Current Status of Time Series Analysis in Hydrological Sciences - Hydrologic Time Series Analysis: Theory and Practice - page 121

Zhang et al. (2010b) employed repeated monotonic trend tests with varying
beginning and ending time for detecting changes in streamflow in tributaries
within the Susquehanna River Basin, USA. The method was employed to
analyze long-term streamflow trends and detect change for annual minimum,
median, and maximum daily streamflow for eight unregulated watersheds
within the basin. Monthly baseflow and storm runoff were investigated. The
results showed a considerable increase in annual minimum flow for most of
the examined watersheds and a noticeable increase in annual median flow for
about half of the examined watersheds. Both these streamflow increases were
abrupt, with only a few years of transition centered around 1970. The abrupt
change in annual minimum and median flows appeared to occur in the summer
and fall seasons. The abrupt change in annual minimum and median flows
was considered as a result of increased flows in the summer and fall seasons.
The results also indicated that there is no long-term significant increasing or
decreasing change in annual maximum flow in the examined watersheds.
Zhao et al. (2010) investigated seasonal and annual trends of streamflow
and the correlations between streamflow and climatic variables in five sub-
basins in Poyang Lake basin in the southeast China over 50-year period. The
Theil-Sen approach and the non-parametric Mann-Kendall test were applied
to identify the trends in the annual and seasonal streamflow, precipitation and
evapotranspiration series. It was found that annual and seasonal streamflow of
all the stations had increasing trends except Lijiadu station in wet season.
Only 37.5% hydro-stations in annual streamflow increased significantly, while
most stations increased at 95% significance level in dry season. Trends in
annual and seasonal precipitation during the whole period were generally not
as significant as those in evapotranspiration. The correlations between
streamflow and climate variables (precipitation and evapotranspiration) were
detected by the Pearson's test. The results showed that streamflow in the
Poyang Lake basin are more sensitive to changes in precipitation than potential
evapotranspiration.
6.3.2 Surface Water Quality
Hirsch et al. (1982) presented techniques for the exploratory analysis of monthly
water quality data for monotonic trends. The first procedure is a nonparametric
test for trend detection, which is applicable to the datasets having seasonality.
The second procedure, the seasonal Kendall slope estimator, is an estimator of
trend's magnitude. The third procedure provides a means for testing the
temporal change in the relationship between constituent concentration and
streamflow.
El-Shaarawi et al. (1983) examined the temporal changes in some water
quality parameters (i.e., pH, alkalinity, total phosphorous, and nitrate
concentrations) using the 1975-1980 data of the Niagara-on-the-lake in Ontario.
The moving averages, Spearman's rank correlation coefficients and regression
methods, which model the seasonal cycle, were used in this study. It was