TRMM卫星数据介绍.pdf-资料库

qq_27421469-8609639-4744300845400565472.pdf-第1页.png

第1页 / 共22页

qq_27421469-8609639-4744300845400565472.pdf-第2页.png

第2页 / 共22页

qq_27421469-8609639-4744300845400565472.pdf-第3页.png

第3页 / 共22页

qq_27421469-8609639-4744300845400565472.pdf-第4页.png

第4页 / 共22页

qq_27421469-8609639-4744300845400565472.pdf-第5页.png

第5页 / 共22页

qq_27421469-8609639-4744300845400565472.pdf-第6页.png

第6页 / 共22页

qq_27421469-8609639-4744300845400565472.pdf-第7页.png

第7页 / 共22页

qq_27421469-8609639-4744300845400565472.pdf-第8页.png

第8页 / 共22页

University of Utah TRMM precipitation and cloud feature database Description Version 1.0 Chuntao Liu Department of Meteorology, University of Utah 135 S 1460 E, Room 809, Salt Lake City, UT 84112-0110 (o) 801-581-3336 (fax) 801-585-3681 liu.c.t@utah.edu http://www.met.utah.edu/zipser/pub/projects/trmm/ 2007. 7

Table of content Introduction 1. 2. Level-1 2.1 collocation between 1B11 and 2A25 2.2 Parallax correction 2.3 Colocation between 1B01 and 2A25 2.4 LIS data manipulation 2.5 Output parameters 3. Level-2 3.1 old definition 3.2 parameters in the old definition 3.3 New definitions 3.4 Additional parameters in the new definitions 3.5 Parameters from NCEP reanalysis 4. Level-3 4.1 precipitation data 4.2 Cloud and precipitation feature processing 4.3 Combined level-3 product 5. References 6. Appendix A. Other by-products B. Websites C. Reading software

1. Introduction The Tropical Rainfall Measuring Mission (TRMM, Kummerow et al., 1998) is a joint mission between NASA and the National Space Development Agency (NASDA) of Japan designed to monitor and study tropical rainfall. Onboard instruments including Precipitation Radar (PR), TRMM Microwave Imager (TMI), Visible Infrared Radiometers (VIRs), Cloud and Earth Radiant Energy Sensor (CERES) and Lightning Imaging Sensor (LIS) provide invaluable measurements of atmosphere. One direction of our research is to generalize the precipitation and cloud features from TRMM measurements and study the radar, passive microwave and lightning characteristics of precipitating systems in the Tropics. A database of PR and TMI rain estimates, VIRS IR brightness temperature and LIS lightning data inside and outside the PR swath in these precipitation and cloud features is constructed. Using this database, many valuable researches have been accomplished, including rainfall estimates validation (Nesbitts et al., 2004), diurnal cycle of precipitation systems (Nesbitt and Zipser, 2003), global distribution of storms with LIS-detected lightning (Cecil et al., 2005), deep convection reaching the tropical tropopause layer (Liu and Zipser, 2005), rainfall production and convective organization (Nesbitt et al. 2006), and the categorization of extreme thunderstorms by their intensity proxies (Zipser et al., 2006). This document describes the TRMM cloud and precipitation database construction procedures and output parameters in three levels of processing as shown in Figure 1.

Figure 1. Flow chart of three levels of the University of Utah TRMM feature database. 2. Level-1 As shown in Figure 1, level-1 data are produced with a combination of the version 6 orbital 1B01, 1B11, 2A12 (Kummerow et al., 2001), 2A23, 2A25 (Iguchi et al., 2000) and LIS granules after TMI-PR parallax correction and TMI-PR-LIS-VIRS nearest neighbor collocation. The output data is saved for each satellite orbit. The details of these procedures and calculated parameters are introduced in this section. 2.1 Collocation between 1B11 and 2A25 The orbit data stored in TMI 1B11 have two resolutions. One is on the low resolution (104 pixels in cross scans) for 10, 19, 21, 37 GHz channels including the brightness temperatures. Another is saved on the high resolution (208 pixels in cross scans) for

