Seasonal Spatial Patterns of Surface Water Temperature, Surface Heat Fluxes and Meteorological Forcing Over Lake Geneva

In many lakes, surface heat flux (SHF) is the most important component controlling the lake’s energy content. Accurate methods for the determination of SHF are valuable for water management, and for use in hydrological and meteorological models. Large lakes, not surprisingly, are subject to spatially and temporally varying meteorological conditions, and hence SHF. Here, we report on an investigation for estimating the SHF of a large European lake, Lake Geneva. We evaluated several bulk formulas to estimate Lake Geneva’s SHF based on different data sources. A total of 64 different surface heat flux models were realized using existing representations for different heat flux components. Data sources to run the models included meteorological data (from an operational numerical weather prediction model, COSMO-2) and lake surface water temperature (LSWT, from satellite imagery). Models were calibrated at two points in the lake for which regular depth profiles of temperature are available, and which enabled computation of the total heat content variation. The latter, computed for 03.2008-12.2013, was the metric used to rank the different models. The best calibrated model was then selected to calculate the spatial distribution of SHF. Analysis of the model results shows that evaporative and convective heat fluxes are the dominant terms controlling the spatial pattern of SHF. The former is significant in all seasons while the latter plays a role only in fall and winter. Meteorological observations illustrate that wind-sheltering, and to some extent relative humidity variability, are the main reasons for the observed large-scale spatial variability. In addition, both modeling and satellite observations indicate that, on average, the eastern part of the lake is warmer than the western part, with a greater temperature contrast in spring and summer than in fall and winter whereas the SHF spatial splitting is stronger in fall and winter. This is mainly due to negative heat flux values (net cooling) and stronger wind forcing, and consequently stronger mixing, in cold seasons.

Surface heat flux variability of a large lake: Lake Geneva, Switzerland

David Andrew Barry, Damien Bouffard, Abolfazl Irani Rahaghi, Ulrich Lemmin

The heat budget of a lake is a fundamental component of physical limnology, and is strongly dependent on the surface heat flux. However, the surface energy exchange depends on several factors, making it difficult to estimate. In this study we employed several bulk formulas to estimate Lake Geneva’s surface heat flux. Combination of different surface heat flux terms leads to a surface heat exchange model which requires various data. Different data sources were used in the heat flux estimates. Meteorological data were taken from an operational numerical weather prediction model, namely COSMO-2 (run by the Swiss meteorological service), while satellite imagery was used for the lake surface water temperature (LSWT). In order to find the best combination of the bulk formulas and to calibrate the model, the temporal evolution of the heat budget was estimated using long-term time series of vertical temperature profiles. Vertical temperature profiles at two points (one in the Lake Geneva’s large basin and one in its small basin) were used. A sensitivity analysis was performed to find the key parameters, and more significantly the optimal combination of different heat flux terms. Finally, the spatio-temporal surface heat flux variation was calculated according to the proposed model. In addition, the relationship between variability of the surface heat flux and meteorological forcing was assessed. The different models, which are of differing complexity, gave reasonably consistent results, with differences attributed to simplifications inherent in them. The modeling results revealed that the LSWT and wind forcing are dominant factors underlying Lake Geneva surface heat flux spatial variation, while its temporal variability is mainly due to the global radiation and air temperature changes. In conclusion, the bulk heat balance approach is a useful tool to estimate various heat flux terms as well as their monthly or seasonally contributions. But, in large lakes where the LSWT is highly variable, the variable surface heat flux would be unavoidable.

2014

The Importance of Systematic Spatial Variability in the Surface Heat Flux of a Large Lake: A Multiannual Analysis for Lake Geneva

David Andrew Barry, Andrea Cimatoribus, Abolfazl Irani Rahaghi, Ulrich Lemmin

The spatiotemporal surface heat flux (SurHF) distribution over Lake Geneva, the largest lake in Western Europe, was estimated for a 7‐year period (2008–2014). Data sources included hourly maps of over‐the‐lake assimilated meteorological data from a validated numerical weather model and lake surface water temperature (LSWT) from satellite imagery. A set of bulk algorithms, previously optimized and calibrated at two locations in Lake Geneva, was used. Results indicate a systematic long‐term average spatial range of >40 Wm‐2 between different parts of the lake and little year‐to‐year variability. This variability is mainly due to topographically induced wind sheltering over parts of the lake, which in turn produces spatial variability in the sensible and latent heat fluxes. These results are supported by a systematic spatial heat content variability obtained from long‐term temperature profile measurements in the lake. During spring, a lower SurHF spatial range was evident. Unlike other seasons, the spring spatial variability of air‐water temperature differences and, to a lesser extent, the global radiation variability resulting from sheltering by the mountainous topography were the main drivers of the SurHF spatial variability. Analysis of the atmospheric thermal boundary layer showed stable conditions from March to early June and unstable conditions for the rest of the year. This regime change can explain the low SurHF spatial variability observed during spring. The results emphasize that spatial variability in meteorological and LSWT patterns, and consequently in the spatiotemporal SurHF data, should be considered when assessing the time evolution of the heat budget of large lakes.

2019

A Multiscale Surface Water Temperature Data Acquisition Platform: Tests on Lake Geneva, Switzerland

David Andrew Barry, Abolfazl Irani Rahaghi, Ulrich Lemmin

An improved understanding of surface transport processes is necessary to predict sediment, pollutant and phytoplankton patterns in large lakes. Lake surface water temperature (LSWT) which may spatially and temporarily vary over the lake surface reflects meteorological and climatological forcing more than any other physical lake parameter. There are different data sources for LSWT mapping, including remote sensing and in situ measurements. The available satellite data might be suitable for detecting the large-scale thermal patterns, but not the meso- or small scale processes. Lake surface thermography, investigated in this study, has finer resolution compared to satellite images. Thermography at the meso-scale provides the ability to ground-truth satellite imagery over scales of one to several satellite image pixels. On the other hand, thermography data can be used as a control in schemes to upscale local measurements that account for the surface energy fluxes and vertical energy budget. Independently, since such data can be collected at high frequency, they can be also useful in capturing the surface signatures of meso-scale eddies and thus to quantify mixing processes. In the present study, we report results from a Balloon Launched Imaging and Monitoring Platform (BLIMP), which was developed in order to measure the LSWT at meso-scale. The BLIMP consists of a small balloon which is tethered to a boat and equipped with the thermal and RGB cameras, as well as other instrumentation for location and communication. Several deployments were carried out on Lake Geneva. In a typical deployment, the BLIMP is towed by a boat, and collects high frequency data from different heights (i.e., spatial resolutions) and locations. Simultaneous ground-truthing of the BLIMP data is achieved using an autonomous craft that collects a variety of data, including in situ surface/near surface temperatures, radiation and meteorological data in the area covered by the BLIMP images. With suitable scaling, our results show good consistency between in situ, BLIMP and concurrent satellite data. In addition, the BLIMP thermography reveals (hydrodynamically-driven) structures in the LSWT – an obvious example being mixing of river discharges.