on a 3-hour temporal resolution and a 0.25-degree by 0.25-degree spatial resolution in a global belt extending from 50 degrees South to 50 degrees North latitude.
The 3B-42 estimates are produced in four stages; (1) the microwave estimates precipitation are calibrated and combined, (2) infrared precipitation estimates are created using the calibrated microwave precipitation, (3) the microwave and IR estimates are combined, and (4) rescaling to monthly data is applied. Each precipitation field is best interpreted as the precipitation rate effective at the nominal observation time.
a. High Quality (HQ) microwave estimates
All of the available passive microwave data are converted to precipitation estimates prior to use, then each data set is averaged to the 0.25° spatial grid over the time range ±90 minutes from the nominal observation time. All of these estimates are adjusted to a "best" estimate using probability matching of precipitation rate histograms assembled from coincident data. The algorithm takes the TCI as the calibrating data source. However, the coincidence of TCI with any of the sensors other than TMI is highly sparse, so we establish a TCI—TMI calibration, then apply that to TMI calibrations of the other sensors to estimate the TCI-calibrated values. The TCI—TMI relationship is computed on a 1°x1° grid for each month using that month’s coincident data to accommodate the somewhat different climatologies of the two estimates. Preliminary work showed that the TMI calibrations of the other sensors’ estimates are adequately represented by climatologically based coefficients representing large areas. In the case of the TMI—SSM/I calibration, separate calibrations are used for five oceanic latitude bands (40-30°N, 30-10°N, 10°N-10°S, 10-30°S, 30-40°S) and a single land area for each of the four three-month seasons. The TMI—AMSR-E and TMI—AMSU-B calibrations are set in the form of a single climatological adjustment for land and another for ocean. The AMSU-B calibration has two additional issues. First, the NESDIS algorithm changed on 31 July 2003, so separate sets of calibrations are provided for the two data periods. Second, in both periods the AMSU-B fractional occurrence of precipitation in the subtropical highs is notably deficient. After extensive preliminary testing, the authors judged it best to develop the ocean calibration in regions of significant precipitation and apply it everywhere. In all cases the calibration is a simple match-up of histograms.
The calibration interval is chosen to be a month to ensure stability and representativeness, except the TMI—AMSR-E calibration is computed with 2 months for stability.
Once the estimates are calibrated for each satellite and audited for >40% "ambiguous pixels", the grid is populated by the "best" data from all available overpasses, although the most likely number of overpasses in the 3-hr window for a given grid box is either one or zero. When there are multiple overpasses, data from TCI, TCI-adjusted TMI, TCI-adjusted AMSR-E, and TCI-adjusted SSM/I are averaged together, and TCI-adjusted AMSU-B estimates are used if none of the others are available for the grid box. Tests show that the histogram of precipitation rate is somewhat sensitive to the number of overpasses averaged together when that number is small. Accordingly, in the future we expect to test a scheme taking the single "best" overpass in the 3-hr period.
b. Variable Rain Rate (VAR) IR estimates
3B-42 uses two different IR data sets for creating the complete record of 3-hrly 0.25° gridded Tbs. In the period 1 January 1998 to 6 February 2000, each grid box’s histograms in the 1°x1° 3-hourly GPCP IR histograms is zenith-angle corrected, averaged to a single Tb value for the grid box, and plane-fit interpolated to the 0.25° grid. For the period from 7 February 2000 onwards, the CPC Merged IR is averaged to 0.25° resolution and combined into hourly files as ±30 minutes from the nominal time. The amount of imagery delivered to CPC varies by satellite operator, but international agreements mandate that full coverage is provided for the 3-hourly synoptic times (00Z, 03Z, …, 21Z). Histograms of time-space matched HQ precipitation rates and IR Tbs, each represented on the same 3-hourly 0.25° grid, are accumulated for a month, and then used to create spatially varying calibration coefficients that convert IR Tbs to precipitation rates. As in the HQ, the calibration interval for the IR is a calendar month, and the resulting adjustments are applied to data for the same calendar month. This choice is intended to keep the dependent and independent data sets for the calibrations as close as possible in time. In fact, the full month of data in the estimates includes the dependent data. A second ambiguous screening is performed on the matched microwave data after accumulation; compared to instantaneous screening, the monthly screening provides better control of artifacts.
