Physics and Chemistry of the Earth, Part B: Hydrology, Oceans and Atmosphere最新文献

LINCOM is a fast linearised and spectral wind flow model for use over hilly terrain. It is designed to rapidly generate mean wind field predictions which provide input to atmospheric dispersion models and wind engineering applications. The thermal module, LINCOM-T, has recently been improved to provide reasonably robust results over a range of stability conditions. The results predicted for idealised terrain only are presented here. Meteorological data used to initialise the model are normally obtained from measurements or from outputs from larger scale numerical models. These standard data types have therefore been used to calculate the meteorological parameters required by LINCOM-T. The effect of the formulation of these parameters on the perturbed velocity field has been investigated in detail.

Currently, more deposition of reduced nitrogen species to the UK is measured than can be accounted for from the calculated emissions. A newly developed long-range transport model was used to study this budget. Changes in horizontal resolution of model and an improved parameterization of dry deposition of ammonia gave an improved agreement between measured and modelled deposition. However, the model still estimated only 65% of the measured deposition with a UK emissions term of 283 ktonnes N yr⁻¹. Scaling the emissions by factors of × 1.2 and × 1.3 still resulted in the model underestimating of reduced nitrogen deposition. Modelled aerosol ammonium and ammonia gas concentrations for the base-case were compared with observations. The ammonia gas concentrations appear to be modelled reasonably well but the aerosol ammonium concentrations were underestimated by a factor of at least 2. This is an important finding as this indicates that other parameters related to wet deposition simulation are unlikely to be in gross error, implying that the UK emissions term is too small. It was concluded that an emission source term of the order 400 ktonnes N yr⁻¹ is required to reconcile the model with the measurements.

A laboratory experimental set - up for studying the behaviour of sediment in presence of a turbulent field with zero mean flow is presented. Particular interest is shown on the initiation of sediment motion and in the sediment lift - off. Some examples of the results obtained with this set-up are shown.

A turbulent velocity u′ lower than that u estimated by the Shield diagram is required to start sediment motion. The minimum u′ required to start sediment lift - off, is a function of sediment size, cohesivity and resting time. The lutocline height depends on u′, and the vorticity at the lutocline seems constant for a fixed sediment size. Combining grid stirring and image analysis, sediment vertical fluxes and settling speeds could be measured.

Several impoundment schemes have been built or are proposed in the UK and worldwide. Impounding fundamentally alters the dynamics of estuaries with consequences in terms of sedimentation patterns and rates as well as water quality. This paper presents some initial findings on the relationships of water quality to river flow based on work done on the Tees impoundment in Northeast England (a total exclusion system) in the summer of 1999 and the winter of 2000. Eleven water quality surveys were undertaken to measure a range of water quality parameters (BOD, alkalinity, pH, Eh, conductivity, dissolved oxygen, temperature, suspended solids, nitrate plus nitrite, ammonium and orthophosphate, and a range of metals) over a variety of states of river flow. Statistical analysis was used to determine whether the major water quality parameters were controlled by processes internal to the impoundment, i.e. with water depth and distance downstream, or external inputs to the impoundment. Those reported as controlled by external inputs include nitrate and nitrite, phosphate, dissolved oxygen, dissolved metals, conductivity, alkalinity, and temperature. pH, and to a minor extent dissolved oxygen and nutrients, show internal control. Understanding the controls on water quality parameters will allow more sustainable management such schemes.

A 3-D Eulerian transport-chemistry model, REGINA (REGIonal high resolutionN Air pollution model), based on models developed over the last decade at the National Environmental Research Institute (DREAM, DEHM, ACDEP and DEOM) is currently under development. The model will be applied to studies of air pollution phenomena (both monitoring, forecasts and scenarios) over Denmark, where there are extensive coastal areas that require a high resolution model in order to resolve the effects of e.g. land-sea interactions. The goal is to obtain a nested grid model capable of high resolution operation. It is very important to implement numerical methods suited for the high model resolution in order to make sure that significant errors are not introduced by the numerics. In this paper the model concepts of REGINA will be described. Special attention will be on the numerical methods and numerical test results concerning the advection and chemistry. Furthermore examples of model results and some preliminary validations with measurements from EMEP will be shown. The numerical test results show, that the methods chosen for the model are very accurate with small numerical errors. The next step will be to implement high resolution input data (emission, land use and meteorological data) and physical parameterisations.

