120
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
conditions and plant species. In general, ET rates near seashores would be lower than
ET rates inland under the same weather and plant conditions, which are the cases in
many coastal cities.
Understanding ET allows efficient water management planning for land use applications and agricultural activities. Various factors affecting ET include temperature,
relative humidity, wind speed, and solar energy; for example, a sunny day with strong
wind and dry air would naturally increase the ET rates. Soil types and conditions
also affect ET rates. Clays can likely hold water better than sand or silts because of a
strong chemical interaction between water and clays at the molecular level; therefore,
ET rates in clays are normally lower than in sand at the same weather conditions.
However, long-term ET rates in clay are normally higher, because sand will lose
water much quicker and dry out in a short time period, assuming a limited amount of
available water, whereas clay would slowly lose water over a longer time.
ET can be estimated by using either remote the sensing technology or mathematical models. Large-scale ET estimation is critical to numerous practices from
regional water resources management to local irrigation scheduling (Bastiaanssen
et al. 1998a,b; Kite and Droogers 2000; Schuurmans et al. 2003). The National
Oceanic and Atmospheric Administration (NOAA) GOES (Jacobs et al. 2008),
USGS Landsat (Bastiaanssen et al. 1998a,b), and the NASA MODIS (Nagler et al.
2005a,b) satellites all provide estimations of ET.
In this study, daily average ET data derived from GOES data and hydrologic
models can be retrieved from the USGS Web site directly (i.e., USGS spatiotemporal GOES-based data; http://hdwp.er.usgs.gov/et.asp). USGS produced retrospective
potential evapotranspiration (PET) and reference evapotranspiration (RET) estimates throughout Florida at a 2-km and daily resolution using a combination of
satellite (NOAA GOES) and land-based (weather stations) methods to compute ET.
The overall effort may provide gridded estimates of solar radiation, net radiation,
PET, RET, and actual ET at a 2 km × 2 km grid scale and a daily time scale from
2002 to 2008 for the entire state of Florida. The satellite-derived solar insolation
data set required calibration to correct for biases embedded in temporal-, seasonal-,
and cloudiness-related models (Jacobs et al. 2008). This was achieved through a
comparison with available ground-based pyranometer measurements (Jacobs et al.
2008). Upon calibration, the quality of the solar insolation product was improved
(Jacobs et al. 2008). Because RET is used mainly for agricultural use, PET data were
downloaded for the first day of each month during the study period to retrieve the ET
monthly maps. To harmonize the overall consistency in terms of spatial resolution,
ET data were finally resampled at a 1-km scale to be comparable with soil moisture
and EVI data sets.
6.3 RESULTS AND DISCUSSION
6.3.1 geneRation of Soil MoiStuRe MaPS
After constructing the soil moisture estimation algorithm, 16-day EVI (Figure 6.2)
and daily LST/emissivity L3 Global 1-km MODIS satellite images (MOD13A2) for
the first day of each month during the study period were processed and input into the
Multiscale Hydrologic Remote Sensing: Perspectives and Applications
conditions and plant species. In general, ET rates near seashores would be lower than
ET rates inland under the same weather and plant conditions, which are the cases in
many coastal cities.
Understanding ET allows efficient water management planning for land use applications and agricultural activities. Various factors affecting ET include temperature,
relative humidity, wind speed, and solar energy; for example, a sunny day with strong
wind and dry air would naturally increase the ET rates. Soil types and conditions
also affect ET rates. Clays can likely hold water better than sand or silts because of a
strong chemical interaction between water and clays at the molecular level; therefore,
ET rates in clays are normally lower than in sand at the same weather conditions.
However, long-term ET rates in clay are normally higher, because sand will lose
water much quicker and dry out in a short time period, assuming a limited amount of
available water, whereas clay would slowly lose water over a longer time.
ET can be estimated by using either remote the sensing technology or mathematical models. Large-scale ET estimation is critical to numerous practices from
regional water resources management to local irrigation scheduling (Bastiaanssen
et al. 1998a,b; Kite and Droogers 2000; Schuurmans et al. 2003). The National
Oceanic and Atmospheric Administration (NOAA) GOES (Jacobs et al. 2008),
USGS Landsat (Bastiaanssen et al. 1998a,b), and the NASA MODIS (Nagler et al.
2005a,b) satellites all provide estimations of ET.
In this study, daily average ET data derived from GOES data and hydrologic
models can be retrieved from the USGS Web site directly (i.e., USGS spatiotemporal GOES-based data; http://hdwp.er.usgs.gov/et.asp). USGS produced retrospective
potential evapotranspiration (PET) and reference evapotranspiration (RET) estimates throughout Florida at a 2-km and daily resolution using a combination of
satellite (NOAA GOES) and land-based (weather stations) methods to compute ET.
The overall effort may provide gridded estimates of solar radiation, net radiation,
PET, RET, and actual ET at a 2 km × 2 km grid scale and a daily time scale from
2002 to 2008 for the entire state of Florida. The satellite-derived solar insolation
data set required calibration to correct for biases embedded in temporal-, seasonal-,
and cloudiness-related models (Jacobs et al. 2008). This was achieved through a
comparison with available ground-based pyranometer measurements (Jacobs et al.
2008). Upon calibration, the quality of the solar insolation product was improved
(Jacobs et al. 2008). Because RET is used mainly for agricultural use, PET data were
downloaded for the first day of each month during the study period to retrieve the ET
monthly maps. To harmonize the overall consistency in terms of spatial resolution,
ET data were finally resampled at a 1-km scale to be comparable with soil moisture
and EVI data sets.
6.3 RESULTS AND DISCUSSION
6.3.1 geneRation of Soil MoiStuRe MaPS
After constructing the soil moisture estimation algorithm, 16-day EVI (Figure 6.2)
and daily LST/emissivity L3 Global 1-km MODIS satellite images (MOD13A2) for
the first day of each month during the study period were processed and input into the