Opportunities and Limits of Using Meteorological Reanalysis Data for Simulating Seasonal to Sub-Daily Water Temperature Dynamics in a Large Shallow Lake

Marieke Frassl, Bertram Boehrer, Peter Holtermann, Weiping Hu, Knut Klingbeil, Zhaoliang Peng, Jinge Zhu, Karsten Rinke
2018 Water  
In lakes and reservoirs, physical processes control temperature dynamics and stratification, which are important determinants of water quality. In large lakes, even extensive monitoring programs leave some of the patterns undiscovered and unresolved. Lake models can complement measurements in higher spatial and temporal resolution. These models require a set of driving data, particularly meteorological input data, which are compulsory to the models but at many locations not available at the
more » ... vailable at the desired scale or quality. It remains an open question whether these meteorological input data can be acquired in a sufficient quality by employing atmospheric models. In this study, we used the European Centre for Medium-Range Weather Forecasts' (ECMWF) ERA-Interim atmospheric reanalysis data as meteorological forcing for the three-dimensional hydrodynamic General Estuarine Transport Model (GETM). With this combination, we modelled the spatio-temporal variation in water temperature in the large, shallow Lake Chaohu, China. The model succeeded in reproducing the seasonal patterns of cooling and warming. While the model did predict diurnal patterns, these patterns were not precise enough to correctly estimate the extent of short stratification events. Nevertheless, applying reanalysis data proved useful for simulating general patterns of stratification dynamics and seasonal thermodynamics in a large shallow lake over the year. Utilising reanalysis data together with hydrodynamic models can, therefore, inform about water temperature dynamics in the respective water bodies and, by that, complement local measurements. generally more susceptible to atmospheric forcing like irradiance, air temperature and wind friction [5]. This does not only refer to variables related to hydro-and thermodynamics but also affects water quality variables. For example, higher water temperatures combined with high nutrient concentrations lead to the frequent occurrence of persistent cyanobacteria blooms [6, 7] . Usually, surface scums of cyanobacteria are not distributed homogeneously, but are constantly transported by wind and occur at a spatially-variable severity (e.g., [8] ). This can result in larger horizontal than vertical gradients in shallow lakes [9, 10] . Pronounced spatial patterns also have been observed for the occurrence of hypoxic zones, which lead to the dissolution of reduced metals and, thus, water quality deterioration (e.g., [11, 12] ). This spatial variability, which is primarily caused by hydrodynamics, is difficult to cover through observations alone: ground-based measurements by moorings or ship-based measurements cannot be implemented in the required spatial and temporal resolution. While remote sensing develops into a promising technology to support the monitoring of lake surfaces (see the reviews by [13, 14] , the special issues [15, 16] and the overview by [17]), it does not resolve the vertical dimension in the water column. In order to complement local measurements and to expand information along spatial and temporal scales, numerical models have been proven useful [18] . Depending on the question, they can be designed from one-dimensional models that resolve the vertical dimension to rather complex three-dimensional models. The latter have the benefit that they resolve both horizontal and vertical heterogeneity, as well as three-dimensional circulation patterns. In oceanography, three-dimensional models have a long history of application (see e.g., [19] ). They became more popular in limnological studies in the early 2000s [20] [21] [22] ) and are now used frequently to understand physical (e.g., [23, 24] ) and biological (e.g., [25, 26] ) processes in lakes and to assess the effects of lake management on oxic conditions [27] or phytoplankton dynamics [28] . In shallow lakes, three-dimensional models are commonly applied to analyze the occurrence of anoxia [29] , resuspension events [30], cyanobacteria blooms [31] and to support management, e.g., by identifying critical nutrient loads [32, 33] . Lake Taihu, a shallow lake similar to Lake Chaohu, has a long history of three-dimensional model applications (for an overview, see [34] ). Three-dimensional lake models have the advantage being able to resolve most physical processes in the water body, e.g., long internal waves, upwelling and to analyse spatial phenomena like the distribution of surface scums (e.g., [35] ) or the flux of nutrients through the benthic boundary layer [20, 36] under pre-defined conditions. One open-source and publicly available three-dimensional hydrodynamic model is the General Estuarine Transport Model (GETM). GETM was originally developed for coastal ocean applications (see the review by [37]), but was also successfully applied to lakes (e.g., [38, 39] ). Important features for lake modelling are adaptive terrain-following coordinates [40] . This facilitates a proper vertical resolution of boundary layers. Within GETM, non-hydrostatic effects [41] , which are required for high-resolution studies, can optionally be included. State-of-the-art vertical turbulence closure is provided via an interface to the General Ocean Turbulence Model (GOTM; [42] ). Three-dimensional models require a large amount of data for forcing and validation. In countries with a large territory or areas that are not easily accessible due to a lack of infrastructure (e.g., in some tropical countries or arctic regions), the high demand of input data is hard to satisfy since the necessary data to drive a complex model are in many cases not easily available at the required spatial and temporal resolution (e.g., [43] ). Alternatively, local forcing data could be obtained from e.g., high-resolution atmospheric models (e.g., [44] ). However, these local atmospheric models need to be nested into large-scale global models [45, 46] or are often specific for a certain meteorological variable and developed for a specific region (e.g., [47] ). The question arises, whether coarse meteorological and hydrological data from global models can be used directly for driving local lake models. This methodology would be easily transferable to other water bodies worldwide and would help to apply lake models in regions with scarce monitoring data. Reanalysis projects assimilate weather forecast models with local observations leading to global datasets of atmospheric circulation [48] . The same data assimilation technique is applied over the
doi:10.3390/w10050594 fatcat:sv7xrdsrpvbtlpzczt4thlydma