An Introduction to TRMM and

its Precipitation Radar (PR)

Arash Mashayekhi
CASA REU Program
Sandra Cruz-Pol, Assoc. Prof. ECE UPRM

The Big Picture

Why TRMM?
Tropical Rain Measurement Mission
tropical rainfall Drives the Climate Machine
Need to understand the Water Cycle
TRMM: the first space-borne rain radar (PR)
and microwave radiometric data

About TRMM:
Launched:
November 28, 1997

Circular Orbit altitude:

350 km
Inclination:
approx. 35 deg.
Orbit Duration:
91 minutes (16 Orbits a day)
Time Spent over Puerto Rico during
each orbit:
1.14 minutes
Total Time spent over Puerto Rico
Each Day:
18.2 minutes

TRMM Primary Instruments for

Measuring Precipitation:
1.
2.
3.

Precipitation Radar (PR)

TRMM Microwave Imager (TMI) radiometer
Visible and Infrared Scanner (VIRS)
Two Additional Instruments:
1. Cloud and Earth Radiant Energy Sensor
(CERES)
2. Lightning Imaging Sensor (LIS)

Microwave Imager

Introduction:

passive microwave sensor designed to

provide quantitative rainfall information
TMI

Provides Valuable Information on:

Specifications:
Quantity
of the water vapor,
Quantity
of the10.65
cloud to
water
Frequency:
85.5 GHz
Intensity of the rainfall in the
Horizontal Resolution: 6 to 50 km
atmosphere.
Swath Width: 760 km

Visible and Infrared Scanner

Introduction:

senses radiation coming up from the Earth in

five spectral regions, ranging from visible to
infrared

It is used to:

Specifications:
Delineate rainfall
Determine
Wavelength:
.63 tothe
12brightness
um
(visibleResolution:
and near2infrared)
or
Horizontal
km
temperature
of the
Swath
Width: 720 (infrared)
km
source emitting radiation

Cloud and Earth Radiant

Energy Sensor
o

Introduction
o

The data from the CERES instrument will be used to study the
energy exchanged between the Sun; the Earths atmosphere,
surface and clouds; and space.

Specifications:
Gathers information on:
Wavelength: .5 to 50 um
Cloud propertiesCloud
Effects
Horizontal
Resolution: 10 km
cloud-amount,
Swath Width:
+ 80thickness,
degreesand
altitude,
the size of the cloud
particles

Lightning Imaging Sensor

Introduction:

The Lightning Imaging Sensor is a

small, highly sophisticated instrument
that will detect and locate lightning
over the tropical region of the globe.
the sensor will provide
Specifications:
information that could lead to
advanced
future
Wavelength:
.77765lightning
mm
sensors capable of significantly
Horizontal Resolution: 4 km
improving weather
"nowcasting."
Swath Width: 600 km

Precipitation Radar

Introduction:

The Precipitation Radar is the first active space borne radar

designed to provide three-dimensional maps of storm structure

PR will provide valuable information on:

Rain size, speed, and altitude
Intensity and distribution of the rain
Rain type
Storm depth
Melting layer altitude: The height at which
snow melts into rain

Precipitation Radar

Specifications:
o

Frequency : 13.8 GHz (Ku-band)

o

More than four times higher than that of a typical ground based
radar (NEXTRAD ~ 3 GHz, S-band)

Horizontal Resolution: 4.3 km

Swath Width: 215 km

Vertical Profile of Rain and Snow: 19.3 km

o
o

Able to detect rainfall rate down to .7 millimeters/hr

Able to separate vertical rain echo samples of 250 meters

Precipitation Radar

Specifications (Contd):

Power Consumption: 224 W

Solid state power amplifiers (128) are used to conserve power

Target Area:

phased array antenna that steers the beam electronically

Precipitation Radar

TRMM Precipitation Radar Algorithm

Level 1

IB21
IC21
Level 2

2A21
2A23
2A25

Level 3
3A25
3A26

TRMM Precipitation Radar Algorithm

Level 1 (IB21, IC21)

IB21
Calculates

received power by performing extensive

internal calibrations

Data in IB21 include:

