Monitoring Drought in the United States: Status and Trends - Monitoring and Predicting Agricultural Drought: A Global Study - page 123

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

and Atmospheric Administration (NOAA) and are frequently used in as-
sessments of agricultural conditions around the United States. The PDSI
has been used historically in policy decisions by the U.S. Department of
Agriculture (USDA) regarding requests for drought relief and by states as
triggers for response actions as part of state drought plans. However, the
PDSI has limitations that diminish its application, bringing into question
the practice of basing agricultural policy decisions solely on the PDSI. These
limitations have been well documented (Alley, 1984; Karl and Knight,
1985; Willeke et al., 1994; Kogan, 1995; McKee et al., 1995; Guttman,
1998). The most significant limitations of the PDSI related to monitoring
agricultural drought include: (1) an inherent time scale (about 8 months)
that makes it difficult to detect emerging drought conditions and shorter-
length drought periods; (2) the characteristic that all precipitation is treated
as rain, which does not account for snowfall, snow cover, and frozen
ground, thus questioning the reliability of the real-time winter PDSI val-
ues and preplant soil moisture estimates (one of the PDSI inputs); (3) the
characteristic that the natural lag between precipitation and runoff is not
considered and that no runoff occurs until the water capacities of the sur-
face and subsurface soil layers are full, which leads to an underestimation
of runoff; and (4) wide variations in the “extreme” and “severe” classifi-
cations of the PDSI values, depending on location in the country.
If a drought index is going to be spatially comparable and useful for
agricultural policy decisions, extreme and severe classifications must occur
consistently and with low frequency (Guttman et al., 1992). An additional
concern is that the PDSI does not do well in the mountainous western
United States, especially since a majority of that region's precipitation falls
during the winter as snowfall. The PDSI can also retain values reflecting
drought well after a climatological recovery from drought has occurred.
All of these limitations reveal the importance of caution when using the
PDSI for monitoring agricultural conditions and making related policy
decisions.

[123

Line
——
0.0
——
Long
PgEn

[123

St andardized Precipitation Index

The SPI was developed to address some of the problems inherent in the
PDSI. McKee and his colleagues at Colorado State University (McKee et
al., 1993, 1995) designed the SPI to be a relatively simple year-round in-
dex for monitoring drought and water supply conditions in Colorado. The
SPI supplemented information provided by the PDSI and the surface water
supply index (SWSI) developed by Shafer and Dezman (1982). The SPI is
based on precipitation alone, whereas the SWSI incorporates snowpack,
streamflow, precipitation, and reservoir storage. Calculation of the SPI for
the specified time period for any location requires long-term monthly pre-
cipitation data for at least 30 years (i.e., the longer the data set, the more
reliable the SPI values). The probability distribution function is determined
from the long-term record by fitting a function to the data. The cumulative
distribution is then transformed using equal probability to a normal distri-