Operational remote sensing capabilities in water-wetland suryeys, those that can be repeated
with a probability of success of at least 85 percent, appear to be limited to those mentioned
above. A long list of promising uses can be compiled, but most of the studies from which this
list would be drawn have not been replicated. Some of the work is so qualitative as to be incon-
clusive, and some authors seem to have jumped from an image to their conclusions without adequate-
ly verifying the cause-effect relationships that created the image. Until results of such stud-
ies are replicated, quantified, and verified their results will remain as important observations
of unknown operational utility. This does not mean that these results are not important in water-
wetland research, for they are, especially when so little is known about the dynamics of wetland
circulations.
The problem just described exists in all areas of remote sensing, but may be more severe in
the water-wetland field because there are fewer basic data from which to work. Considerable
emphasis has been placed on collecting spectral reflectance and emittance data for terrestrial
vegetation, rocks, and soils for nearly fifty years, but similar data for water and wetlands are
Scarce. Difficulties of access, real concern for the fragility of wetland ecosystems, and lack of
understanding of wetland dynamics are factors that contribute to this scarcity. Use of calibrated
multispectral scanning systems (MSS) in conjunction with careful "ground" data acquisition can
help overcome these difficulties.
PROMISING SPECTRAL BANDS
The most promising spectral bands for remote sensing of water and wetlands are tabulated be-
low. The smallest band-width shown is 0.06 micrometer (um), but MSS with narrower band capabil-
ities have shown promise for detecting and classifying various contaminants at the surface.
Land-use analysis was not considered in preparing this table.
WAVELENGTH BAND
POTENTIAL USES IN REMOTE SENSING OF WATER AND WETLANDS
(in ym) CHARACTERISTICS SOME POTENTIAL USES IN REMOTE SENS
0.3 - 0.4 UV Detect bioluminescence; map oil spills; map vegetation.
0.46 Blue-Green Bathymetry to 20 meters in clear water; detect suspended chlorophyll
Green-Yellow Bathymetry; detect suspended sediment; map vegetation; track currents
Orange-Red Detect suspended sediment; track currents; detect flourescence;
determine emergent plant biomass
Reflective IR Map high turbidity areas; determine emergent plant biomass
Reflective IR Map shorelines; map vegetation boundaries
Reflective IR Discriminate snow and ice from clouds
8.0 - 13.0 Thermal IR Detect upwellings; map currents and heated effluent plumes;
delineate vegetation and wetland boundaires
100+ Microwave Determine sea state; map oil spills; determine water content and
dynamics of snow and ice fields
Both LANDSAT-1, launched during the XII ISP Congress in July 1972, and LANDSAT-2, launched in
January 1975, carry an MSS haying an instantaneous-field-of-yiew (IFOV) of 80 meters in four broad
spectral bands (0.5-0.6, 0.6-0.7, 0.7-0.8, and 0.9-1.1 ym). Four years work with LANDSAT data
have confirmed the utility of satellite remote sensing in some earth resource surveys. Informa-
tion needed by many users cannot be extracted from LANDSAT data, however, because band-width is
too wide or resolution too coarse. For example, suspended inorganic sediments are broad-band
backscatterers but chlorophyll in suspension, while highly reflective at wavelengths « ^0.52 um
is primarily absorptive between 0.52 and 0.58 um. Suspended chlorophyll cannot be consistently
discriminated from inorganic sediments in any single, or combination of, LANDSAT bands but appear
separable if both 0.46 to 0.52 and 0.52 to 0.58 um bands are available. Addition of a thermal
channel to the LANDSAT-C MSS, now scheduled for launch in 1977, will make that sensor more useful
than its predecessors but will do nothing to overcome the band-width and IFOV limitations of the
present Systems. This has been recognized and one plan for the MSS to be carried by LANDSAT-D in
the 1980's includes seven narrower bands, and an IFOV of 30 to 40 meters. Even this IFOV will t
meet the one to five meter resolution needs of many users, and satellite data will not re lace "e
aircraft or ground data for some time to come, if ever. T pers
,
SOME HUMAN PROBLEMS
to CEN Gf Tryst wextns prablens in furthering remote sensing are those attributable directly
iStics oi human beings. All too often persons famili i ing pabi
to. iar with remote sensing capabil-
ities do not know what information is i Api FR OU
needed by a potential user, and illi Fi
ic ,; ; ; and are unwilling to find out.
eM ay to convince users to use information that the remote sensing technologist knows
: ted by the fact that th king i Sensing are
he : SC ! ose working in remote sensing are
Ue enely ianerane of the realities of Socio-economic processes through which decisions Are made
ile ote sensing Systems can provide better information, that information must be what the
ecision-makers want or the information cannot be cost effective :
It 1 2 ;
rove ag Fru that some persons needing information do not know what remote sensing can
3 not see any need to find out. Sometimes this i i eri
oe 3 a ; eti is is the result of prior ex erience
ose who have over-sold remote sensing capabilities; but, some potential users are so tied
to traditional measures that the illi
y are unwilling to acce [t1
measures are inadequate to meet their needs. E PE ces) even ERSTE he tredicions)
