Conleth Cunnane

Methods and merits of regional flood frequency analysis

Abstract Regional flood frequency analysis (RFFA) is discussed in the context of regional smoothing of hydrological data, the concept of regional homogeneity is explained and the effect of inter-site dependence of flood data is discussed. A description of the principal features of twelve RFFA schemes is given, not all to the same degree of algebraic detail. Common elements of these methods are identified and the case in favour of RFFA is presented. The choice of RFFA scheme and the accompanying choice of distribution is discussed. The properties of RFFA methods and intercomparison of them as flood quantile estimators are referred to but the details reported by others are not reproduced here.

A criterion of efficiency for rainfall-runoff models

Abstract Two criteria are proposed for expressing the efficiency of conceptual rainfall—discharge models. The first assesses the model as a means of converting the input factors into discharge by comparison with a forecast based only on the seasonal regime of the river. The second expresses the efficiency of the model under the assumption that it is to be used with an updating procedure to provide a forecast of discharge over a prescribed lead time.

Review of Statistical Models for Flood Frequency Estimation

The relation between flood frequency estimates and the economic decision-making process is briefly discussed and selected developments in frequency analysis are traced in tabular form. The main types of frequency analysis procedures and models are surveyed including at-site, at-site/regional and regional only cases. The at-site/regional cases include Index Flood, Bayesian and TCEV methods. Criteria for selecting a frequency analysis procedure are discussed under two headings, descriptive ability and predictive ability. The former relates to a model’s ability to preserve statistics of observed flood series while the latter relates to quantile estimating ability in a robust manner. The relative merits of different types of flood quantile estimators are discussed, beginning with at-site estimators and the search for a robust at-site estimator. This is followed by the principal results available about at-site/regional quantile estimators. Further topics considered are regional homogeneity of flood statistical behaviour, the effect of spatial and temporal interdependence of flood magnitudes, and the detection and use of historical floods.

Modelling point rainfall occurrences with the modified Bartlett-Lewis rectangular pulses model

Abstract A six parameter stochastic point process model, known as the modified Bartlett-Lewis Rectangular Pulses Model, is applied to fairly long hourly rainfall data recorded at Valentia (relatively a wet location) and Shannon Airport (relatively a dry location), Ireland. Five different sets of statistics of the rainfall data of each month, assuming local stationarity within the month, are used to estimate the parameters and to simulate model output. The problems of parameter stability/sensitivity and identification are discussed and it has been shown that the sensitivity of the model parameters to the choice of six statistics can be avoided by estimating the six parameters by optimization from 16 statistics namely mean, variance, lag-1 autocorrelation corfficient and proportion dry of hourly, 6-hourly, 12-hourly, and 24-hourly rainfalls. Some useful properties of the rainfall depth process are analysed using the notion of event-based statistics. The conditional distributions of rainfall depth and maximum intensity, mean event profiles, and various other features of the rainfall depth process obtained from the model simulated samples compare favourably with the historical ones.

Evaluation of the usefulness of historical and palaeological floods in quantile estimation

Abstract The methods of incorporation of historical floods and palaeological information into flood frequency analysis, and the usefulness of doing so, have been evaluated by many hydrologists. These evaluations are not in complete agreement. The results of a Monte Carlo study are presented comparing different simulation procedures and assessing the value of historical floods for at-site flood frequency analysis on the assumption of a Gumbel (EVI) distribution. It is shown that historical floods and palaeological information provide a useful source of information additional to the recorded series, and have great value in flood frequency analysis when floods are drawn from the Gumbel distribution. Simulation procedures based on type II censoring result in the largest bias and root mean square error in quantile estimation. This may be due to their assumption of type II censoring in the production of their simulated samples, an assumption that has some limitations. In the present work it was found that the type I censored-data maximum likelihood estimator is a robust model for the Gumbel distribution and that the type II censored-data maximum likelihood estimator performs poorly when the data are in fact obtained by type I censoring.

Split sampling technique for selecting a flood frequency analysis procedure

Abstract The use of a split sampling technique for selecting a flood frequency analysis procedure is discussed in detail with the aid of simulation experiment results. Although this technique is unable to identify the most likely statistical distribution which generated given flood data, it is demonstrated that it provides useful results with which to compare the relative merits of different flood frequency estimation procedures.

Expected probabilities of exceedance for non-normal flood distributions

Abstract The concept and importance of expected probability of exceedance of flood quantiles is discussed. Theoretical derivation of expected probability of exceedance values for different sample lengths, has been so far limited to the normal distribution. A relatively simple numerical algorithm is introduced to calculate such values for non-normal flood distributions. The sensitivity of the expected probability to different combined flood frequency analysis procedures is demonstrated. The usefulness of the expected probability as an appropriate indicator to gauge relative merits of different flood frequency analysis procedures is also illustrated with a simple example.

A quasi physical snowmelt runoff modelling system for small catchments

Runoff forecasting in the case of seasonally snow covered small catchments with shallow snowpacks requires application of a quasi physical approach wherein the dominant snow accumulation and melting processes are accounted for by an intensive physically based modelling approach and transformation of the snowmelt and the rainfall to streamflow is accounted for by a conceptual modelling approach. In the case of shallow snowpacks both high and low water saturation can occur more frequently and therefore the physically based multilayer snowmelt model must account for capillary pressure gradients as well as gravity drainage. One such physically based snowmelt model entitled UCGVDSM which accounts for coupled transport of mass and energy into the snowpack, is first validated on point snowmelt data of the Kuhtai station located in Austria. UCGVDSM is then applied to the Ticha Orlice catchment (96.8 km 2 ) located in the Czech Republic. It is shown how the constraints of data availability for application of the physically based snowmelt model can be handled to reproduce accurately, the snow water equivalent (SWE). the snow depth (H) and the melt water flux (qmelt) The snowmelt rates thus obtained for the snowcover periods are then incorporated along with the rainfall and the evapotranspiration data into the Soil Moisture Accounting and Routing model (SMAR), a conceptual rainfall runoff model It is shown that incorporating a number of statistical modelling techniques into the SMAR model has no effect on the model performance while accounting for physical processes improves the model performance Finally, an updating component is incorporated into the SMAR model to allow its application in a forecasting mode.

Modelling coupled transport of mass and energy into the snowpack—model development, validation and sensitivity analysis

Abstract A numerical simulation model entitled UCGVDSM is proposed for modelling coupled transport of mass and energy into the snowpack. The model is tested on point snowmelt data for station Kuhtai located in the Austrian Alps. Model performance is .assessed on the basis of model reproduction of three state variables—snow water equivalent, melt water flux and depth of snowpack. The model also provides information on internal processes of the snowpack. The range of all the parameters used in the model is extensively examined through a sensitivity analysis. The search for optimum values of the parameters is combined with sensitivity analysis.