Results - Response of Flood Events to Land Use and Climate Change - page 202

Table 6.21 Statistics of homogeneity test (z-value) for selected variables

Runs test
Parameters Station z 95 % confidence interval (P value)

Precipitation Barcelonnette -2.063 .039

Discharge Barcelonnette -11.392 .000

Max-temperature Barcelonnette -16.014 .000

Min-temperature Barcelonnette -16.180 .000

Precipitation Jausiers -1.24 .216

Precipitation Uvernet -0.49 .620

Precipitation Condamine -0.39 .696

Precipitation Allos -1.66 .095

Precipitation SainPaul -1.78 .075

Median

6.12.2 Detection of Homogeneity

Statistical analysis was done on the available monthly data of precipitation,
maximum discharge (these data are maximum values of discharges which occurred
at the Barcelonnette station based on recorded hourly discharge), and maximum
temperature during the period from 1961 to 2009. Due to a lack of high temporal
resolution precipitation data, only the available monthly precipitation series was
used in this study. Z-score is known as a standardized value. To obtain z-scores for
a variable for each case, subtract the variable's mean value and divide it by the
standard deviation. Z-scores are useful for finding outliers and for comparing
values of variables that are measured on different scales. The Z value of the
homogeneity test was analyzed in the SPSS package. The z-score column is
compared to a standard normal table. Thus, at a 5 % significance level, a Z-score
with an absolute value greater than 1.96 indicates nonhomogeneous. The signifi-
cant values are presented in bold italic format. In the period between 1961 and
2009, three stations, Jausiers, Uvernet, and Condamine, represent nonsignificant Z-
values for precipitation. On the other hand, precipitation variables in Sain-Paul and
Allos-Nord represent significant Z-values. At the Barcelonnette station, the Z-
value is also significant with 95 % confidence in the lower and upper parts for
precipitation, discharge, and maximum/minimum temperature parameters. This
may imply the existence of abrupt climate change in these stations (Table 6.21 ).
At two other stations close to Barcelonnette, Allos-Nord and Saint Paul, there is
an abrupt point in precipitation where the descriptive statistics of the variables
(such as mean and variance) begin to change suddenly. To find the time of this
change, a sequential Mann-Kendall test was applied.