An evaluation of a traditional and a neural net modelling approach to flood forecasting for an upland catchment

This study evaluates two (of the many) modelling approaches to flood forecasting for an upland catchment (the River South Tyne at Haydon Bridge, England). The first modelling approach utilizes ‘traditional’ hydrological models. It consists of a rainfall–runoff model (the probability distributed model, or PDM) for flow simulation in the upper catchment. Those flows are then routed to the lower catchment using two kinematic wave (KW) routing models. When run in forecast‐mode, the PDM and KW models utilize model updating procedures. The second modelling approach uses neural network models, which use a ‘pattern‐matching’ process to produce model forecasts.Following calibration, the models are evaluated in terms of their fit to continuous stage data and flood event magnitudes and timings within a validation period. Forecast times of 1 h, 2 h and 4 h are selected (the catchment has a response time of approximately 4 h). The ‘traditional’ models generally perform adequately at all three forecast times. The neural networks produce reasonable forecasts of small‐ to medium‐sized flood events but have difficulty in forecasting the magnitude of the larger flood events in the validation period. Possible modifications to the latter approach are discussed. © Crown copyright 2002. Reproduced with the permission of Her Majesty's stationery office. Published by John Wiley & Sons, Ltd.     

Climate and water budget change of a Mediterranean coastal watershed, Ravenna, Italy

It is generally difficult to quantify exactly the freshwater going in or out of the coastal watersheds along the northern Adriatic Sea because, on one hand, excess water is drained and pumped into the sea to prevent flooding but, on the other hand, water is brought onto the land from far away for irrigation. Fragmentation of water authorities makes it difficult to collect all the necessary information. Climate change and increasing salinization of the coastal aquifers make it imperative, however, to better know the quantities of freshwater involved in these small basins. The water budget of a small coastal agricultural watershed along the Adriatic Sea in Italy (The Quinto Basin near Ravenna) is presented here considering different land uses. The evaporation of open water and the evapotranspiration of wetlands, pine forests, bare soil and irrigated agriculture are calculated based on the Penman–Monteith equation and the Cropwat program. The current water budget is based on average climate data from 1989 to 2008 and drainage and irrigation data. Predictions for future evapotranspiration, net irrigation and hydrologic deficit are calculated with climate data from IPCC (The Fourth Assessment Report (AR4) 200, Climate change 2007). From the study results, the soil type may determine whether or not a crop will need more or less irrigation in the future. Regulations on land use should therefore consider which crop type can be grown on a specific soil type. Water budget analysis in scenarios A1b and A2 both show an increase of water deficits in the summer and an increase of water surplus in the winter. This is explained by the fact that a larger percentage of the rain will fall in winter and not during the growth season. The open water evaporation will decrease under future climate scenarios as a result of increased relative humidity in winter and decreased wind velocity. This may have a positive effect on the water cycle. The current irrigation is very abundant, but has beneficial effects in contrasting soil salinization and saltwater intrusion into the coastal aquifer.     

The 'remarkable' M31 globular cluster 037–B327 revisited

The M31 globular cluster candidate     has long been known to be an extremely red, non-stellar object. The first published spectrum of this object is used to confirm that it is a globular cluster belonging to M31, with rather typical values of     and     . Using the spectroscopic metallicity to predict the intrinsic colours, we derive a reddening value of     , in good agreement with the value obtained using reddening-free parameters. The extinction-corrected magnitude of     is     (absolute magnitude     , which makes it the most luminous globular cluster in M31. We examine van den Bergh's argument regarding the brightest and most-reddened globular cluster in M31; we find that the brightest clusters are more heavily reddened than average, but this can be explained by selection effects rather than a different R _V in M31.     

Space and time in river bank erosion research: a review

The intention of this article is to identify the ways in which ideas about space and time are manifest in river bank erosion research, thus linking abstract theoretical discussion with the 'practice' of geomorphological research. Bank erosion research to date has involved a range of space- and time-scales, different perspectives resulting from these views. Scale linkage via mathematical translation has been attempted in some cases, but studies are often contextualized through loosely hierarchical ideas. The application of hierarchy theory and notions of space–time are considered, concluding with the identification of questions that might usefully be considered in future research     

Retrieval of boreal forest LAI using a forest reflectance model and empirical regressions

Spectral invariants provide a novel approach for characterizing canopy structure in forest reflectance models and for mapping biophysical variables using satellite images. We applied a photon recollision probability (p) based forest reflectance model (PARAS) to retrieve leaf area index (LAI) from fine resolution SPOT HRVIR and Landsat ETM+ satellite data. First, PARAS was parameterized using an extensive database of LAI-2000 measurements from five conifer-dominated boreal forest sites in Finland, and mixtures of field-measured forest understory spectra. The selected vegetation indices (e.g. reduced simple ratio, RSR), neural networks and kNN method were used to retrieve effective LAI (L_e) based on reflectance model simulations. For comparison, we established empirical vegetation index-LAI regression models for our study sites. The empirical RSR–L_e regression performed best when applied to an independent test site in southern Finland [RMSE 0.57 (24.2%)]. However, the difference to the best reflectance model based retrievals produced by neural networks was only marginal [RMSE 0.59 (25.1%)]. According to this study, the PARAS model provides a simple and flexible modelling tool for calibrating algorithms for LAI retrieval in conifer-dominated boreal forests. The advantage of PARAS is that it directly uses field measurements to parameterize canopy structure (LAI-2000, hemispherical photographs) and optical properties of foliage and understory.     

