Drought Monitoring Methods

The following models have been developed for monitoring droughts in China, and an operational drought monitoring system is also in place.

Water Balance Model

The water budget for the root zone of a soil column can be written as (Yang and Tian, 1991):

d t where W is the water content of the root zone (cm), P is precipitation (cm), I is irrigation (cm), E is evapotranspiration (cm), R is runoff (cm), D is the discharge from groundwater (cm), and G is recharge to the groundwater (cm). Ignoring, for the moment, recharge and discharge and incorporating irrigation into rainfall, the above equation can be integrated over time t (days or weeks) to derive a simple equation for the incremental change in the soil water storage:

The above parameters are difficult to estimate. Typically, soil moisture is estimated from the amount of rainfall and crop water demand, depending on air temperature, solar radiation, humidity, wind speed, potential evapotranspiration (Ep), and land cover conditions (Monteith and Unsworth, 1990). Simple water balancing can be performed if initial storage (W0) is known and if operating characteristics for the area that define relationships between moisture availability [ma(WB)] and runoff fraction (rr) are also known. For example:

where Wf is the amount of water available in a soil profile at field capacity; Ww is the amount of water in the profile at which evapotranspiration stops and plant growth ceases; and Wmax is extractable water that usually equals Wf — Ww. It has been sufficient (but not necessary) to use the simplest linear forms of these equations. Water availability is defined as (Jupp et al., 1997):

ma(WB) = max {0, min [1, j3(Wt — Ww)/Wmax]| [28.5]

Yang and Tian (1991) developed a drought index (D) following a water-balancing approach:

where Wr is the amount of water required by the vegetation for its normal growth and respiration. Derivation of D, which is closely related to ma, needs adequate meteorological data, spatial information about the hydro-logical properties of soils, and the nature of the land cover. If ma or D can be determined during various periods by remote sensing techniques (chapters 7 and 8), it is possible to validate water-balancing models.

Remote Sensing

Remote sensing provides physical measurements of components of the energy balance—daily input of solar radiation and its various forms, such as convection, evaporation, and storage in the earth or oceans. In particular, the Advanced Very High Resolution Radiometer (AVHRR) data provide information on shortwave absorption and the condition of the cover and surface temperature.

If Rn denotes net wavelength radiation (W/m2) and G denotes the heat flux into the soil (W/m2), the net energy available at the earth's surface will be Rn — G, which can be partitioned as:

where E is the evapotranspiration flux (m/sec) of water vapor; k is latent heat of vaporization of water (J/m3), H is sensible heat flux (W/m2) or the energy involved in the movement of the air and its transfer to other objects (such as trees, grass, etc.), and kE is the energy required to transform water from liquid into vapor and can be computed as follows (Monteith and Unsworth, 1990):

The above components may be computed in various ways using resistance models (Monteith and Unsworth, 1990) of differing levels of complexity introduced by land surface structure and cover. In each case, the sensible heat flux is driven by the differences between the distribution of temperatures among the surface components and the air temperature. For simple (one-layer) or more complex models (Jupp, 1990), the remotely sensed aggregated surface temperature may be sufficient to estimate evapotranspiration (ET).

Not all energy balance models contain explicit relationships between soil moisture and surface temperature. Many of those that do are complex and may contain many parameters that are difficult to measure. Most, however, can be solved for the case when soil moisture is at field capacity and is not a limiting factor. The effective ET (XEp, where Ep is potential ET) corresponds to a situation where soil water and vegetation condition do not limit ET while the atmospheric demand and land cover types and structures are the same. The moisture availability at other times (when soil water is below field capacity) is then defined as ma(EB) = XE/XEp [28.9]

Crop Water Stress Index Using an integrated version of the moisture availability defined in equation 28.9, Jackson et al. (1983) defined the crop water stress index (CWSI) as

Epd where the subscript d denotes that the quantity is integrated over a day. Remote sensing data can be used for computing water balancing by providing estimates for ET (Jupp et al., 1997; Tian et al., 1997). Based on the range of CWSI and W/Wf, the surface conditions can be classified as heavy drought, dry, light dry, normal, and moist conditions (table 28.2).

Normalized Difference Temperature Index Jackson et al. (1983) found that an approximation of ma equals a normalized difference temperature index (NDTI):

where T^ is theoretical surface temperature if no water were available, and T0 is theoretical temperature if maximum amount of water were available (i.e., the temperature corresponding to Ep).

The NDTI was successfully used to map surface temperatures for drought monitoring in the North China Plain (Tian et al., 1997). To estimate evapotranspiration, a one-layer model was used for a vegetation-covered surface (Jupp et al., 1997), and a two-layer model was used for partial vegetation-covered surface (Jupp et al., 1997). A soil-thermal inertia model was used for low vegetation-covered or bare surface for estimating soil moisture (Yu and Tian, 1997).

AGRICULTURAL DROUGHT IN CHINA 361 Table 28.2 The categories of drought

Grade

Crop water satisfaction index

Heavy drought Dry

Light dry

Normal

Moist

>0.913 0.765-0.913 0.617-0.764 0.322-0.616 <0.321

Normalized Difference Water Index

The spectral signature of vegetation canopy at 1.55-1.75 ^m waveband is sensitive to water content in the canopy. Remotely sensed data at the shortwave infrared can be used to monitor water content of vegetation canopy. A normalized difference water index (NDWI) was developed to monitor drought during the cropping season (Tian et al., 2002):

where PSW is the reflectance at shortwave infrared (e.g., 1.55-1.75 ^.m), and PR is the reflectance at red band (0.61-0.68 ^.m). The NDWI is correlated with soil moisture (Tian et al., 2002) at root zone and therefore can be used to monitor drought conditions in crops.

Anomalous NDVI Model

The anomalous NDVI (Chen, et al., 1994) can also be used to monitor drought conditions:

where NDVI is the normalized difference vegetation index based on AVHHR data, and NDVI is an NDVI averaged for multiple (more than 10) years.

Combination Model

The combination model (CM) is defined as

ANDVI =

NDVI - NDVI

NDVI

where TS is surface temperature that can be derived from channel 4 and channel 5 data of the AVHHR thermal data (Xin et al., 2003).

Figure 28.3 Operational system for drought early warning in the North China Plain.

Was this article helpful?

0 0

Post a comment