140
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
7.1 INTRODUCTION
7.1.1 Motivation of Study
The occurrence of historical droughts led to studies on their impact and assessment
methods. Droughts differ from most natural hazards in several important ways:
(1) a slow-onset, creeping phenomenon occurs; (2) duration varies from event to
event; (3) there is no universal definition; (4) no single drought index can identify
precisely the onset and severity of the event; (5) spatial extent can be much greater
than that of other natural hazards, making assessment difficult; (6) the core area or
epicenter can change over time, reinforcing the need for continuous monitoring; and
(7) impacts are generally difficult to quantify with cumulative effects. In particular,
monitoring these phenomena in a fast-growing urban region where the multitemporal changes of land use and land cover (LULC) can affect holistic drought assessment is a challenge (Tadesse et al. 2005).
The early quantitative indices based on climatic and meteorological observations
include the Palmer drought severity index (PDSI; Palmer 1965), rainfall anomaly
index (van Rooy 1965), and Palmer crop moisture index (Palmer 1968). Current
drought measurement relies on biophysical parameters such as vegetation indices
(VIs), land surface temperature (LST), soil moisture, albedo, and evapotranspiration (ET). Vegetation health is an essential indicator, and vegetation cover was once
considered a good surrogate index for drought monitoring (Tadesse et al. 2005). The
most frequently used VIs are the normalized difference vegetation index (NDVI;
Rouse et al. 1974), the soil-adjusted vegetation index (SAVI; Huete 1988), the modified SAVI (MSAVI; Qi et al. 1994), and the enhanced vegetation index (EVI; Huete et
al. 1999). With the aid of NDVI, other vegetative drought indices such as the vegetation condition index (VCI) and temperature condition index (TCI) have been shown
useful for drought detection (Kogan 1995; Bhuiyan et al. 2006). Some recent drought
monitoring models were developed with the aid of satellite remote sensing imageries in relation to those VIs and LST using a combination of LST from thermal band
data versus VIs from visible and near-infrared (NIR) data (Bayarjargal et al. 2006;
Ghulam et al. 2007). To gain more insight into the relationship between vegetation
vigor and moisture availability, several more remote sensing–based drought indices
were developed (Ji and Peters 2003). Some early drought indices such as the Keetch–
Byram drought index are also starting to include the El Niño/southern oscillation
information to address global climate change impacts (Brolley et al. 2007).
In most urban drought events, drought might simultaneously turn pastures brown,
threaten shrubs and trees, and result in low vegetation cover and high LST. In the
last two decades, to reflect the drought impacts with multiple aspects, many satellitederived indices have been specifically developed to function as drought indicators
of plant water content, water stress, VIs, LST, soil moisture, and ET (Brolley et al.
2007; Kimura 2007; Ghulam et al. 2007). However, a composite drought indicator
7.5 Conclusions .................................................................................................. 160
Acknowledgment ................................................................................................... 161
References ............................................................................................................. 161
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
7.1 INTRODUCTION
7.1.1 Motivation of Study
The occurrence of historical droughts led to studies on their impact and assessment
methods. Droughts differ from most natural hazards in several important ways:
(1) a slow-onset, creeping phenomenon occurs; (2) duration varies from event to
event; (3) there is no universal definition; (4) no single drought index can identify
precisely the onset and severity of the event; (5) spatial extent can be much greater
than that of other natural hazards, making assessment difficult; (6) the core area or
epicenter can change over time, reinforcing the need for continuous monitoring; and
(7) impacts are generally difficult to quantify with cumulative effects. In particular,
monitoring these phenomena in a fast-growing urban region where the multitemporal changes of land use and land cover (LULC) can affect holistic drought assessment is a challenge (Tadesse et al. 2005).
The early quantitative indices based on climatic and meteorological observations
include the Palmer drought severity index (PDSI; Palmer 1965), rainfall anomaly
index (van Rooy 1965), and Palmer crop moisture index (Palmer 1968). Current
drought measurement relies on biophysical parameters such as vegetation indices
(VIs), land surface temperature (LST), soil moisture, albedo, and evapotranspiration (ET). Vegetation health is an essential indicator, and vegetation cover was once
considered a good surrogate index for drought monitoring (Tadesse et al. 2005). The
most frequently used VIs are the normalized difference vegetation index (NDVI;
Rouse et al. 1974), the soil-adjusted vegetation index (SAVI; Huete 1988), the modified SAVI (MSAVI; Qi et al. 1994), and the enhanced vegetation index (EVI; Huete et
al. 1999). With the aid of NDVI, other vegetative drought indices such as the vegetation condition index (VCI) and temperature condition index (TCI) have been shown
useful for drought detection (Kogan 1995; Bhuiyan et al. 2006). Some recent drought
monitoring models were developed with the aid of satellite remote sensing imageries in relation to those VIs and LST using a combination of LST from thermal band
data versus VIs from visible and near-infrared (NIR) data (Bayarjargal et al. 2006;
Ghulam et al. 2007). To gain more insight into the relationship between vegetation
vigor and moisture availability, several more remote sensing–based drought indices
were developed (Ji and Peters 2003). Some early drought indices such as the Keetch–
Byram drought index are also starting to include the El Niño/southern oscillation
information to address global climate change impacts (Brolley et al. 2007).
In most urban drought events, drought might simultaneously turn pastures brown,
threaten shrubs and trees, and result in low vegetation cover and high LST. In the
last two decades, to reflect the drought impacts with multiple aspects, many satellitederived indices have been specifically developed to function as drought indicators
of plant water content, water stress, VIs, LST, soil moisture, and ET (Brolley et al.
2007; Kimura 2007; Ghulam et al. 2007). However, a composite drought indicator
7.5 Conclusions .................................................................................................. 160
Acknowledgment ................................................................................................... 161
References ............................................................................................................. 161