TRMM and GPM Data Products G.J. Huffman NASA/Goddard Space Flight Center Introduction TMPA IMERG

1. TRMM and GPM Data Products • G.J. Huffman • NASA/Goddard Space Flight Center • Introduction • TMPA • IMERG • Transitioning from TRMM to GPM • Final Comments

2. 1. Introduction – Basic Products For both TRMM and GPM there are a variety of products based on the sensors and combinations of sensors The sensor datasets tend to be used by precipitation specialists • access is open to all The multi-satellite datasets tend to be the most useful for non-expert users * national algorithms

3. 1. Introduction – Goals A diverse, changing, uncoordinated set of input precip estimates, with various • periods of record • regions of coverage • sensor-specific strengths and limitations infraredmicrowave latency 15-60 min 3-4 hr footprint 4-8 km 5-30+ km interval 15-30 min 12-24 hr (up to 3 hr) (~3 hr) “physics” cloud top hydrometeors weak strong • additional microwave issues over land include • scattering channels only • issues with orographic precip • no estimates over snow

4. 2. TMPA – Flow Chart (1/2) Computed in both real and post-real time, on a 3-hr 0.25° grid Microwave precip: • intercalibrate to TMI/PR combination for P • intercalibrate to TMI for RT • then combine, conical- scan first, then sounders IR precip: • calibrate with microwave Combined microwave/IR: • IR fills gaps in microwave

5. Calibrate High-Quality (HQ) Estimates to “Best” Instant-aneous SSM/I TRMM AMSR AMSU MHS HQ coefficients Merge HQ Estimates 3-hourly merged HQ Match IR and HQ, generate coeffs 3-hourly IR Tb 30-day IR coefficients Apply IR coefficients Hourly HQ-calib IR precip Merge IR and HQ estimates 3-hourly multi-satellite (MS) Compute monthly satellite-gauge combination (SG) Monthly gauges Monthly SG Monthly Climo. Adj. Coeff. Rescale 3-hourly MS to monthly SG Rescaled 3-hourly MS 2. TMPA – Flow Chart (1/2) Computed in both real and post-real time, on a 3-hr 0.25° grid Microwave precip: • intercalibrate to TMI/PR combination for P • intercalibrate to TMI for RT • then combine, conical- scan first, then sounders IR precip: • calibrate with microwave Combined microwave/IR: • IR fills gaps in microwave

6. Calibrate High-Quality (HQ) Estimates to “Best” Instant-aneous SSM/I TRMM AMSR AMSU MHS HQ coefficients Merge HQ Estimates 3-hourly merged HQ Match IR and HQ, generate coeffs 3-hourly IR Tb 30-day IR coefficients Apply IR coefficients Hourly HQ-calib IR precip Merge IR and HQ estimates 3-hourly multi-satellite (MS) Compute monthly satellite-gauge combination (SG) Monthly gauges Monthly SG Monthly Climo. Adj. Coeff. Rescale 3-hourly MS to monthly SG Rescaled 3-hourly MS 2. TMPA – Flow Chart (2/2) Production TMPA • monthly MS and GPCC gauge analysis combined to Satellite-Gauge (SG) product • weighting by estimated inverse error variance • 3-hrly MS rescaled to sum to monthly SG Real-Time TMPA • 3-hrly MS calibrated using climatological TCI, 3B43 coefficients RT retrospective processing starts March 2000 • start date due to IR dataset • driven by user feedback new

7. 2. TMPA – Dominant Controls on Performance Each product (not just TMPA) should tend to follow its calibrators • over land – the GPCC gauge analysis • over ocean – satellite calibrator • climatological calibration only sets long-term bias, not month-to-month behavior • current work with U. Wash. group uncovering regional variations Fine-scale variations • land and ocean: occurrence of precipitation in the individual input datasets • inter-satellite calibration attempts to enforce consistency in distribution • event-driven statistics depend on satellites, e.g. bias in frequency of occurrence Differences between sensors tend to be noticeable • different sensors “see” different aspects of the same scene • limited opportunities to “fix” problems with the individual inputs on the fly • satellite sensors tend to be best for tropical ocean • satellite sensors and rain gauge analyses tend to have more trouble in cold areas and complex terrain