85GHz channels. The collocation between PR 2A25 and TMI 1B11 are performed only on the high TMI resolution inside PR swath. The idea is not interpolating the pixels to PR coordinates. Rather, we assign a TMI pixel to each PR pixel. The method of ― the nearest neighbor‖ is applied to assign these TMI pixels. As the result, each PR pixel has a corresponding TMI pixel. Then we save the indices of these TMI pixels for future use. The collocation for low resolution can be easily obtained by degrading the indices from high resolution grids. 2.2 Parallax correction Because TMI scans with 52o conical angle and PR scans nadir, there could be a problem if the microwave scattering signals are from elevated hydrometeors, such as high convective cells. For this reason, we used a simple parallax correction method that simply move the TMI data coordinates data backwards for one scan shown as Figure 2. After this correction, there are better correspondences between PR and TMI measurements for high convective cells. However, the correspondences between PR and TMI for shallow precipitations become worse because of the overcorrection. This could lead to problems when calculate the microwave scattering properties inside a shallow precipitation system defined by PR surface rainfall area. Figure 2. Schematic diagram of parallax correction. 2.3 Collocation between 1B01 and 2A25 Since VIRS scans in nadir, it is relatively easier to collocate VIRS data with PR data. We simply applied the nearest neighbor to degrade the VIRS radiance data onto PR pixel coordinates. Then the brightness temperatures at five VIRS channels at each PR pixel are calculated from radiances at the nearest VIRS pixel.

2.4 LIS data manipulation LIS data collocation method was developed by Chris West and Dan Cecil in 1999. First, the observation view time is interpolated into 0.1 degree resolution, then use nearest neighbor method to assign each one of the flash event to TMI pixel coordinates. Following the collocation between PR and TMI, each flash can be assigned to a PR pixel for further analysis. 2.5 Output parameters We have chosen some interesting parameters from 1B01, 1B11, 2A25, 2A12, and some derived parameters for storing into the level-1 products. These parameters include: Parameters from PR 2A25 Orbit 1 integer version 1 float rays 1 integer, 49 scans 1 integer year Float array (scans) month Float array (scans) day Float array (scans) hour Float array (scans) minute Float array (scans) second Float array (scans) lon Float array (rays, scans) lat Float array (rays, scans) Rangebinnum* Float array (7,rays,scans) nearsurfz Float array (rays,scans) Orbit number Version number Number of rays in each scan Number of scans in the orbit Year Month Day Hour Minute Second Longitude Latitude Range bin number Near surface reflectivity (0.01 dBZ) Near surface rain (mm/hr) Path integrated attenuation Z-R retrieval method Z-R retrieval parameters scan indices of pix with echoes Ray indices of pix with echoes nearsurfrain Pia* method Zrparamnode* Scan# Ray# Pr_dbz# Float array (rays,scans) Float array (3,rays,scans) Float array (rays,scans) Float array (5,rays,scans) Float array (valid scans) Float array (valid scans) Float array (valid scans, 80) Reflectivity profiles with echoes Colohi^ Float array (rays,scans) (0.01 dBZ) Indices of TMI pixels for each PR pixels *Detail see interface control specification TSDIS.MDL-02.5 volume 4, 1-20 # In order to reduce the file size, we only save the reflectivity profiles with valid echoes. For example, one may use lon[ray[i], scan[i]] to obtain the longitude of the reflectivity profiles pr_dbz[i,*]. ^ These indices can be used to find the collocated TMI measurements for each PR