Location

of Earth surface and surface clutter

System noise level
Land/Ocean Flag
And many more

TRMM Precipitation Radar Algorithm

Some Examples of IB21 Data:

Navigation

X, Y, Z Components of Space Craft Velocity and Position

Latitude
Longitude
Altitude
Sensor Orientation

Min. Echo Flag

0 : No Rain
10: Rain possible but maybe noise
20: Rain Certain
Land / Ocean Flag
0: Water
1: Land

TRMM Precipitation Radar Algorithm

Level 1 (IB21, IC21)

Output:

Radar Reflectivity Factor

Almost same file format as that of IB21:

Power replaced by Radar Reflectivity Factor

Noise replaced by Dummy Variable

Level 2 (2A21, 2A23, 2A25)

Primary

Objective:

Compute Path Integrated Attenuation (PIA) using the

Surface Reference Techniques (SRT).

Input

Data: IB21
Output used by: 2A25, 3A25, and 3A26

TRMM Precipitation Radar Algorithm

Level 2 (2A21, 2A23, 2A25)

Main Objectives:

Classification of Rain Types

Output of Rain / No Rain Flag
Computation of estimated height of freezing level
Output of the height of storm top

Input Data: IC21

Output used by: 2A25, 2B31, 3A25, 3A26

Level 2 (Contd) (2A21, 2A23, 2A25)

Main Objectives:
Input Data: IC21, 2A21, 2A23
Output used by: 3A25, 3A26

Correct for the Rain Attenuation in measured Radar Reflectivity

Estimate instantaneous 3-D distribution of rain

TRMM Precipitation Radar Algorithm

Level 3 (3A25, 3A26)

Objective:

calculate various statistics over a month from the level 2

Four types of statistics are calculated:

1.
2.
3.
4.

probabilities of occurrence
means and standard deviations
histograms
correlation coefficients

Level 3 (3A25, 3A26)

Objective:

Compute rain rate statistics

Compared to 3A25

statistics produced from 3A25 are conditioned either on the presence of

rain or on the presence of a particular type of rain but statistics from 3A26
are unconditioned.

Data for Rain event

We requested data for a strong rain event
that occurred in Puerto Rico last May
2004.
Dates May 14, 15, 21
We have corresponding data for NWS
NEXRAD in Cayey, PR and rain gauges
around the island.
Our goal is to compare these data sets

How does the data look like?

Data files are huge: 30MB for each 1.1
minute. Total of over 1GB for the event.
There are several (~20) products

Ave rain
Near surface rain
Sigma zero
Rain flag
Zeta
PIA

Ave Rain: Digital Array Viewer

Sigma 0: Digital Array viewer

Need to Filter

We only need
Near surf rain
Quality flag
?

And of course Latitude/Longitude, Date,

Time to map over Puerto Rico
This filtering should considerably reduce
the data file sizes.

Rain algorithm

Once we filter the data

Need to develop code in IDL to convert to
arrays in text
Compare actual rain algorithm being used by
NWS. The Rosenfelt tropical convective

Z 250R

1.2

PR Rain Characterization

Look at different algorithms per region

Elsner & Carter, 2000 ; Vasquez & Roche,
1997 suggest that the island be divided into
~6 rain regions each with a different
algorithm for 3 seasons.

Tropical Environment
Tropical weather is especially
difficult to forecast due to
several factors including:

Easterly trade winds caused

forced convection
Complex topography of the
island

In the fall, we plan to use CSU

disdrometer to help further
characterize rainfall in PR.

Credits
TRMM Official Website
TRMM Education and Outreach Scientist :
Dr Jeffrey B. Halverson
Responsible NASA Official:
Dr.Robert Adler
http://trmm.gsfc.nasa.gov/

NASA Official Website

Editor: Jim Wilson
NASA Official: Brian Dunbar
Last Updated: July 6, 2004
http://www.nasa.gov/home/index.html

Japan Aerospace Exploration Agency (JAXA) Official Website

http://www.jaxa.jp/index_e.html

National Space Development Agency of Japan (NASDA) Official Website

http://www.nasda.go.jp/index_e.html
Tropical Rainfall Measuring Mission
TRMMPrecipitation Radar Algorithm Instruction Manual For Version 6