HBV-modellen är för prognosändamål kalibrerad i 32 områden. Under våren 1985 användes den för prognoser i 27 av dessa områden. Prognoser med arealnederbördsmetoden gjordes i 43 områden.
Prognoserna var av 3 typer:
HBV-modellen användes för alla 3 typerna av prognoser, medan arealnederbördsmetoden endast är tillämpbar på prognoser av typ 1.
HBV-modellen är för prognosändamål kalibrerad i 37 områden. Under våren 1987 användes den för prognoser i 28 av dessa områden. Prognoser med Arealnederbördsmetoden gjordes i 43 områden.Prognoserna var av 3 typer:
HBV-modellen användes för alla 3 typerna av prognoser, medan Arealnederbördsmetoden endast är tillämpbar på prognoser av typ 1.
HBV-modellen är en hydrologisk-matematisk modell för beräkning av vattenföring utgående från meteorologiska data. Grundversionen av modellen finns beskriven i rapport RHO (Bergström,1976). En sammanfattande beskrivning av modellstruktur och prognosmetodik ges i rapport HO 7 (Persson, 1983). Arealnederbördsmetoden baseras på regressionssamband mellan tillrinningsvolym och nederbörd. Det finns beskriven i HB-rapport nr 52 (Häggström 1982)
I det följande redovisas prognosresultaten 1987 tillsammans med en utvärdering av effekterna av den nederbörd, som verkligen föll under prognosperioderna. Avsnitt 1 ger en sammanfattning av nederbörd, temperatur, magasinsförhållanden och vattenföring. Avsnitt 2 behandlar HBV-modellen. Avsnitt 3 behandlar resultat för Arealnederbördsmetoden samt för en prognos baserad på snötäckets vatteninnehåll.
Rapporten riktar sig i första hand till dem, som tagit del av prognoserna, och förutsätter viss kännedom om prognosmetoderna. Motsvarande rapporter för 1979-1986 års prognossäsonger finns tillgängliga vid SMHI.
The evapotranspiration routines in existing conceptual hydrological models have been identified as one of the weaknesses which appear when these rnodels are used for the simulation of hydrological effects of a changing clirnate. The hydrological models in operational use today usually have a very superficial description of evapotranspiration. They have, nevertheless, been able to yield reasonable results. The objective of this paper is to analyse and suggest modifications of existing evapotranspiration routines in conceptual hydrological models to make them more appropriate for use in simulation of the effects of a changing climate on water resources.
The following rnodifications of the evapotranspiration routine were formulated and tested in the HBV model: Ternperature anomaly correction of evapotranspiration, potential evapotranspiration by a simplified Thornthwaite type formula, interception submodel, spatially distributed evapotranspiration routine and alternative forrnulations of lake evapotranspiration. Sensitivity analyses were thereafter rnade to illustrate the effects of uncertainty in the hydrological model structure versus those of the uncertainty in the climate change predictions.
[1] We aimed at investigating whether the spatial variability of infiltration in field soils, as visualized through dye infiltration experiments, is characterized by a multiscaling behavior. Digitized high-resolution dye images from three sites in an experimental catchment in Tunisia were analyzed using three indicators of scaling: empirical probability distribution functions, power spectra, and raw statistical moments. The two former indicators suggested a general scaling behavior of the data, which through the moments' analysis was found to be of multiscaling type. Random cascade processes are frequently used to model multiscaling processes, and we fitted the "universal multifractal'' (UM) model of Schertzer and Lovejoy [1987] to our data. The UM model closely reproduced the empirical K(q) functions, and simulated fields reproduced key features in the observed ones. The results indicate that multiscaling random cascade modeling is useful for statistically describing flow processes and solute transport under field conditions.
The small-scale variability and scaling properties of solute transport dynamics were investigated by laboratory experiments. Dye-stained water was applied at a constant flux in a Plexiglas Hele-Shaw cell filled with soil material, and the transport process was registered by digital photographs of the cell front, producing a very high resolution in both time (minutes) and space (millimeters). The experiments comprised different cell materials (uniform and natural sand) and different water fluxes. Scaling properties of vertically integrated dye mass distributions were analyzed and modeled using so-called breakdown coefficients (BDCs), which represent the multiplicative weights in a microcanonical random cascade process. The pdfs of BDCs varied with scale, indicating self-affine scaling, and were accurately approximated by beta distributions with a scaling parameter. Two versions of BDC-based random cascade models were used to simulate mass distributions at different time steps. The results support the applicability of random cascade models to subsurface transport processes.
