Keywords Foliage and soil water equilibrium • Radar brightness • Soil water
content • SWAT model • Synthetic aperture radar • Vegetation cover
15.1 Introduction
Soil water content (SWC) is a measure of the total amount of water, including water
vapour, contained within a column of soil above the saturated zone above the
groundwater table. The variable is one of the most critical regarding: (i) the
duration and intensity of droughts; (ii) the production of crops; and (iii) the amount
of soil erosion and runoff. The most standard protocol in estimating SWC is based
on the “gravimetric” or “volumetric” method (Hassan et al. 2007). Both methods
estimate SWC accurately, but only provide point measurements given the extent of
work required to process a single field sample. Consequently, soil sampling fails to
provide the spatial information needed for land-management applications. Remote
sensing platforms, however, can provide greater detail of SWC at high spatial
resolution, but small enough for implementation at the land-management unit
(Hassan et al. 2007).
Radar remote sensing is an effective way of mapping SWC under adverse
weather conditions, including under cloudy conditions. Since 1991, a number of
microwave sensors have been launched (e.g., ERS, JERS, RADARSAT, and
ENVISAT), creating opportunity to study and map SWC at a multitude of spatiotemporal resolutions. Radar-based methods of estimating SWC have been broadly
classed into five main algorithm-types based on: i.e., (i) semi-empirical Synthetic
Aperture Radar (SAR) formulations, (ii) multi-temporal SAR for SWC change
detection, (iii) SAR data fusion of images from both passive and active microwave
sensors, (iv) SAR-data fusion of images from microwave and optical sensors, and
(v) SAR and microwave scattering properties (Moran et al. 2004). The method is
based on processing multiple radar images from identical passes (i.e., same polarization and incident angles), but for different times. Due to the simplicity and ease
with which the “multi-temporal SAR for SWC change detection” method can
remove artifacts created by uneven terrain and changes in vegetation cover, enormous opportunity exists with using the method in an operational setting. Table 15.1
summarizes various “multi-temporal SAR for SWC change detection”-based
methods developed over the past few years.
Here, we propose to (i) develop a new wetness index (WI) based on a chronological series of RADARSAT-1 ScanSAR images of the same area taken at
different times over the July–September period of 2005, and (ii) evaluate its
potential to estimate SWC in areas of diverse vegetation cover, including dense
forests in humid, forest-dominated landscapes of southcentral New Brunswick
(NB), Canada, and the potato-growing area of northwestern NB. In the latter
evaluation, we use SWC calculated with the Soil Water Assessment Tool
(SWAT) as field-measurements of SWC were not available to us.
302
Q.K. Hassan and C.P.-A. Bourque
content • SWAT model • Synthetic aperture radar • Vegetation cover
15.1 Introduction
Soil water content (SWC) is a measure of the total amount of water, including water
vapour, contained within a column of soil above the saturated zone above the
groundwater table. The variable is one of the most critical regarding: (i) the
duration and intensity of droughts; (ii) the production of crops; and (iii) the amount
of soil erosion and runoff. The most standard protocol in estimating SWC is based
on the “gravimetric” or “volumetric” method (Hassan et al. 2007). Both methods
estimate SWC accurately, but only provide point measurements given the extent of
work required to process a single field sample. Consequently, soil sampling fails to
provide the spatial information needed for land-management applications. Remote
sensing platforms, however, can provide greater detail of SWC at high spatial
resolution, but small enough for implementation at the land-management unit
(Hassan et al. 2007).
Radar remote sensing is an effective way of mapping SWC under adverse
weather conditions, including under cloudy conditions. Since 1991, a number of
microwave sensors have been launched (e.g., ERS, JERS, RADARSAT, and
ENVISAT), creating opportunity to study and map SWC at a multitude of spatiotemporal resolutions. Radar-based methods of estimating SWC have been broadly
classed into five main algorithm-types based on: i.e., (i) semi-empirical Synthetic
Aperture Radar (SAR) formulations, (ii) multi-temporal SAR for SWC change
detection, (iii) SAR data fusion of images from both passive and active microwave
sensors, (iv) SAR-data fusion of images from microwave and optical sensors, and
(v) SAR and microwave scattering properties (Moran et al. 2004). The method is
based on processing multiple radar images from identical passes (i.e., same polarization and incident angles), but for different times. Due to the simplicity and ease
with which the “multi-temporal SAR for SWC change detection” method can
remove artifacts created by uneven terrain and changes in vegetation cover, enormous opportunity exists with using the method in an operational setting. Table 15.1
summarizes various “multi-temporal SAR for SWC change detection”-based
methods developed over the past few years.
Here, we propose to (i) develop a new wetness index (WI) based on a chronological series of RADARSAT-1 ScanSAR images of the same area taken at
different times over the July–September period of 2005, and (ii) evaluate its
potential to estimate SWC in areas of diverse vegetation cover, including dense
forests in humid, forest-dominated landscapes of southcentral New Brunswick
(NB), Canada, and the potato-growing area of northwestern NB. In the latter
evaluation, we use SWC calculated with the Soil Water Assessment Tool
(SWAT) as field-measurements of SWC were not available to us.
302
Q.K. Hassan and C.P.-A. Bourque