Agricultural Drought in North-Central Mexico - Monitoring and Predicting Agricultural Drought: A Global Study - page 137

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

three months (SPI-3) and 12 months (SPI-12). The SPI-3 is intended to help
dry-land farming where production depends on seasonal rainfall, whereas
SPI-12 can be used by federal and state agencies that manage state water
resources. Finally, the SPI values are loaded on to a Geographic Informa-
tion System (GIS) for showing spatial distribution of droughts and their
severity levels ( http://www.sequia.edu.mx/). Further GIS analyses are per-
formed to estimate damages to agricultural areas, other natural resources,
and society.
The advantage of using a large scale (state size) is that regional resources
affected by drought can be better evaluated. A main disadvantage, how-
ever, is that the geographic limits of major drought events affecting large
areas might not be included in the drought monitor. The current moni-
toring activities of the CEISS are confined to the state of Chihuahua. The
center has plans to extend its monitoring activities to the northern part of
Mexico. The SMN is also planning drought monitoring at the national level
by applying the SPI methodology in addition to the precipitation deviation
method described earlier.

[137

Line
——
0.0
——
Norm
PgEn

Pr edicting Crop Yields using Standardized Precipitation Index

The SPI is a qualitative indicator of drought. To use it for quantitative
assessment of agricultural drought, the quantitative relationship between
SPI and yields of maize and bean crops was examined for the north-central
states of Mexico. Table 10.2 shows the basic variation in main yields of
maize and bean during the last two decades.
The rainfall data were collected from two databases (ERIC I, 1995;
ERIC II, 1999) of the Instituto Mexicano de Tecnología del Agua (IMTA;
i.e., Mexican Institute of Water Technology) to compute SPI-3. In addi-
tion, maize and bean yield data were collected from the Sistema Integral
de Información Agroalimentaria y Pesquera (SIIAP; i.e., Agricultural and
Food Information Integral System) created by the federal agency, Secre-
taría de Agricultura, Ganadería, Desarrollo Rural, Pesca y Alimentación
(i.e., Agency of Agriculture, Cattle, Rural Development, Fishing and Food;
SIIAP, 2003) and INEGI (1993, 1995, 1998, 1999a, 1999b, 2000b, 2000c).
Results of these correlations are presented in figure 10.4. The results
show positive but low correlation coefficients (from 0.26 to 0.54) for bean
and maize yields with SPI-3. In addition to SPI-3, the correlation analysis
was attempted using SPI-4, SPI-5, and SPI-6, but the results showed even
lower correlations. Aguascalientes, Durango, and Zacatecas are the states
that indicate moderate correlations for both crops. Correlations for Chi-
huahua and Coahuila are the lowest (figure 10.4b,c). SPI-3 seasonal values
for the state of Durango show the best correlation with the economic values
of bean and maize crops.
A possible reason for low correlations is that the SPI values used in the
correlation analysis were averaged for an entire state. One way to improve
these results is by splitting the state into regional agricultural areas and

[137