Spatial and Temporal Variability Analysis in Rainfall Using Standardized Precipitation Index for the Fuhe Basin, China - Intelligent Computing for Sustainable Energy and Environment - page 453

value by using an inverse normal (Gaussian) function. For further details on the
specifics of the procedure the reader can refer to the reference [11].
Associated with the SPI index, positive values indicate greater than the mean
for wet conditions and negative values indicate less than the mean for drought.
Normally, part of the SPI range is arbitrarily split into 7 level conditions [11].
The specific partitions are shown in Table 1, where the drought and the flood
categories correspond to specific classification ranges of

> 1. Following the
SPI calculation the process is smoothened by using a moving average window.
In other words, for the SPI-10 days calculation, the 10th day SPI is based on
the mean precipitation from the first day to the 10-th day in the same period,
the 11th day SPI is based on the mean precipitation from the second day to the
11-th day in the same period, etc. In the present study, the value of the SPI-10
days is calculated as shown in Fig. 2.

| SPI |

Fig. 1. The schematic diagram of the spatial hydrological stations distribution in the
Fuhe basin

4 Results and Discussions

4.1 Years Trends and Analysis
Fig. 3 shows the annual average rainfall distribution from the year 1970 to the
year 2008 in 17 hydrometric stations over the Fu river basin. It can be seen that