pixels. For example, one may use tmi.rain[colohi[I,j]] to find the 2A12 rainfall estimates for PR pixel (i,j) at longitude lon[I,j] and latitude lat[I,j]. Parameters from PR 2A23 1 float version Raintype2a23 Integer array (rays, scans) HBB2A23 HFREEZ2A23 Stormh Integer array (rays, scans) Integer array (rays, scans) Integer array (rays, scans) Version number Rain type 100-153: strat 200-293: convective Height of bright band (m) Height of freezing level (m) Storm height (m) *Detail see interface control specification TSDIS.MDL-02.5 volume 4, 1-9 Parameters from PR 1B01 boost Ch1 Ch2 Ch3 Ch4 Ch5 Lon Lat Ch4_rain 1 integer Float array (rays, scans) Float array (rays, scans) Float array (rays, scans) Float array (rays, scans) Float array (rays, scans) Float array (261,*) Float array (261,*) Float array (261,*) 0: before, 1: after boost 0.63 micron TB at PR pixels (K) 1.6 micron TB at PR pixels (K) 3.75 micron TB at PR pixels (K) 10.8 micron TB at PR pixels (K) 12.0 micron TB at PR pixels (K) Longitude of full VIRS swath (K) Latitude of full VIRS swath (K) 10.8 micron TB of full VIRS swath (K) Orbit number Version number Number of high res rays in each scan Number of low res rays in each scan Number of scans in the orbit Year Month Day Hour Minute Second Parameters from 1B11 1 integer Orbit version 1 float 1 integer (208) hiRays 1 integer (104) loRays Scans 1 integer Float array (scans) year Float array (scans) month day Float array (scans) Float array (scans) hour Float array (scans) minute second Float array (scans) Float array (hirays, scans) High resolution longitude lonHI Float array (hirays, scans) High resolution latitude latHI Lonlo Float array (lorays, scans) low resolution longitude low resolution latitude Float array (lorays, scans) latlo 10 GHz vertical polarization TB (K) Float array (lorays,scans) V10 H10 Float array (lorays,scans) 10 GHz horizontal polarization TB (K) 19 GHz vertical polarization TB (K) Float array (lorays,scans) V19 H19 Float array (lorays,scans) 19 GHz horizontal polarization TB (K)

V21 V37 H37 V85 H85 Float array (lorays,scans) Float array (lorays,scans) Float array (lorays,scans) Float array (hirays,scans) Float array (hirays,scans) 21 GHz vertical polarization TB (K) 37 GHz vertical polarization TB (K) 37 GHz horizontal polarization TB (K) 85 GHz vertical polarization TB (K) 85 GHz horizontal polarization TB (K) TMI surface rainfall (mm/hr) Surface flag 37 GHz polarization corrected TB (K) 85 GHz polarization corrected TB (K) Float array (hirays,scans) Float array (hirays,scans) Confidence of retrieval Float array (hirays,scans) Float array (hirays,scans) Float array (hirays,scans) Float array (valid pix, 14) Cloud water profile (mg/m3) Float array (valid pix, 14) Cloud ice profile(mg/m3) Parameters from PR 2A12 Rain Confidence* Surfaceflag* PCT37 PCT85 Cld_water* Cld_ice* Precip_water* Float array (valid pix, 14) Precipitation water profile (mg/m3) Float array (valid pix, 14) Precipitation ice profile (mg/m3) Precip_ice* Profile_ray# Indices of rays with surface rain Float array (valid pix) Profile_scan# Float array (valid pix) Indices of scans with surface rain *Detail see interface control specification TSDIS.MDL-02.5 volume 4, 1-1 # In order to reduce the file size, we only save the hrdrometeor profiles over rainfall area. For example, one may use lonhi[profile_ray[i],profile_ scan[i]] to obtain the longitude of the cloud water profiles cld_water[i,*]. Above parameters are saved into ―HDF‖ format with naming rules as ―1Z99.yymmdd.orbit.version.HDF‖, and there is an IDL program ―read_pf_level1_hdf.pro‖ for access these level-1 files. 3. Level-2 The first step to create the level-2 data is to define the features. There are two groups of feature definitions with development of the database. The old-definition (1999-2005, Nesbitt et al., 2000) is a ―hybrid definition‖ using information from both PR and TMI. The new definition was developed recently (Sep 2006, Liu et al., 2007) by using ―pure‖ information from individual measurements. Currently all TRMM data are processed with both groups of definitions. This section will introduce these definitions separately. 3.1 Old definition The first TRMM Precipitation Feature (PF) was developed by Dan Cecil, Steve Nesbitt and Ed Zipser around 1998-1999 (Nesbitt et al., 2000). The concept was to use the information from both TMI and PR, and defined the PFs with area of PR pixels with 20 dBZ at near surface or TMI 85GHz Polarization Corrected Temperature (PCT, Spence et al., 1989) colder than 250 K. Then summarize the precipitation, convective properties inside the PF area. By using this definition, many valuable

资料库

TRMM卫星数据介绍.pdf

相关推荐

课程资源

热门标签

最新资料