Bruno Tassin

Application of MCMC-GSA model calibration method to urban runoff quality modeling

Abstract In stormwater quality modeling, estimating the confidence level in conceptual model parameters is necessary but difficult. The applicability and the effectiveness of a method for model calibration and model uncertainty analysis in the case of a four parameters lumped urban runoff quality model are illustrated in this paper. This method consists of a combination of the Metropolis algorithm for parameters’ uncertainties and correlation assessment and a variance-based method for global sensitivity analysis. The use of the Metropolis algorithm to estimate the posterior distribution of parameters through a likelihood measure allows the replicated Latin hypercube sampling method to compute the parameters’ importance measures. Calibration results illustrate the usefulness of the Metropolis algorithm in the assessment of parameters’ uncertainties and their interaction structure. The sensitivity analysis demonstrates the insignificance of some parameters in terms of driving the model to have a good conformity with the data. This method provides a realistic evaluation of the conceptual description of the processes used in models and a progress in our capability to assess parameters’ uncertainties.

Impact of internal waves on the spatial distribution of Planktothrix rubescens (cyanobacteria) in an alpine lake

The vertical and horizontal distribution of the cyanobacterium, Planktothrix rubescens, was studied in a deep alpine lake (Lac du Bourget) in a 2-year monitoring program with 11 sampling points, and a 24-h survey at one sampling station. This species is known to proliferate in the metalimnic layer of numerous deep mesotrophic lakes in temperate areas, and also to produce hepatotoxins. When looking at the distribution of P. rubescens at the scale of the entire lake, we found large variations (up to 10u2009m) in the depth of the biomass peak in the water column. These variations were closely correlated to isotherm displacements. We also found significant variations in the distribution of the cyanobacterial biomass in the northern and southern parts of the lake. We used a physical modeling approach to demonstrate that two internal wave modes can explain these variations. Internal waves are generated by wind events, but can still be detected several days after the end of these events. Finally, our 24-h survey at one sampling point demonstrated that the V1H1 sinusoidal motion could evolve into nonlinear fronts. All these findings show that internal waves have a major impact on the distribution of P. rubescens proliferating in the metalimnic layer of a deep lake, and that this process could influence the growth of this species by a direct impact on light availability.

1-D coupled physical-biological model of the northwestern Mediterranean Sea

Abstract A 1-D coupled physical-biological model of the ecosystem of the Ligurian Sea is described. The physical model has been adapted from a thermal dispersive lake model (Tassin, 1986), and the biological model from the sub-models AQUAPHY (Lancelot et al., 199l a) and H.S.B. (Billen and Servais, 1989). The biological model considers two types of phytoplankton, diatoms and non-siliceous flagellates, to describe the response of the phytoplankton community and the related food-web to different types of fertilization (by river, atmospheric deposition or deep water upwelling) with different N:Si ratios. This distinction, and the detailed description of the microbial loop, helps to distinguish between new and regenerated primary production. The model has been calibrated on FRONTAL 1986 data, obtained monthly at a long-term station (43°24′N, 07°52′E). The quantitative biomass, its repartition between diatoms and flagellates, and its spatio-temporal variability are reasonably well reproduced by the model. From these simulations, an annual new production of 62 g C m−2 is computed, contributing 35% of total annual primary production (180 g C m−2). This calculation reflects the oligotrophic state of the system, where regenerated production occurs most of the year while the biomass remains low. A Monte-Carlo sensitivity analysis performed on the parameters driving the model leads to a confidence interval for the simulation and also to a better understanding of how the model functions.

Tracing calcite precipitation with specific conductance in a hard water alpine lake (Lake Bourget)

Abstract The authors discuss the reliability of a new method used to depict precisely the authigenic calcite precipitation process in the water column of a mesotrophic alpine lake (Lake Bourget, Savoie, France). Complete chemical analysis of dissolved major elements shows that, in the epilimnion, calcium concentration and alkalinity are the only chemical species presenting large variations. The behaviour of these two parameters is only related to calcite precipitation. The method is based on in situ specific conductance measurements ( C 25 ). Experimental values were compared with theoretical calculations. The results demonstrate clearly that dissolved calcium concentration can be directly related to measured specific conductance. Hence, a mathematical relationship was developed to compute calcium concentration profiles in the water column with a centimetre spatial resolution through the whole water column. Finally, authigenic calcite production can be observed and quantified on a time scale covering two years of experimental data. Calcite production is estimated to 4.8xa0mol m −2 yr −1 . Extrapolation to the whole lake surface leads to 16,000±2000 tons yr −1 as CaCO 3 . Specific conductance is an easy-to-measure in situ parameter which could be used to trace authigenic calcite production in other hard water alpine lakes.