8. 3. IMERG – Introduction Want to go to finer time scale, but the “good stuff” (microwave) is sparse • 30 min of data shows lots of gaps • extra gaps due to snow in N. Hemi. • 5 imagers, 3 sounders here Aqua AMSR-E F15 SSMI N17 AMSU Aqua AMSR-E TMI F13 SSMI F14 SSMI TCI N16 AMSU 30-min HQ Precip (mm/h) 00Z 15 Jan 2005 N15 AMSU N15 AMSU N17 AMSU GPM developed the concept of a unified U.S. algorithm that takes advantage of • Kalman Filter CMORPH (lagrangian time interpolation) – NOAA • PERSIANN with Cloud Classification System (IR) – U.C. Irvine • TMPA (inter-satellite calibration, gauge combination) – NASA • all three have received PMM support Integrated Multi-satellitE Retrievals for GPM (IMERG)

9. 3. IMERG – Introduction Want to go to finer time scale, but the “good stuff” (microwave) is sparse • 30 min of data shows lots of gaps • extra gaps due to snow in N. Hemi. • 4 imagers, 3 sounders here Interpolate between PMW overpasses, following the cloud systems. The current state of the art is • estimate cloud motion fields from geo-IR data • move PMW swath data using these displacements • apply Kalman smoothing to combine satellite data displaced from nearby times Currently being used in CMORPH, GSMaP (Japan) Introduces additional correlated error Aqua AMSR-E F15 SSMI N17 AMSU Aqua AMSR-E TMI F13 SSMI F14 SSMI TCI N16 AMSU 30-min HQ Precip (mm/h) 00Z 15 Jan 2005 N15 AMSU N15 AMSU N17 AMSU GPM developed the concept of a unified U.S. algorithm that takes advantage of • Kalman Filter CMORPH (lagrangian time interpolation) – NOAA • PERSIANN with Cloud Classification System (IR) – U.C. Irvine • TMPA (inter-satellite calibration, gauge combination) – NASA • all three have received PMM support Integrated Multi-satellitE Retrievals for GPM (IMERG)

10. 3. IMERG – Heritage The Adjusted GPI (early ‘90’s) led to GPCP GPCP concepts were first used in TRMM, then blended with new multi-satellite concepts IMERG adds morphing, Kalman smoother, and neural-network concepts AGPI SGM GPCP V1 GPCP V2,2.1 SG GPCP V3 GPCP V1,1.1 1DD GPCP V1,1.1 Pentad TRMM V4,5 3B43 V6,7 TRMM 3B43 IMERG final TRMM V4,5 3B42 V6,7 TRMM 3B42 IMERG late thin arrows denote heritage IMERG early TRMM 3B42RT TRMM 3B42RT TRMM 3B42RT TRMM 3B42RT CMORPH KF-CMORPH Time PERSIANN PERSIANN-CCS

11. 3. IMERG – Notional Requirements Resolution – 0.1° [i.e., roughly the resolution of microwave, IR footprints] Time interval – 30 min. [i.e., the geo-satellite interval, then aggregated to 3 hr] Spatial domain – global, initially covering 60°N-60°S Time domain – 1998-present; later explore entire SSM/I era (1987-present) Product sequence – early sat. (~4 hr), late sat. (~12 hr), final sat.-gauge (~2 months after month) [more data in longer-latency products] unique in the field Instantaneous vs. accumulated – accumulation for monthly; instantaneous for half-hour Sensor precipitation products intercalibratedto TRMM before launch, later to GPM Global, monthly gauge analyses including retrospective product – explore use in submonthly-to-daily and near-real-time products; unique in the field Error estimates – still open for definition;nearly unique in the field Embedded metadata fieldsshowing how the estimates were computed Operationally feasible, robust to data drop-outs and (strongly) changing constellation Output in HDF5 v1.8 – compatible with NetCDF4 Archiving and reprocessing for near- and post-RT products; nearly unique in the field

