Change Point Detection (1909-2018)

Pettitt Test

For more information, see here.

  • Null Hypothesis: There are no abrupt changes in the mean of the data.
  • Alternative Hypothesis: There is one abrupt change in the mean of the data.

Results:

  • The Pettitt test yielded a p-value of 0.022.
  • At a significance level of 0.05, we reject the null hypothesis.
  • Therefore, there is evidence of an abrupt change in the mean in 1974.

MKS Test

For more information, see here.

  • Null Hypothesis: There are no trend changes in the data.
  • Alternative Hypothesis: There is at least one trend change in the data.

Results:

  • The MKS test yielded a p-value of 1.
  • At a significance level of 0.05, we fail to reject the null hypothesis.
  • Therefore, there is no evidence of a trend change.

Trend Detection (1909-2018)

White Test

For more information, see here.

  • Null Hypothesis: The data is homoskedastic.
  • Alternative Hypothesis: The data is heteroskedastic.

Results:

  • The White test yielded a p-value of 0.087.
  • At a significance level of 0.05, we fail to reject the null hypothesis.
  • Therefore, there is no evidence of heteroskedasticity in the data.

MW-MK Test

For more information, see here.

  • Null Hypothesis: There is no monotonic trend in the standard deviation of the data.
  • Alternative Hypothesis: There is a monotonic trend in the standard deviation of the data.

Results:

  • The MW-MK test yielded a p-value of 0.976.
  • At a significance level of 0.05, we fail to reject the null hypothesis.
  • Therefore, there is no evidence of a monotonic trend in the standard deviation.

Mann-Kendall Test

For more information, see here.

  • Null Hypothesis: There is no monotonic trend in the mean of the data.
  • Alternative Hypothesis: There is a monotonic trend in the mean of the data.

Results:

  • The Mann-Kendall test yielded a p-value of 0.007.
  • At a significance level of 0.05, we reject the null hypothesis.
  • Therefore, there is evidence of a monotonic trend in the mean.

Spearman Test

For more information, see here.

  • Null Hypothesis: There is no autocorrelation in the data.
  • Alternative Hypothesis: There is autocorrelation in the data.

Results:

  • The Spearman test yielded a least lag of 1.
  • Therefore, there is no evidence of autocorrelation in the data.

Sen’s Trend Estimator (Means)

For more information, see here.

Identified Trend: y = -48.82x + 232.035

Runs Test (Means)

For more information, see here.

  • Null Hypothesis: The input vector is random.
  • Alternative Hypothesis: The input vector is not random.

Results:

  • The Runs test yielded a p-value of 0.439.
  • At a significance level of 0.05, we fail to reject the null hypothesis.
  • Therefore, there is evidence that the residuals are distributed randomly.

Flood Frequency Analysis

Approach: NS-FFA

Split Point(s): None

Model Structures:

Subperiod Trend in Location Trend in Scale
1909-2018 TRUE FALSE

‘L-distance’ Distribution Selection (1909-2018)

Recommended Distribution: GNO

Distribution Metric
GUM 0.0040
NOR 0.1741
LNO 0.0560
GEV 0.0039
GLO 0.0426
GNO 0.0012
PE3 0.0149
LP3 0.0018
WEI 0.0334

‘MLE’ Parameter Estimation (1909-2018)

Distribution: GNO

Estimated Parameters: 224.0497, -35.5154, 54.6325, -0.3689

‘RFPL’ Uncertainty Quantification (1909-2018)

Slice: 1925

Effective Return Period CI (Lower) Return Level Estimate CI (Upper)
2 199.52 215.17 231.79
5 250.73 269.09 290.17
10 282.53 304.68 333.49
20 310.86 338.76 379.56
50 344.99 383.00 445.12
100 369.02 416.42 498.30

Slice: 1975

Effective Return Period CI (Lower) Return Level Estimate CI (Upper)
2 186.41 197.41 209.14
5 236.16 251.33 270.14
10 266.78 286.93 315.58
20 294.21 321.00 362.87
50 327.44 365.24 429.21
100 351.05 398.66 482.70

Slice: 2025

Effective Return Period CI (Lower) Return Level Estimate CI (Upper)
2 159.61 179.66 199.46
5 210.76 233.57 258.66
10 242.08 269.17 302.78
20 270.18 303.25 349.40
50 304.11 347.48 415.25
100 328.08 380.90 468.46

Model Assessment (1909-2018)

Statistic Value
Model Validation: Coefficient of Determination 0.9918262
Model Validation: RMSE 0.0961502
Model Validation: Bias -0.0000020
Akaike Information Criterion (RMSE) -247.2609347
Bayesian Information Criterion (RMSE) -236.4955432
Akaike Information Criterion (MLL) 1189.4657004
Bayesian Information Criterion (MLL) 1200.2310920