Soil Moisture

Soil Properties

Soil Temperature

Vegetation and Land Cover

Aircraft Remote Sensing

Satellite Remote Sensing

Contents

1. Contact Information

2. Data Availability

3. PALS SGP99 Mission Overview

4. Data Description

5. Reference

1. Contact Information

2. Data Availability

The PALS data files from SGP99 reside on DAAC anonymous FTP. You may access them from this document,

PALS Radiometer Data (ASCII) PALS Radar Data (ASCII)

or directly via FTP at

ftp disc.gsfc.nasa.gov

login: anonymous

password: < your internet address >

cd http://disc.sci.gsfc.nasa.gov/data/sgp99/air_remote_sensing/pals

3. PALS SGP99 MISSION OVERVIEW

3.1 Objectives

The Passive/Active L/S-band sensor (PALS) was developed to study the use of multi-frequency, multi-polarization, passive and active data for remote sensing of ocean salinity and soil moisture (Wilson et al., 2001). The PALS instrument was flown for the first time during the July 1999 Southern Great Plains Soil Moisture Experiment (SGP99) near Chickasha, Oklahoma. The objectives of the PALS flights were: (1) to perform an engineering evaluation of the integrated passive-active sensor design, including noise performance, calibration stability, and radio-frequency interference (RFI); (2) to acquire simultaneous multi-polarization radiometer and radar L- and S-band data for development of improved models of land surface emission and backscatter, and for assessment of the potential of multi-channel/multi-sensor retrieval of soil moisture; and (3) in conjunction with data from C-band radiometers flown during SGP99, to compare the L-, S-, and C-band passive microwave signatures, and to optimize channel combinations over the 1- to 7-GHz frequency range, for improved sensor and algorithm design for possible future spaceborne systems.

3.2 Sensor Description

PALS was designed for high accuracy, sensitivity and stability, as driven by the ocean salinity application. The instrument operates at L band (1.41-GHz radiometer and 1.26-GHz radar) and S band (2.69-GHz radiometer and 3.15-GHz radar) with dual polarization (fully polarimetric radar). Two conical horn antennas are used (1.2-m-diameter at L band and 0.6-m-diameter at S band), preset at a fixed incidence angle between 35^o and 55^o. The instrument was designed to be flown on a C-130 aircraft, with the antennas viewing out the rear door directed downwards and to the rear of the aircraft. Views of the C-130 aircraft and the instrument mounted at the rear are shown in Figure 1. The instrument is non-scanning, thus a single-footprint track is sampled along the flight path. The key instrument characteristics are listed in Table I. Additional details of the instrument design and engineering performance are provided in Wilson et al. (2001).

The L- and S-band radar and radiometer frequencies are not identical due to separate frequency-spectrum allocations for passive and active systems. To minimize interference the radiometers incorporate narrowband filters (20 MHz at L band and 5 MHz at S band). At a nominal flight altitude of 1 km, and incidence angle of 40^o, the instantaneous 3-dB footprints of the horn antennas are approximately 300 x 400 m in all channels. The horns are shared between the radiometer and radar subsystems. The radar operates at a pulse repetition frequency (PRF) of 2.86 kHz (350 ms period) with an 8% duty cycle, switching cyclically between VV, HH, and VH transmit/receive modes. The radiometers integrate for 300 ms between the radar pulses, switching cyclically between the antenna, antenna plus noise diode, and internal reference load, first in V-polarization and then in H-polarization. The radar and radiometer outputs are averaged by the instrument data system at 0.5-second intervals. Aircraft data including location, altitude, attitude and heading, and downward-looking thermal IR temperatures, are recorded at 1-s intervals and inserted into the PALS data stream. In post-processing of the data, the 0.5-s radiometer and radar data are averaged and time-registered to the aircraft data at the 1-s intervals. Because of the relatively slow ~0.07 km s^-1 C-130 aircraft velocity, the radiometer effective integration time available at 1-km altitude is about 3 s per footprint. The resulting radiometric sensitivities (DT) per footprint are 0.15 K and 0.35 K at L and S bands, respectively. Though the antennas are time-shared between the radiometer and radar channels, the high sampling rate relative to footprint size permits collocation of the time-averaged radiometer and radar footprints. The radar pulse width places a lower limit of about 3,000 ft on the altitude at which the radar can operate reliably. Thus, in SGP99 PALS was operated predominantly at this altitude, except for the low-altitude lake radiometer calibration flights during which the radar was turned off.