12. 3. IMERG – Box Diagram The flow chart shown is for the final product • institutions are shown for module origins, but • package is an integratedsystem • “the devil is in the details” • (near-)RT products will use a cut-down of this processing GSFC UC Irvine CPC 3 Compute even-odd IR files (at CPC) 9 IR Image segmentation feature extraction patch classification precip estimation 1 Receive/store even-odd IR files 4 Compute IR displacement vectors 10 2 Build IR-PMW precip calibration Import PMW data; grid; calibrate; combine 5 Forward/backward propagation 11 Recalibrate precip rate 12 Import mon. gauge; mon. sat.-gauge combo.; rescale short-interval datasets to monthly 8 7 Apply Kalman filter Build Kalman filter weights Post-RT 13 RT Apply climo. cal. prototype6

13. 3. IMERG – Multiple Runs Multiple runs serve different users’ needs for timeliness • more delay usually yields a better product • pioneered in TMPA Early – first approximation; flood, now-casting users • current input data latencies at PPS support ~4-hr delay • truly operational users (< 3 hr) not well-addressed Late – wait for full multi-satellite; crop, flood, drought analysts • driver is the wait for microwave data for backward propagation • expect delay of 12-18 hr Final – after the best data are assembled; research users • driver is precip gauge analysis • GPCC gauge analysis is finished ~2 months after the month

14. 3. IMERG – Output Data Fields Output dataset includes intermediate data fields • users and developers require • processing traceability • support for algorithm studies 0.1° global CED grid • 3600x1800 = 6.2M boxes • files are big • but dataset compression means smaller disk files • PPS will provide subsetting “User” fields in italics, darker shading, similar to TMPA

15. 3. IMERG – Sample Day The fine time resolution is intended to provide adequate sampling for fast systems Some “flashing” illustrates that we still need to tune the coefficients

16. 3. IMERG – Probability of Liquid Phase Several GPM products are providing precipitation phase • all are diagnostic, driven by ancillary data (likely JMA forecast for RT, GANAL product for post-real time) This first example is based on Kienzle (2008) – temperature-only • Day-1 IMERG will consider surface temperature and humidity PPLP 1 March 2011

17. 3. IMERG – Testing “Baseline” code delivered November 2011 “Launch-ready” code delivered November 2012 “Frozen” code delivered September 2013 Changes to input algorithms are delaying operational testing to December 2013 • shake out bugs and conceptual problems • start quasi-operational production of “proxy” GPM data • likely we can release parallel TMPA and IMERG products PMM GV is key to • establishing calibration and confidence in the individual sensor retrievals and the IMERG processing • long-term evaluation of IMERG performance in a variety of climate zones and landform cases

18. 4. Transitioning from TRMM to GPM – Plan IMERG will be computed at launch (February 2014) with TRMM-based coefficients 6-12 months after launch expect to re-compute coefficients and run a fully GPM-based IMERG • compute the first-generation TRMM/GPM-based IMERG archive, 1998-present • all runs will be processed for the entire data record • when should we shut down the TMPA legacy code? Contingency plan if TRMM ends before GPM is fully operational: • institute climatological calibration coefficients for the legacy TMPA code and TRMM-based IMERG • continue running • particularly true for Early, Late • NEW! TRMM fuel is now forecast to last into 2016

19. 4. Transitioning from TRMM to GPM – Data Set Differences The same satellite “counts” data are used, but differences in • radiance computations (Level 1C) • retrieval algorithms

20. 5. Final Comments The TMPA continues to run until IMERG is “ready” IMERG beta (TRMM-calibrated) versions will need early test users • we expect some start-up issues, given the changes in calibration and input data Full GPM-based IMERG should be available Q4 2014 • we plan to cover the entire TRMM/GPM era in the first retrospective processing The discussion continues in the “Meet the Developer Brownbag” at 1 p.m. george.j.huffman@nasa.gov