Accurate measurements of solute concentration are needed to conduct studies of solute transport process in unsaturated soil. In this paper we present a method of obtaining accurate measurements in and space using dye infiltration and image analysis. The soil color was related to the dye concentration in the soil (C-s) using 74 small calibration samples. The overall root mean square error (RMSE) 0.057 g dm(-3), however, for C-s < 0.75 g dm(-3), the RMSE was only 0.032 g dm(-3). Variability of the concentration estimates at the pixel scale could be reduced by using an average filter. We used the calibration relationship during four infiltration experiments in a 0.95 by 0.975 m large Plexiglas Hele-Shaw cell to calculate dye concentration patterns. Using the first and second order time moments, the dispersivity lambda was calculated for nine different artificial column widths, from 0.0014 (local-scale) to 0.72 m (meso-scale). The horizontally averaged lambda proved to be identical for column widths from 0.0014 to 0.045 m. For larger scales, lambda gradually increased. We noticed that the two experiments with higher flow (1 and 2) and the two experiments with lower flow (3 and 4) showed an almost identical variation of meso-scale lambda with depth. We concluded that above a specific critical value of theta (similar to 0.22 m(3) m(-3)), solute mixing is enhanced, leading to a lower lambda, and that solute transport can be described as a convective-dispersive process. When theta is lower than this critical level, part of the porosity is deactivated and mixing between individual stream tubes decreases, which implies that transport then occurs as a stochastic-convective process.
Four unsaturated solute transport experiments with different water fluxes were conducted in a Hele-Shaw cell filled with uniform sand. The transport of the dye tracer used was recorded with a camera and the dye concentration was calculated using image analysis. The concentrations fields were analysed in terms of time moments and converted into vertical solute transport velocity V. Both mean value and standard deviation of V increased with water flux. The autocorrelation function exhibited a linear decrease for short lags. The pronounced variability of V suggested a description in terms of scaling properties, and a scaling regime was indeed found from the resolution 1.8 mm up to almost 0.1 m. The upper limit corresponds roughly to a characteristic scale of fingering structures seen in the dye concentration images. Indications of a second scaling regime at larger scales were found. In the small-scale scaling regime, the power spectrum exponent beta was generally slightly below 1 and the intermittency parameter C(1) was on average 0.00025. The moment scaling K(q) functions were convex, implying a multiscaling process.
This report describes the HBV model application for inflow forecasting on a daily time step to the Tarbela dam in the Indus River. The model application forms part of the new Tarbela control center for Pakistan's Water and Power Development Authority (W APDA). The new control system will be an ABB S.P.I.D.E.R. SCADA system, delivered on a turnkey basis by ABB Network Control of Sweden. SMHI have as subconsultants to the ABB Network Control been responsible for the set up, calibration, training and delivery of the inflow forecasting system based on the HBV model.
Korrelationskoefficienter för månadsmedelvärden av temperatur och nederbörd har beräknats för 9 svenska meteorologiska stationer, de flesta med observationer under mer än 100 år.
Korrelationen mellan två direkt på varandra följande månaders medeltemperatur varierar mellan 0.10 och 0.54 med de högsta värdena under vintern. Motsvarande samband för nederbörden liksom för temperatur mot nederbörd är betydligt svagare. Temperaturkorrelationerna mellan aktuell månad n och månad n + 2 har genomsnittsvärden på 0.11 - 0.17.
Persistensens betydelse för resultatet av vårflödesprognoserna har undersökts för två områden genom att enbart utnyttja den hälft av åren då utgångsmånadens medeltemperatur varit mest lik medeltemperaturen under det aktuella årets utgångsmånad. Därigenom har det genomsnittliga felet i prognosvolymen minskat något eller med 5.3 procent för Sädvajaure och med 3.3 procent för Trängslet.