[4]. Freshwater wetlands can be an important source of groundwater recharge and
water in arid and semiarid areas [87]. However, wetlands have been exposed to a
wide range of stress-inducing alterations, including dredge and fill operations,
hydrologic modifications, pollutant runoff, eutrophication, impoundments, and
fragmentation by roads and ditches.
For more than three decades, remote sensing techniques have been used effectively to detect, map, and monitor tidal wetlands [88–92]. For instance, the US Fish
and Wildlife Service (FWS) through its National Wetlands Inventory has provided
federal and state agencies and the private sector with scientific data on wetlands
location, extent, status, and trends. To accomplish this important task, FWS has
used multiple sources of aircraft and satellite imagery and on-the-ground observations [92]. Most states have also created a wide range of tidal wetland inventories,
using both aircraft and satellite imagery. The aircraft imagery frequently included
natural color and color infrared images. The satellite data consisted of both highresolution (1–4 m) and medium-resolution (10–30 m) multispectral imagery.
The Landsat TM has been a reliable source for land cover data [93]. Its 30 m
resolution and spectral bands have proven adequate for observing land cover
changes over large areas (e.g., the Horn of Africa). Freshwater wetlands have
been mapped using Landsat TM and other medium-resolution data [94–97]. The
availability of high spatial resolution (0.4–4.0 m) satellite data has significantly
improved the capacity for mapping isolated and upstream freshwater wetlands [89,
98, 99]. However, the cost per sq. km. of imagery and its analysis increases very
rapidly from using medium-resolution to high-resolution imagery. Therefore, large
wetland areas or entire watersheds should be mapped using medium-resolution
sensors, such as Landsat TM at 30 m, and only small, critical areas should be
examined with high-resolution sensors, such as IKONOS at 1–4 m resolution [98].
Airborne geo-referenced digital cameras providing color and color infrared
digital imagery are particularly suitable for accurate mapping of small freshwater
wetland sites or interpreting satellite data. For example, in Fig. 6, the wetlands map
shown on the left was derived from an airborne ADS-40 digital camera image
shown on the right. At a spatial resolution of 0.5 m, the ADS-40 digital imagery was
able to identify three key species of marsh vegetation (i.e., Phragmites, Typha, and
Spartina). Digital cameras are often used on small aircraft flown at low altitudes
(e.g., 200–500 m) and can be used to guide and supplement field data collection [98,
100]. Most digital cameras are capable of recording reflected visible to nearinfrared light. In some cases digital camera spectral bands can be matched with
specific satellite imaging band, e.g., blue, green, red, and near-infrared bands
matching the bands of the IKONOS satellite multispectral imager [101]. Digital
camera imagery can be integrated with GPS information and used with geographic
information system software for a wide range of modeling applications [102].
Groundwater discharges in wetlands can also be identified by the unique vegetation species they may support. For example, in Spartina alterniflora tidal marshes
groundwater discharges have been identified because they decrease the local
salinity and thus attract other species to grow, such as invasive Phragmites australis
Using Remote Sensing to Map and Monitor Water Resources in Arid and Semiarid. . .
49
water in arid and semiarid areas [87]. However, wetlands have been exposed to a
wide range of stress-inducing alterations, including dredge and fill operations,
hydrologic modifications, pollutant runoff, eutrophication, impoundments, and
fragmentation by roads and ditches.
For more than three decades, remote sensing techniques have been used effectively to detect, map, and monitor tidal wetlands [88–92]. For instance, the US Fish
and Wildlife Service (FWS) through its National Wetlands Inventory has provided
federal and state agencies and the private sector with scientific data on wetlands
location, extent, status, and trends. To accomplish this important task, FWS has
used multiple sources of aircraft and satellite imagery and on-the-ground observations [92]. Most states have also created a wide range of tidal wetland inventories,
using both aircraft and satellite imagery. The aircraft imagery frequently included
natural color and color infrared images. The satellite data consisted of both highresolution (1–4 m) and medium-resolution (10–30 m) multispectral imagery.
The Landsat TM has been a reliable source for land cover data [93]. Its 30 m
resolution and spectral bands have proven adequate for observing land cover
changes over large areas (e.g., the Horn of Africa). Freshwater wetlands have
been mapped using Landsat TM and other medium-resolution data [94–97]. The
availability of high spatial resolution (0.4–4.0 m) satellite data has significantly
improved the capacity for mapping isolated and upstream freshwater wetlands [89,
98, 99]. However, the cost per sq. km. of imagery and its analysis increases very
rapidly from using medium-resolution to high-resolution imagery. Therefore, large
wetland areas or entire watersheds should be mapped using medium-resolution
sensors, such as Landsat TM at 30 m, and only small, critical areas should be
examined with high-resolution sensors, such as IKONOS at 1–4 m resolution [98].
Airborne geo-referenced digital cameras providing color and color infrared
digital imagery are particularly suitable for accurate mapping of small freshwater
wetland sites or interpreting satellite data. For example, in Fig. 6, the wetlands map
shown on the left was derived from an airborne ADS-40 digital camera image
shown on the right. At a spatial resolution of 0.5 m, the ADS-40 digital imagery was
able to identify three key species of marsh vegetation (i.e., Phragmites, Typha, and
Spartina). Digital cameras are often used on small aircraft flown at low altitudes
(e.g., 200–500 m) and can be used to guide and supplement field data collection [98,
100]. Most digital cameras are capable of recording reflected visible to nearinfrared light. In some cases digital camera spectral bands can be matched with
specific satellite imaging band, e.g., blue, green, red, and near-infrared bands
matching the bands of the IKONOS satellite multispectral imager [101]. Digital
camera imagery can be integrated with GPS information and used with geographic
information system software for a wide range of modeling applications [102].
Groundwater discharges in wetlands can also be identified by the unique vegetation species they may support. For example, in Spartina alterniflora tidal marshes
groundwater discharges have been identified because they decrease the local
salinity and thus attract other species to grow, such as invasive Phragmites australis
Using Remote Sensing to Map and Monitor Water Resources in Arid and Semiarid. . .
49