Journal of Hydrologic Engineering

Bangladesh suffered damage on account of the most catastrophic floods of 1987, 1988, and 1998, resulting in untold suffering of the people. This paper demonstrates the technique to develop a flood hazard map and a land development priority map for countermeasures against flood damage. To create the final products, the flood hazard map and land development priority map, National Oceanic and Atmospheric Administration (NOAA) advanced very high resolution radiometer (AVHRR) data for the flood events of 1988, 1995, and 1998 were incorporated with geographic information system data. Flood-affected frequency and flood depth categories were estimated using NOAA AVHRR images. Special attention was paid to population density for the construction of the land development priority map, because highly dense populated areas represent the highly important urban and industrial areas of Bangladesh. The land development priority map offers a new opportunity for flood risk management, planning, design, and operation of flood control measures in Bangladesh, and should be useful in assigning priorities for the development of at-risk areas.

Depth-area relationships, such as those published by the National Weather Service in TP 40 and the NOAA Atlas 2, enable conversion of point rainfall depths to areal average depths for the same storm duration and recurrence interval. This problem of conversion is most germane to hydrologic analyses for moderate to large drainage basins, where point rainfall depths are not representative of the spatial distribution of a storm event. Historically, depth-area relationships have been developed on the basis of data from dense networks of recording gauges. However, with the ongoing accumulation of radar-rainfall records, radar-rainfall data represent an alternative to gauging data. This paper summarizes what is believed to be the first study made under the auspices of the National Weather Service (NWS) for evaluation of the potential of NEXRAD radar-rainfall data for development of geographically fixed depth-area relationships. Objectives were to evaluate the use of radar-rainfall data for development of depth-area relationships and to identify potential obstacles that might hinder use of such data. Data analyzed for this study are those recorded for the Arkansas-Red Basin River Forecast Center (ABRFC), and span the period of time from May 1993, to September 2000. Conclusions of this study are that data heterogeneities and shortness of data records are major factors limiting development of depth-area relationships on the basis of radar-rainfall data. Possible biases in radar estimates of extreme rainfall are also of concern. Depth-area curves developed for the ABRFC, presented herein, are reasonably consistent with those presented in NWS publications but should only be considered as preliminary.

Multiyear (1982–1990) monthly areal evapotranspiration (AET) was modeled with the Morton approach at the Solar and Meteorological Surface Observation Network stations within the conterminous United States. The AET values were correlated with satellite-derived, monthly maximum-value-composited Normalized Difference Vegetation Indices (NDVI) at half-degree resolution over the growing season (April–October). Generally, the strongest monthly correlation was obtained when the NDVI values were related to the AET estimates of the previous month. Geographically, both the monthly and growing-season averaged NDVI-AET relationships were best over the prairie (with an r=0.66±0.21 and 0.55±0.22, and a RMSE=26.75±12.62 and 6.24±1.67 $\mathrm{mm}{\mathrm{month}}^{-1}\text{,}$ respectively) and worst along the coastline and in the most humid, southeast region of the conterminous United States, where the Morton approach may function improperly.

The Richard’s equation was used to numerically model 1D flow within unsaturated topsoil and solved using a two-step Crank-Nicolson scheme. Under constant water supply, the Darcy velocity obtained at the upper boundary, the ground surface, is regarded as the correct infiltration rate. Based on this, three infiltration-capacity formulas by Horton, Philip, and Green-Ampt were evaluated for three types of soil. Results demonstrated that all three formulas provide similar fits to the numerical results, but the Horton formula differs most as compared to the other two formulas in terms of infiltration rate. The cumulative infiltration curves were not distinguishable from each other. By perturbing the rainfall intensity, duration, and initial soil moisture content, regression results demonstrated that only the saturated hydraulic conductivity $K_s$ can be used as a good estimator for parameter k (or $f_c)$ in all formulas. The other two parameters, however, will deviate from their original values. Only the late-time infiltration parameter (k or $f_c)$ can be predicted a priori using measured $k_s$ values.

In an urban fractured-rock aquifer in the Mt. Eden area of Auckland, New Zealand, disposal of storm water is via “soakholes” drilled directly into the top of the fractured basalt rock. The dynamic response of the groundwater level due to the storm water infiltration shows characteristics of a strongly time-varying system. A dynamic fuzzy modeling approach, which is based on multiple local models that are weighted using fuzzy membership functions, has been developed to identify and predict groundwater level fluctuations caused by storm water infiltration. The dynamic fuzzy model is initialized by the fuzzy clustering algorithm and optimized by the gradient-descent algorithm in order to effectively derive the multiple local models—each of which is associated with a locally valid model that represents the groundwater level state as a response to different intensities of rainfall events. The results have shown that even if the number of fuzzy local models derived is small, the fuzzy modeling approach developed provides good prediction results despite the highly time-varying nature of this urban fractured-rock aquifer system. Further, it allows interpretable representations of the dynamic behavior of the groundwater system due to storm water infiltration.

Feed-forward multilayer neural networks are widely used as predictors in several fields of applications. The purpose of this study is to investigate the performance of neural networks as potential models capable of forecasting daily streamflows. Once an appropriate network has been identified, a comparison approach is used to evaluate it against a conceptual model presently in use by the Alcan Company. The Mistassibi River, located in northeastern Quebec, serves as the case study, and results based on mean square errors and Nash coefficients show that artificial neural networks outperform the deterministic model PREVIS for up to 5-day-ahead forecasts. Moreover, the results obtained with the neural network are also superior to the ones obtained with a classic autoregressive model coupled with a Kalman filter.