Sharply defined water density gradients (due principally to salinity contrasts but also to temperature effects) and velocity gradients occur both longitudinally and transversely in many estuaries. These gradients are related to frontal systems (fronts) which, at the water surface, are typically manifest as lines or bands of foam, floating debris or distinct changes in the colour and transparency of the water. Fronts, which form by several mechanisms, including tidal intrusion, axial convergence, advective flow and flow separation, acts as temporary barriers, serve to inhibit exchange of water masses and to entrap fine particulate materials. It is suggested that, in terms of suspended sediment transport through and within estuaries that are characterized by fronts, such features should be considered as “sieves” in the estuarine sediment transfer system. Where characterised by many fronts, an estuary as a whole should most appropriately be considered as a complex of sieves, which collectively create a dynamic “sieve regime”.

One of the many types of neural networks that found application in hydrology is the probabilistic neural networks. Probabilistic neural networks are based upon the Parzen approximation of probability densities by (Gaussian) kernels. The advantages of probabilistic neural networks are that they learn extremely quickly, give probabilistic interpretation and by this not only produce estimation of the mean but also give insight into the other statistics of the errors.

When (in higher dimensions) the observations tend to cluster around lower dimensional subspaces, the classical approach fails by not being able to take this into account. The solution proposed here is to use a local version, based upon Gaussian kernels with locally estimated covariances. This concept resembles the “local and global embedding dimension” used in (classical) deterministic time series analysis.

As an example, results on predicting discharges in a small catchment will be presented. Inputs are lagged discharges. If the time discretisation scale is rather small, and one uses many lags, the input space becomes high dimensional but the observations by the mutual dependence between the components of the input fill only a lower dimensional subspace of this. It will be shown that this new technique offers better results in these cases.

There are growing demands for effective public participation in river basin management. These are posed among others by the proposed EU Water Framework Directive as well as international conventions and policy documents. The demands will raise a need for guidance on exactly how the public is to be involved. In public participation ‘decision support tools’ (DST) and integrated modelling can have a role to play. Many computer tools not only can serve as tools for analysis for experts, but also as vehicles for communication, training, forecasting and experimentation. Illustrated by selected examples of software a range of tools and their potential applications are presented. Essential for the added-value of computer tools in stakeholder participation, in comparison to other methods like expert presentations, fact sheets, etc., is the way they are used. Lessons can be drawn from projects where computer tools were used in other policy areas. Within an European research project on energy and climate policy (ULYSSES) integrated models on global climate change were used in integrated assessment (IA) focus groups. A web-based tutorial for the use of computer models was developed. It gives guidelines for the design and setup of participatory arrangements in conjunction with computer models. Besides procedural recommendations lessons can be drawn concerning the kind of output the tools should provide and the early involvement of users is modelling and software development.

The overall objective of the EU project ARSGISIP is to promote the application of Earth Observation (EO) techniques and GIS integration for the parameterization of hydrological, erosion, and solute transport models. Activities of the first two project years are presented, emphasizing the common environmental problem of nutrient leaching within three representative basins located in Germany, Austria and Sweden. The scientific approach is focussed on physical parameters describing soil and vegetation cover by means of multi-resolution and multi-spectral land use classifications using Landsat-5 TM and IRS-IC PAN as well as ERS-2 radar data. Results confirm that major land use classes and crops can be classified with high accuracy with multi-temporal optical data. Multi-temporal ERS-2 radar data support the separation of major crop types.