Temporal evolution of urban wet weather pollution: analysis of PCB and PAH in sediment cores from Lake Bourget, France

In order to evaluate the historical impact of urban stormwater runoff on a lacustrine environment, eight sediment cores were collected in October 2006 in Lake Bourget (Savoie, France). Sediment samples were analysed for polycyclic aromatic hydrocarbons (PAH) and polychlorobyphenyls (PCB) in order to correlate the vertical profiles of contamination and the evolution of the urbanisation on the watershed for the last 100 years. Overall, we observe that vertical profiles of PCB and PAH concentrations in Lake Bourget are showing a downward trend, probably resulting from the pollution control measures. Concentrations of PCB and PAH measured in surface sediments are approximately 5-10 fold less than maximal concentrations found in the cores. Maximal PAH concentrations were found at the bottom of the cores ( approximately 40 cm) which approximately correspond to the beginning of the 20th century. Maximal PCB concentrations measured in cores L5 and L6 are respectively 850 and 790 microg/g dw (dryweight). The peak of PCB corresponds to the maximum of PCB production and consumption in the 1970s. In surface sediments total PCB concentrations range between 47 and 79 microg/kg dw. Those concentrations are particularly high compared to other values measured in lake sediments and highlight the problem of a local source of PCB contamination in the watershed. General trends were observed, yet the part of urban stormwater runoff in the total contamination could not be determined.

Global warming as a driving factor for cyanobacterial blooms in Lake Karaoun, Lebanon

AbstractThe Middle East region suffers already from the gradual effects of climate change and will be among the most vulnerable regions in the future. As a result, productivity should undergo losses due to high temperatures, drought, floods, and soil degradation which threaten food security of Levantine countries. Since water is the critical factor in the region, even slight changes in air temperature and rainfall patterns will have considerable impact. It has been proven that potential climate change may disrupt, on one hand, most ecosystems through changes in their physicochemical conditions, and on the other hand the species which are living in these ecosystems. Then, the biodiversity can be found challenging. In this study, the effects of climate change on population and phytoplankton communities of Lake Karaoun were investigated since 1992. The climate regime shifts have been shown to alter the lake ecosystem. In the past, Lake Karaoun was characterized by a highly diversified microflora dominated by...

Contribution of mathematical modeling to lake ecosystem understanding: Lake Bourget (Savoy, France)

This paper demonstrates how mathematical modeling can contribute to improve understanding of lake behavior. Since the 60s Lake Bourget, one of the largest in France, had been suffering from eutrophication which was checked in 1980 by the diversion of the main sewers entering the lake. A research program was implemented between 1987 and 1990, including an on-site sampling campaign conducted concurrently with thermal and biogeochemical modeling of lake behavior. The model helped provide a better understanding of the ecosystem, displaying some processes hitherto misunderstood: (1) Winter overturn does not reach the bottom of the water column when the weather is mild. This leads to a incomplete reoxygenation of the hypolimnion and to redox conditions inducing the release of orthophosphate from the sediment, (2) Grazing by herbivorous zooplankton is getting more important in the control of spring algal growth as eutrophication of the lake regresses, (3) Settling of particulate phosphorus seems a complex and very important process in Lake Bourget, showing high sedimentation rates for particulate mineral phosphorus.

Long-term temperature evolution in a deep sub-alpine lake, Lake Bourget, France: how a one-dimensional model improves its trend assessment

Water temperature and the alternation of stratification and mixing count amongst key drivers of lake ecology. Reliable long time-series of water temperature are rare. Here, we investigated how a numerical model can fill the gaps in heterogeneous time-series and make it possible to identify a significant trend in the lake thermal regime. We computed the mean water temperature and the Schmidt stability, an indicator of the stratification strength, in a deep and well-stratified lake (Lake Bourget, France), between 1976 and 2008. We first used temperature measurements and then a one-dimensional vertical model, which we describe here. The model performs as well as the best existing models. During the 1976–2008 periods, whereas no statistically significant trend came out of the measurements, we found in the simulation results an increase of 0.12xa0°C per decade in the water temperature and of 5.9 days per decade in the stratification duration. This shows that the temperature history of a lake can be reconstructed and a reliable long-term trend computed when weather data and a calibrated temperature model are available. Our results also suggest that different local climatic influences cause a slower warming in this lake of the south-western Alps than in other sub-alpine lakes.

Impact of a flood event on the biogeochemical behaviour of a mesotrophic alpine lake : Lake Bourget (Savoy)

A flood event which occurred during the onset of thermal stratification and of algal growth (March 1996) was studied in mesotrophic Lake Bourget (France). Both physical and biogeoehemical processes occurring during this episode were assessed. The dominant effect was a decrease of nutrient concentrations along the river—lake ecotone. This phenomenon seemed mainly linked to biological factors: stimulation of the spring algal and bacterial growth. After the flood, the algae aggregated with allochthonous particles brought by the river and were rapidly eliminated by sedimentation. The effect of this early spring flood, seemed spatially restricted and rather neutral as regards lake water quality. In the lake region crossed by the river plume, the overall effect after the flood was a decrease of P available for the phytoplankton. In the open water, the P concentration was probably not modified by the load supplied by the flood.

Forecasting of water quality in lakes: A predictive use of a one-dimensional model. Application to lake Bourget (Savoie, France)

In order to assess the efficiency of eutrophication control programs, predictive models are necessary. We propose a methodology for implementing such a model, based on the coupling of a biogeochemical model of a lake and the use of long term time series of meteorological data. This methodology is applied to lake Bourget (Savoie, France). It allows to obtain both mean and standard deviation (first and second order moments) of the state variables of the model on a 1 y period. The sensitivity of the model to the various forcing variables, as well as to the initial conditions is analyzed as well as the linear or non-linear behavior of the model. Finally, the propagation of the uncertainties (standard deviations) in time and space, for both water temperature and dissolved oxygen are assessed.