Voronoi grids conforming to 3D structural features

Flow simulation in a reservoir can be highly impacted by upscaling errors. These errors can be reduced by using simulation grids with cells as homogeneous as possible, hence conformable to horizons and faults. In this paper, the coordinates of 3D Voronoi seeds are optimized so that Voronoi cell facets honor the structural features. These features are modeled by piecewise linear complex (PLC). The optimization minimizes a function made of two parts: (1) a barycentric function, which ensures that the cells will be of good quality by maximizing their compactness; and (2) a conformity function, which allows to minimize the volume of cells that is isolated from the Voronoi seed w.r.t., a structural feature. To determine the isolated volume, a local approximation of the structural feature inside the Voronoi cells is used to cut the cells. It improves the algorithm efficiency and robustness compared to an exact cutting procedure. This method, used jointly with an adaptive gradient solver to minimize the function, allows dealing with complex 3D geological cases. It always produces a Voronoi simulation grid with the desired number of cells.     

Water budget management of a coastal pine forest in a Mediterranean catchment (Marina Romea, Ravenna, Italy)

In Mediterranean coastal catchments, water management for preservation of pine forests and other natural areas faces particular challenges. Limited rainfall, water consumption by vegetation as well as subsidence, drainage and salt water intrusion all play an important role. Traditionally forest and water management are carried out independent of one another and do not consider water budget calculations. We show with this study that is very important to have quantitative information of all the components of the water budget as well as the size of the fresh water lenses in the aquifer to be able to integrate the water- and forest management. We use an integrated hydrologic-ecologic methodology based on easily attainable data to assess the monthly water budget of a coastal catchment, Marina Romea (Ravenna, Italy). We present detailed monthly water table records, rainfall data, drainage data, tree density and tree perimeter and use published sap flow measurements of single pine trees (Pinus Pinea) to quantify the actual transpiration of single pine trees in different periods of the year. Transpiration amounts to 10–30 l per day per tree. These values are confirmed by independent estimates of tree transpiration based on our water budget calculations: 9–34 l/tree/day. Because typically there are so many trees in planted pine forests, the total transpiration rates over the whole watershed take up a large percentage (up to 200 %) of the precipitation. In Marina Romea, four monitoring periods out of twelve, the tree transpiration is larger than precipitation. In nine monitoring periods, drainage in the watershed is larger than precipitation or tree water transpiration. The measurements and calculations show that not much freshwater is left to recharge the fresh water lens underneath Marina Romea. Monthly monitoring of groundwater table elevation and salinity in the pine forest of Marina Romea from March 2007 to February 2008 shows that the groundwater table strongly fluctuates and groundwater salinity is constantly very high (up to 17.7 g/l). Analytical calculations based on the Ghyben Herzberg Dupuit principle suggest that even a small continuous annual recharge of 15 mm could form a 2-m deep freshwater lens in the unconfined aquifer. This freshwater lens is not present in the study area and this is due to the fact that tree water transpiration and drainage take out most of the fresh water coming into the watershed. In catchments like Marina Romea, water consumption by the (natural) vegetation and seasonal differences as well as the fact that fresh water lenses are limited in salty surroundings should be taken into account in water and forest management.     

Organochlorine contamination in some odontoceti species from the North Pacific and Indian Ocean

Concentrations of persistent organochlorines were determined in the blubber and melon of 11 species of adult male odontocetis collected from the North Pacific, Indian Ocean and nearby seas. Mean concentrations of DDs (33 μg/g wet wt) were the highest followed by PCBs (32 μg/g wet wt), chlordane compounds (CHLs: 3.7 μg/g wet wt), HCHs (1.1 μg/g wet wt), and HCB (0.32 μg/g wet wt). Odontoceti species inhabiting temperate waters revealed maximum residual concentrations of these contaminants, and the elevated DDT and PCB residues detected seem to suggest that some of the present species might potentially be at high risk. Relatively high DDT concentrations were found in tropical water species, which could be attributed to the current usage of DDT in the tropics and the less movable nature of this compound via long-range atmospheric transport. The HCH levels in animals inhabiting cold and temperate waters were higher than those inhabiting tropical waters, a result that was perhaps reflective of atmospheric transport from the tropical source to the northern sinks. A similar pattern was also observed in PCBs, CHLs and HCB, probably indicating the ongoing discharge of these compounds from mid-latitudes as well as those originating in tropical regions.     

Understanding the evolution of syn-depositional folds: Coupling decompaction and 3D sequential restoration

The analysis of basin dynamics and burial evolution requires a good understanding of sediment compaction. Classically, decompaction of sediments is performed in one dimension at a well location, using either a simple compaction/depth relationship or more complex elasto-plastic models. This paper presents a new approach combining sequential decompaction with 3D restoration to allow for a true 3D basin analysis. Decompaction is performed in 3D after each restoration step, thus taking into account possible tectonic events and lateral thickness variations. Care is taken to apply decompaction to ensure volume continuity especially around faults. This approach is particularly suitable for syn-depositional folds whose growth strata constrain tectonic evolution through time.The proposed approach is applied to the sand-rich turbiditic reservoir analogue of Annot (SE France) where two fictitious wells are used to compare the new 3D technique to a well decompaction analysis. Coupling restoration and decompaction leads to an improved assessment of the basin history: an uplift of the underlying units is identified, which was not detected using decompaction on wells only. Such differences may have a significant impact on possible hydrocarbon maturation models of the basin. Moreover, the geometry of the restored and decompacted models can better constrains the basin history, and influence our understanding of potential hydrocarbon migration pathways.