Geoscience Reference
In-Depth Information
Fig. 7.4. Autocorrelograms for the (a) annual and (b-g) maximum
rainfall series of Kharagpur.
subsequent stochastic analysis. On the other hand, it is obvious from Figs
7.4(b-g ) that the autocorrelograms for the one-day, consecutive 2-, 3-, 4-, 5-
and 6-day maximum rainfall time series are non-persistent, i.e., they are
independent of each other.
7.10 Conclusions
A critical assessment of various time series tests has been carried out to
demonstrate the efficacy of these tests for analyzing hydrologic time series
through a case study on the annual and salient consecutive days' maximum
rainfall series of Kharagpur, India. The time series plots, box plots and normal
probability plots of the one annual and six maximum rainfall time series
revealed no trends. The box plots indicated no severe outliers in both types of
rainfall series (i.e., annual rainfall and consecutive days maximum rainfalls),
except for one or two mild outliers in the 1-day and 2-day maximum rainfall
time series but they were also not found significant. The normal probability
plots indicated normality in the annual rainfall series as well as in the
consecutive 3-, 4-, 5- and 6-day maximum rainfall series, with slight non-
normality in the one-day and consecutive 2-day maximum rainfall series.
However, after removing two-three mild outliers, the D'Agostino-Pearson
Omnibus test suggested normality in all the seven rainfall series. Therefore,
the annual and maximum rainfall time series under investigation are considered
to be normally distributed.
Analysis of the results of three homogeneity tests and four multiple
comparisons tests indicated that the annual and maximum rainfall time series
Search WWH ::




Custom Search