No abstract

Light transmission through ice and light conditions beneath ice have been investigated in the mild winter of the year 2000 in eight Estonian lakes and in one braclush water basin, Santala Bay in the Gulf of Finland. A new system designed for optical measurements beneath the ice was successfully tested. In the water body the vertical profiles of photosynthetically active radiation (PAR), temperature and oxygen were mapped. The concentrations of optically active substances (dissolved organic matter, chlorophyll a, particles) were estimated for water samples and meltwater of ice samples. The PAR band albedo was 0.28-0.76 and transmittance was 1-52% for the icelsnow cover. The light field below ice was much more diffuse than in open water conditions. Euphotic depth was 0.1-5.5 m. The amount of yellow substance in lake ice is very small in comparison with the lake water; lake ice may contain a lot of particles, but their source is atmospheric fallout rather than the water body. In some lakes a depletion of oxygen was observed. There were considerable differences between the fresh and brackish water ice (structure, stratigraphy, amount of impurities), which influenced the underwater light field.

No abstract

The main objective of the present study is to test various methods for describing the absorption spectra of coloured dissolved organic matter (CDOM) and to determine the numerical values of some optical parameters of CDOM in lakes with diverse water quality. First, the parameters of an exponential model in different spectral intervals were determined. In addition, the suitability of some other models for the approximation of CDOM spectra was estimated. Specific absorption coefficients of CDOM were calculated from the absorption coefficients and dissolved organic carbon (DOC) concentrations. The experimental initial data were differences between spectral attenuation coefficients of filtered and distilled water. Two datasets were used: 1) for 13 Estonian and 7 Finnish lakes (altogether 404 spectra between 350 and 700 nm) measured by the Estonian Marine Institute (EMI); 2) for 10 Finnish lakes (73 spectra) measured by the Finnish Environment Institute (FEI). The spectra of CDOM absorption coefficients (aCDOM) were calculated from experimental data taking into account the correction due to scattering properties of colloids in the filtered water. The total content of CDOM in natural waters of Estonian and Finnish lakes was expressed by means of aCDOM at the wavelength of 380 nm. It varied significantly, from 0.71 to 19.5 m−1, the mean value (of all the investigated lakes) being around 6.6 m−1. Slopes of the exponential approximation varied widely, from 0.006 to 0.03 nm−1. Averaged over all lakes values of slope for the interval 380-500 nm obtained from the EMI dataset are close to those obtained from the FEI dataset: from 0.014 nm−1 (without correction) to 0.016-0.017 nm-1 (with different types of correction). These results are in good correspondence with most published data. Attempts to describe the spectra in the region of 350-700 nm by means of hyperexponential functions (∽ exp(-αλη)) show that: (1) η < 1 (in the case of traditional exponential approximation η = 1); (2) a promising idea is to seek the best fit only for wavelengths λ > λ1, where λ1 will be chosen taking into account the real shape of aCDOM spectra. The mean value of the specific absorption coefficient (a*CDOM) at the wavelength 380 nm obtained in this study (0.44 L mg−1 m−1) is close to the values published in the literature, if we assume that a*CDOM (380) is calculated using the data of dissolved organic matter (DOM). The optically non-active fraction of DOM in our study was high and therefore a*CDOM (380) was considerably higher (1.01 L mg−1 m−1) than a*CDOM (380). The results of the present work could be used in the modeling of underwater light field as well as in the interpretation of radiation measurements and optical remote sensing results.

The method suggested earlier for estimating the spectra of diffuse attenuation coefficient of light in the water bodies relying on the beam attenuation coefficient measured from water samples, was improved and applied to different types of lakes. Measurement data obtained in 1994-95 and 1997-98 for 18 Estonian and Finnish lakes were used. The spectra of two characteristics were available for our investigations: 1) beam attenuation coefficient estimated from water samples in the laboratory with a spectrophotometer Hitachi U1000; 2) vertical irradiance (diffuse) attenuation coefficient measured in situ with an underwater spectroradiometer LI 1800UW. A total of 70 spectra were considered. Relying on these data the parameters of our earlier model were changed. The criterion of the efficiency of the new version of our model is the coincidence of the spectra of diffuse attenuation coefficient derived from Hitachi U1000 data (Kdc) with those obtained by underwater irradiance measurements (Kdm). Correlation analysis of the model's results gave the relationship Kdm=1.0023Kdc with correlation coefficient 0.961. The respective values of mean relative difference and standard deviation were 5.4% and 0.55 m−1. This method may be useful in conditions where in situ measuring of underwater irradiance spectra cannot be performed because of weather conditions. As the measurement of the underwater radiation field is often a complicated and expensive procedure, our numerical method may be useful for estimating the underwater light climate.