Microwave Integrated Retrieval System (MIRS) - Validation

Comparison to Radiosonde

Daily collocations to radiosonde observations (RAOBs) are performed for each of the satellite sensors which MIRS processes. The entire set of daily RAOBs are provided by the NOAA PROducts and Validation System (NPROVS) RAOBs which fall within 50km and 2 hours of a satellite observation are accumulated in an archive, along with the collocated satellite observation itself. This section displays the results of comparing the MiRS retrievals with collocated RAOBs, including temperature sounding, humidity sounding and TPW. Other data which are included in the NPROVS dataset and collocated to the RAOBs are also compared, including ATOVS, COSMIC, and GFS forecasts.
For NOAA-18, Metop-A, and DMSP F16 MiRS, the record of collocations extends from April 2008 - present. For NOAA-19, the record extends from March 2009 - present.

Comparison to HIRS

Along with radiosonde observations, the NPROVS dataset includes HIRS skin temperature. MiRS skin temperature is compared to HIRS at all points where both are collocated to RAOBs. The record of collocations is equivalent to that noted in the Comparison to Radiosonde section.

Comparison to CloudSat

This section has been implemented to assess the performance of the MiRS hydrometors, including the cloud liquid water path, rain water path, ice water path as well as the rainfall rate, using the hydrometeors retrieved as part of the CloudSat mission. The thresholds for collocation criterion are 50km and 3 minutes. All CloudSat rainfall rates (1.5 km resolution) that fall within this range for a given MiRS field of view are averaged to be more representative of the overall conditions within the MiRS footprint (50km at nadir for POES/Metop and 70km at nadir for DMSP F16).

Comparison to Stage IV

The hourly NCEP Stage IV precipitation analysis has been used as a reference for the assessment and evaluation of the MiRS rainfall rate technique. Produced by the 12 River Forecast Centers over the Continental U.S. (CONUS), the NCEP Stage IV product is a 4-km resolution precipitation analysis derived from hourly radar precipitation estimates and hourly rain gauge data. The Stage IV precipitation analysis product is much more of an integrated hourly estimate, while the satellite-based MiRS precipitation estimation products represent an instantaneous observation. Because of that, during the comparison, it has been assumed that the Stage IV rainfall rate is a constant amount for an entire hour and is compared with the MiRS precipitation estimates that had occurred within that hour.

Comparison to TRMM

This section contains results from comparing MIRS rainfall rate to collocated TRMM 2B31 rainfall rate. The thresholds for collocation criterion are 50km and 2 minutes. All TRMM rainfall rates (5km resolution) that fall within this range for a given MiRS field of view are averaged to be more representative of the overall conditions within the MiRS footprint (50km at nadir for POES/Metop and 70km at nadir for DMSP F16).
For NOAA-18, Metop-A, and DMSP F16 MiRS, the record of collocations extends from December 2008 - present. For NOAA-19, the record extends from March 2009 - present.
**Note: Due to an instrument anomaly with TRMM PR, the period between June 2009 - present contains no collocations.

Comparison to TRMM_2A12

This section contains results from comparing MIRS rainfall rate and integrated hydrometeor amounts to collocated TRMM 2A12 data. The thresholds for collocation criterion are 50km and 2 minutes. All TRMM scenes (5km resolution) that fall within this range for a given MiRS field of view are averaged to be more representative of the overall conditions within the MiRS footprint (50km at nadir for POES/Metop and 70km at nadir for DMSP F16). Here, comparisons are made to TRMM rainfall rate, integrated rain water path, ice water path, and cloud liquid water. When clicking the Profiles button, panels containing the MiRS hydrometeor profiles and collocated TRMM profiles are displayed, along with the derived rainfall rate found for the scene. The TRMM profiles are interpolated to the MiRS vertical pressure grid. To change the date, parameter, or surface type, select from the dropdown menus and then click Submit Query to display the results. For NOAA-18, NOAA-19, Metop-A, and DMSP F16 MiRS, the record of collocations extends from July 2009 - present.

IPWG Project

The IPWG is dedicated to perform large-scale validations and intercomparisons of daily rainfall estimates from operational and semioperational satellite precipitation estimates against "ground truth" precipitation references based on the rain gauge and/or ground-based radar observations. More information about the IPWG validation/intercomparison can be found at IPWG validation/intercomparison

Comparison to Rain Gauge

For assessment purposes, the MiRS precipitation composite estimate given in millimeters per day has been generated using the rainfall rate samples (given in millimeters per hour) from three MiRS precipitation estimation products: MiRS NOAA-18, Metop-A, and DMSP-F16. Computed from -60 to 60 degree latitudes and with a spatial resolution of 0.25 degree, the daily MiRS precipitation composite has been included as part of the IPWG validation/intercomparison. Currently, the daily MiRS precipitation composite estimate is being validated over CONUS, South America, and Australia, and its incorporation to other regions is in progress. For the validation over CONUS and South America, the daily MiRS precipitation composite estimate extends from 12:00 UTC to 12:00 UTC, while for the validation over Australia, the MiRS composite extends from 00:00 UTC to 24:00 UTC. In this section, rain gauge observations are used as the precipitation reference.

Comparison to Rain Radar

In this section, comparisons between microwave-based precipitation algorithms and rain radar observations are presented. This section is complementary of the Comparison to Rain Gauge validation section.

Comparison to CPC

This section presents comparisons between the daily NOAA/NCEP Climate Prediction Center (CPC) precipitation (based on rain gauge analysis) and the daily MiRS precipitation estimate given in millimeters per day and generated from the instantaneous rainfall rates derived from MiRS NOAA-18, Metop-A, and DMSP-F16 rainfall rate products. In this comparison, the limited temporal and spatial coverage associated with the satellite sensors used to generate the daily MiRS precipitation estimate product must be considered, compared to that of the rain gauges. This comparison is limited to the CONUS.

Performance Summary

This section presents some MiRS peformance summaries.