Information Technology Reference
In-Depth Information
Tabl e 1.
Drought category based on the SPI index
SPI Value Category
>
2.0 Extremely wet
1.5-1.99 Very wet
1.0-1.49 Moderately wet
-0.99-0.99 Near normal
-1.0-1.49 Moderately dry
-1.5-1.99
Very dry
Fig. 2.
The standard precipitation indices (SPIs) series based on the average rainfall
over the Fu river basin from Jan 1970. to Dec 2008 for a total of 468 months and 14244
days.
the inter-annual variability of the annual rainfall is significant. Specifically, the
years 1980, 1997, 1998, 2002 are wet years and have significant rainfalls, and
there are local floods or large area floods in these years. Whereas in the years
1971, 1973, 1978, 2003, 2007, the Fu river basin was subject to severe droughts.
By using a simple linear regression in time and the water quantity, one can
obtain the expression
y
=9
.
9589
x
18130 for the precipitation and years, where
x
denotes the year ranging from 1970 to 2008 and
y
denotes the annual rainfall
as shown by the red line in Fig. 3. Given that the slope is positive this indicates
that the rainfall is increasing year by year in the Fuhe basin, and the overall
trend of the rainfall is gradually increasing.
According to the classification of the SPI index (Table 1), using the 10-
day smoothing moving average, the rainfall frequency within the normal range
(
−
1
<
=
SPI <
= 1), drought (
SPI < −
1), wet (
SPI >
1) from the years
1970 to 2008 are calculated and shown in Fig. 4. It can be seen that the wet
label has been the dominant position since 1978, and specifically in 1986-1989,
1990-1992, 1997-2002, 2005 floods have occurred. The years 1998 and 2002 were
the extremely wet periods with significant floods. Whereas the years 1970-1979
were the relatively dry concentrated periods, especially in 1971 and 1973 where
a sustained and extremely drought period has occurred. Following these years
in 1986, 1996, 2003 and 2008, droughts have occurred again.
In addition, a linear regression is also used to analyze the normal year fre-
quency trends. We have obtained the following linear regression equation as
frequency
=
−
0
.
00046796
year
+1
.
6699 with a norm of the residuals being
0.33434, where year is to be selected from 1970 to 2008. It can be seen from Fig.
4 that the linear regression normal level curve presents a slight downward trend,
which indicates that the normal rainfall turn lower with time and the drought
and flood in recent years have became more intense and the fluctuation of the
weather have become larger than prior years.
−
Search WWH ::
Custom Search