By design, there is no precipitation when the 0.25°x0.25°-average Tb is greater than the local threshold value that matches the frequency of precipitation in the IR to that of the microwave. Increasingly colder Tbs are assigned increasingly large precipitation rates using histogram matching. Those grid boxes that lack coincident data throughout the month, usually due to cold-land dropouts or ambiguous editing, are given calibration coefficients by smooth-filling histograms of coincident data from surrounding grid boxes. Finally, preliminary testing showed that the precipitation rates assigned to the coldest Tbs by strict probability matching tended to show unphysical fluctuations. To ameliorate this effect, a somewhat subjectively chosen coldest 0.17% of the Tb histogram is specified by a fourth-order polynomial fit to a climatology of coldest-0.17%—precipitation rate points around the globe. In each grid box a constant is added to each point on the climatological curve such that it is piecewise continuous with the grid box’s Tb-precipitation rate curve at the 0.17% Tb.
Once computed, the HQ-IR calibration coefficients are applied to each 3-hourly IR data set during the month.
c. Combined HQ and VAR estimates The ultimate goal of this algorithm is to provide the "best" estimate of precipitation in each grid box at each observation time. It is frequently quite challenging to combine different estimates of an intermittent field such as precipitation. The process of combining passive microwave estimates is relatively well-behaved because the sensors are quite similar and GPROF is used for most retrievals. This is not the case for the HQ and VAR fields. We currently take a simple approach for combining the HQ and VAR estimates, namely the physically-based HQ estimates are taken "as is" where available, and the remaining grid boxes are filled with VAR estimates. This scheme provides the "best" local estimate, at the expense of a time series that is built from data sets displaying heterogeneous statistics.d. Rescaling to monthly data
The final step in generating 3B-42 is the
indirect use of rain gauge data. It is highly advantageous to
include rain gauge data in combination data sets.. However,
experience shows that on any time scale shorter than a month the
gauge data are not reported with sufficient density nor reported with
consistent observational intervals to warrant direct inclusion in a
global algorithm that provides sub-monthly resolution. The authors
solved this issue in the GPCP One-Degree Daily combination data set
by scaling the short-period estimates to sum to a monthly estimate
that includes monthly gauge data. Here, we take a similar approach
with the 3B-42 estimates. All available 3-hourly HQ+VAR estimates
are summed over a calendar month to create a monthly multi-satellite
(MS) product. The MS and gauge are combined to create a
post-real-time monthly satellite-gauge combination (SG), which is a
TRMM product in its own right (3B43). Then the field of SG/MS ratios
is computed (with controls) and applied to scale each 3-hourly field
in the month.
File Format
The file content description for Product 3B-42 can be obtained
from the Volume 4 - Level 2 and Level 3 File Specifications provided
by TSDIS. It is available at: http://tsdis02.nascom.nasa.gov/tsdis/Documents/ICSVol4.pdf.
Known Deficiencies
The IR data prior to February 2000 covers the span 40 degrees North to 40 degrees South. After and including February 2000, the data cover 50 degrees North to 50 degrees South. This results in a minor discontinuity in the data record. Also, HQ data sources are introduced at different points in the data record. Therefore, variations in HQ coverage will occur throughout the record, increasing as time progresses. Most critically, the introduction of AMSU-B causes a low bias of almost 10% globally.
Planned Improvements
Efforts are currently focusing on the validation of the Product 3B-42 precipitation estimates with rain gauge data, ground-based radar data, and data from other satellites. Any shortcomings of the algorithm identified during the validation efforts will be addressed with associated enhancements to the algorithm implemented, such as the low bias induced by AMSU-B.
References
Huffman, G.J., R.F. Adler, B. Rudolph, U. Schneider, and P. Keehn, 1995: Global Precipitation Estimates Based on a Technique for Combining Satellite-Based Estimates, Rain Gauge Analysis, and NWP Model Precipitation Information, J. Clim., 8, 1284-1295.
Huffman, G.J., 1997: Estimates of Root-Mean-Square Random Error for Finite Samples of Estimated Precipitation, J. Appl. Meteor., 1191-1201.
Huffman, G.J., R.F. Adler, P. Arkin, A. Chang, R. Ferraro, A. Gruber, J. Janowiak, A. McNab, B. Rudolph, and U. Schneider, 1997: The Global Precipitation Climatology Project (GPCP) Combined Precipitation Dataset, Bul. Amer. Meteor. Soc., 78, 5-20.
Huffman, G.J., R.F. Adler, D.T. Bolvin, G. Gu, E.J. Nelkin, K.P. Bowman, Y. Hong, E.F. Stocker, D.B. Wolff, 2007: The TRMM Multi-satellite Precipitation Analysis: Quasi-Global, Multi-Year, Combined-Sensor Precipitation Estimates at Fine Scale. J. Hydrometeor., 8(1), 38-55.
Huffman, G.J., R.F. Adler, M. Morrissey, D.T. Bolvin, S. Curtis, R. Joyce, B McGavock, J. Susskind, 2001: Global Precipitation at One-Degree Daily Resolution from Multi-Satellite Observations. J. Hydrometeor., 2(1), 36-50.