Table 1

PALS Instrument Characteristics

3.3 SGP99 Site Description

A description of the SGP99 experiment, including field site descriptions and aircraft flight lines, is provided in the SGP99 Experiment Plan. Landcover conditions in July in the little Washita region included bare fields (harvested winter wheat); alfalfa, corn and other crops; grass pasture; and clumped trees (more extensive towards the eastern part of the basin). The topography is gently rolling, with large regions of both coarse and fine soil textures. The field sections are 800 m on a side.

3.4 Flight Lines

The C-130 flew at a nominal altitude of 3,000 feet over SGP99 flight lines 8, 9, 10, 11, and 12 in the Little Washita basin region near Chickasha, OK. It also flew over line 13 near El Reno, OK. The Little Washita flight lines are shown in Fig. 2. On July 8 lines 8 through 12 were flown as shown. On July 9, 11, 12, 13, and 14, line 11 was not flown but seven flight lines were added parallel to and between lines 9 and 10, to provide contiguous mapping of the rectangular region bounded approximately by latitudes 34.905 N and 34.965 N, and longitudes 98.35 W and 97.90 W. For the mapping purpose the east end of line 9 was extended to the same longitude as the east end of line 10,

4. Data Description

4.1 Data Quality

The PALS data were calibrated as described in Wilson et al. (2001). PALS data for all flight lines were georeferenced in latitude and longitude using airborne GPS location, attitude, and altitude data from the C-130 instrumentation, and knowledge of the antenna pointing geometry.

Figures 4(a) and (b) show the PALS 1.41- and 2.69-GHz brightness temperatures along flight line 9a (latitude 34.9094 N, West-to-East) on the days immediately before and after the rain event that occurred on July 10. The vertical (V) and horizontal (H) brightness temperatures are shown, along with surface (skin) temperatures from a nadir-viewing thermal infrared (IR) sensor on board the aircraft. The data shown are 1-s samples. The flight line is about 40 km in length. Each such flight line took about 10 minutes of flight time, not including aircraft turns and alignment at the ends of the lines.

The data quality exhibited in Figures 4(a) and (b) is representative of all data acquired on West-to-East flight lines. There were few problems with RFI on these lines except for localized interference observed in the L-band data when flying directly over electric power transmission lines. More RFI was observed on East-to-West flight lines at both L and S bands. Much of this is low-level interference, thought initially to be solar radiation entering the antenna sidelobes, or reflected off the land surface into the main antenna beam. This is consistent with the orientation of the antenna horns at the rear of the aircraft (pointing downward and eastward at 40^o from nadir) and the position of the sun during the morning flights. Closer investigation determined, however, that the interference occurred mostly as short spikes of 5 to 10 K or higher with a periodicity of about 5 s. It appears likely that a ground-based azimuth-scanning radar at a nearby airfield was the actual source of the interference.

Figure 4(c) shows the PALS 1.26-GHz radar backscatter data (s^o) along line 9a. The VV and VH channels are shown. The s^o data indicate an increase from July 9 to July 11of 3 to 7 dB over the western portion of the flight line (larger in the VV channel) caused by the increased soil moisture. The fluctuations in s^o caused by roughness and vegetation heterogeneity are more pronounced, relative to the soil moisture effect, than with the T_B data. This is consistent with the greater sensitivity of s^o to vegetation and roughness.

4.2 File Nomenclature

Each PALS data file is an ASCII text file containing one flight line of radiometer or radar data. The filenames have the extension '.ftr' for radiometer data and '.red' for radar data. The filename provides the date and approximate start time of the pass as follows:

'aabbccdd.ftr' (radiometer),

or

where: aa = month

bb = day

cc = hour

dd = minute

Table 2 below shows the flight lines and corresponding data files for each flight day of the mission. The letter designation of the flight lines after the hyphen refers to the general heading of the aircraft (n-north, e-east, w-west, s-south). The radiometer and radar instruments were operated by different computers, hence in some cases the start times of the data files are slightly different. "cc hour" refers to local time. Lines containing bad data, and some repeat lines (not all), have been removed from the data.

4.3 Radiometer Data Format

Within each ASCII text file the radiometer data are arranged in 12 columns as follows (Table 3):

4.4 Radar Data Format

Within each ASCII text file the radar data are arranged in 26 columns as follows (Table 4):

5. Reference

Wilson, W. J., S. H. Yueh, S. J. Dinardo, S. Chazanoff, F. K. Li, and Y. Rahmat-Samii (2001): Passive Active L- and S-band (PALS) microwave sensor for ocean salinity and soil moisture measurements, IEEE Trans. Geosci. Rem. Sens. (in press).