Remote Sensing

Remotely sensed data will be very useful for monitoring the evolution of SCS projects through spatiotemporal assessments of vegetation (type, cover, and productivity), water (soil moisture), and energy (soil temperature). In SCS projects, the soil surface is likely going to be covered most of the time by live or dead vegetation (crops, plant litter). Thus, remote sensing cannot be used for direct measurements of soil carbon stocks (Merry and Levine, 1995; vanDeventer et al., 1997) unless the soil is bare, and correlations among soil reflectance, soil color, and SOM concentration can be established (Bhatti et al., 1991; Chen et al., 2000; Henderson et al., 1989). Merry and Levine (1995) found a negative correlation (r = -0.63) between NDVI values derived from the AVIRIS (airborne visible infrared imaging spectroradiometer) sensor and soil carbon density determined from a detailed soil survey (scale 1:12,000) conducted in Orono, Maine. The negative sign of the correlation coefficient suggested that certain types of vegetation with low NDVI values were more adapted to grow in places with enhanced levels of soil carbon densities. (NDVI, the normalized difference vegetation index, is calculated as the ratio of the difference between the near-infrared and red bands over the sum of these bands. Use of the NDVI allows for meaningful comparisons of seasonal and interannual changes in vegetation growth and activity.)

A number of satellite and airborne sensors are capable of gathering data useful for estimating LAI, net primary productivity (NPP), crop yields, and litter cover (Table 19.2). For many years, the AVHRR (Advanced Very High Resolution Radiometer) and Landsat sensors have been the traditional sources of land cover data. More recently, the MODIS (MODerate-resolution Imaging Spectroradiometer) sensor onboard the Terra satellite has been obtaining data on land cover with a spatial and spectral resolution that was not possible with AVHRR. MODIS and AVHRR are able to provide data at rather frequent temporal resolution and a good spatial resolution for monitoring vegetation growth during the growing season. Conversely, the Landsat and SPOT (Système Probatoire pour l'Observation de la Terre) sensors allow for excellent spatial detail of land cover/crop identification, but might not the best sensors for monitoring seasonal vegetation dynamics. The commercial sensors IKONOS and Quickbird offer excellent spatial and temporal resolution, and could be very helpful for agricultural applications and SCS projects. Two airborne sensors might be particularly useful for SCS projects: AVIRIS and LIDAR (light detection and ranging) (Table 19.2).

Hyperspectral images provide spectral images that resemble those obtained by laboratory spectroscopic instruments (Shippert, 2004). Because they are overdetermined, hyperspec-tral images allow for the identification and distinction of spectrally similar, but unique, materials (Shippert, 2004). The remote sensing community has been advancing methodologies to interpret hyperspectral imagery, including the development of libraries for identifying minerals, substances, and vegetation (http://rst.gsfc.nasa.gov/Intro/Part2_24.html). The AVIRIS sensor, aboard the NASA-ER1 plane, has been providing hyper-spectral data of immense value since 1987 (Table 19.2). Another hyperspectral sensor is Hyperion (http://eo1.gsfc.nasa.gov/ technology/hyperion.html), part of EO-1, in operation since December 2000, which serves as testing unit for advancing the development of spaceborne hyperspectral instrumentation.

Remote sensing data have also been used with simulation models to derive spatially explicit estimates of cropland NPP

Table 19.2 Temporal and Spatial Resolutions of Various Satellite and Airborne Sensors Collecting Data Useful for Land Cover/Land Use Classifications, Vegetation Monitoring, and Topographic Mapping

Instruments,

Spectral

Spatial

Temporal

Sensor in

Use

Number of

Bands

Resolution/Swath

Resolution

Operation Since

Category

Bands

(nm)

(m/km)

(d)

Websites and Special Features

Satellite

AVHRR, >20

Operational

Multispectral, 2 + 3

580-1100

1000/2485

1

http://edcsns17.cr.usgs.gov/

years

thermal

bi-weekly

EarthExplorer/

composites

MODIS,

Research

Multispectral, 2

620-876

250

2

http://modis.gsfc.nasa.gov/

December

Multispectral, 5

459-2155

500

2

1999

29 bands for

1000/2330

10

clouds, O3, etc.

ASTER

Research

Multispectral, 3

520-860

15/60

Pointing

http://www.science.aster.ersdac.

December 1999

Multispectral, 6

1600-2430

30/60

or.jp/en/science_info/index.

html

Landsat MSS

Operational

Multispectral, 4

500-1100

80

18

http://landsat7.usgs.gov/index.

Landsat TM

Multispectral, 6

450-2350

30

16

php

Landsat ETM+,

Multispectral, 6

450-2350

30

http://edcsns17.cr.usgs.gov/

> 30 years

Panchromatic

520-900

15

EarthExplorer

SPOT (2, 4), 13

Commercial

Multispectral, 4

500-1750

20/60

Pointing

http://www.spotimage .fr/html/_

years; SPOT

Panchromatic, 1

480-710

2.5-10

1

167_.php

5, May 2002

Stereoscopic

490-690

10

http://www.spot-vegetation.com/

Vegetation

450-1750

1000

Hyperion

Research

Hyperspectral, 220

400-2500

30/7.5

Pointing

http://eo1.gsfc.nasa.gov

/technology/hyperion.html

IKONOS,

Commercial

Multispectral, 4

450-880

4/7

3 pointing

http://www.spaceimaging.com/

September

Panchromatic, 1

450-900

1

QuickBird, October 2001

AVIRIS >16

years AISA

LIDAR, commercial since 1993

Commercial

Research Commercial

Research, Commercial

Multispectral Operational

Multispectral, 4 Panchromatic

450-900

Airborne

Hyperspectral, 224 380-2550

Hyperspectral, varies

User selectable

Active laser sensor, 1045-1065 1

Similar to Landsat 400-900 TM

20/11 1

Variable

Pointing

http://www.spaceimaging.com/

http://aviris.jpl.nasa.gov/

http://www.specim.fi/ (Available from several companies)

> 0.75 horizontal Dependent on http://ltpwww.gsfc.nasa.gov/eib/ 0.15 vertical flight projects/airborne_lidar/lvis/

schedule Several companies provide airborne LIDAR services ~0.25 adjustable Dependent on Several companies provide flight multispectral service using schedule instruments of varying sophistication r

(Prince et al., 2001; Lobell et al., 2002). Satellite estimates of residue cover (plant litter) would also be useful for assessing adoption of conservation practices (no till) and calibrating models. Daughtry et al. (2004) determined spectral reflectance characteristics of dry and wet crop residues and soils over the 400- to 2400-nm wavelength region. Since crop residue cover was linearly related to a cellulose absorption index, and this to NDVI, Daughtry et al. (2004) proposed a method to estimate soil tillage intensity based on these two indices. They concluded that it was possible to use advanced multispectral or hyperspectral imaging systems to conduct regional surveys of conservation practices.

0 0

Post a comment