Scientific publications

An approach to assessing the radiation regime under forest canopy is proposed, based on the integration of ERA5-Land global reanalysis data and field measurements of the leaf area index (LAI). The attenuation of incoming solar radiation was described by an exponential relationship derived from the Beer Lambert-Bouguer law, using the canopy extinction coefficient (k) calculated from ground-based observations. The extinction coefficient was determined for each “sample plot – observation date” combination by comparing measured under canopy radiation with representative daily radiation values above the canopy obtained from ERA5-Land data and adjusted to the level of ground measurements. For each experimental treatment (control and different doses of mineral fertilizers), the median value of k was used in the calculations, which reduced the influence of extreme values and provided a robust characterization of the light regime. Model validation was performed using data from the SAMU automatic weather station installed in the control sample plot for daytime hours. A satisfactory agreement in the temporal dynamics of under-canopy radiation was found (r = 0.73; RMSE = 14.18 W/m²). The low values of the coefficient of determination are associated with the high variability of hourly illumination under the canopy caused by variable cloudiness and local fluctuations in the proportion of direct radiation and do not imply model inadequacy when assessing integral characteristics of the radiation regime. Scenario analysis showed that the calculated under-canopy radiation values strongly depended on the magnitude of the extinction coefficient. Differences between treatments within the scenario analysis reflect the sensitivity of the model to changes in the extinction coefficient and are not considered as independent validation of sample plots with different concentrations of mineral fertilizers. The proposed approach provides a physically grounded and robust description of the undercanopy radiation regime and can be applied to assess the light regime and productivity potential of forest ecosystems where field data availability is limited.