Projected changes in meteorological drought over East Africa inferred from bias-adjusted CMIP6 models

Natural Hazards - The ongoing global warming has caused unprecedented changes in the climate system, leading to an increase in the intensity and frequency of weather and climate extremes. This...     

Energy fluxes and melt rate of a seasonal snow cover in the Moroccan High Atlas

ABSTRACT

In this study, we used an energy balance model and two simple methods based on readily available data to identify the processes driving the point-scale energy and mass balance of the snowpack. Data were provided from an experimental site located at 3200 m. All models were evaluated by comparing observed and modelled snow water equivalents. Performances are variable from one season to the next and the energy balance model gives better results (mean of root mean square error, RMSE = 25 mm and r² = 0.90) than the two simplified approaches (mean of RMSE = 54 mm and r² = 0.70). There are significant amounts of snow sublimation but they are highly variable from season to season, depending on wind conditions (between 7 and 20% of the total). While the main source of energy for melting is net radiation, the amount of heat brought by sensible heat flux is significant for two of the most windy snow seasons.

Editor Z.W. Kundzewicz Associate editor not assigned     

Effective parameters of surface energy balance in heterogeneous landscape

This paper addresses the problem of estimating surface fluxes at large scale over heterogeneous terrain, and the corresponding determination of effective surface parameters. Two kinds of formulation are used to calculate the fluxes of sensible and latent heat: the basic diffusion equations (Ohm's law type) and the Penman-Monteith equations. The strategy explored is based upon the principle of flux conservation, which stipulates that the average flux over a large area is simply the area-weighted mean of the contributions from the different patches making up the area. We show that the application of this strategy leads to different averaging schemes for the surface parameters, depending on the type of flux (latent heat, sensible heat) and on the type of formulation used to express the flux. It appears that the effective value of a given parameter must be appraised for each individual application, because it is not unique, but differs according to the magnitude being conserved and the equation used to express this magnitude. Numerical simulations are carried out to test over contrasted areas the aggregation procedures obtained. The areal fluxes estimated from these effective parameters, together with the areal fluxes calculated by means of a simple areal averaging of the parameters, are compared to the true average fluxes, calculated as area-weighted means of the elementary fluxes. The aggregation procedures obtained prove to be much more accurate for estimating areal fluxes and for closing the energy balance equation than those based upon simple areal averaging of the parameters.     

Understanding hydrological processes with scarce data in a mountain environment

Performance of process‐based hydrological models is usually assessed through comparison between simulated and measured streamflow. Although necessary, this analysis is not sufficient to estimate the quality and realism of the modelling since streamflow integrates all processes of the water cycle, including intermediate production or redistribution processes such as snowmelt or groundwater flow. Assessing the performance of hydrological models in simulating accurately intermediate processes is often difficult and requires heavy experimental investments. In this study, conceptual hydrological modelling (using SWAT) of a semi‐arid mountainous watershed in the High Atlas in Morocco is attempted. Our objective is to analyse whether good intermediate processes simulation is reached when global‐satisfying streamflow simulation is possible. First, parameters presenting intercorrelation issues are identified: from the soil, the groundwater and, to a lesser extent, from the snow. Second, methodologies are developed to retrieve information from accessible intermediate hydrological processes. A geochemical method is used to quantify the contribution of a superficial and a deep reservoir to streamflow. It is shown that, for this specific process, the model formalism is not adapted to our study area and thus leads to poor simulation results. A remote‐sensing methodology is proposed to retrieve the snow surfaces. Comparison with the simulation shows that this process can be satisfyingly simulated by the model. The multidisciplinary approach adopted in this study, although supported by the hydrological community, is still uncommon. Copyright © 2007 John Wiley & Sons, Ltd.     

Sensible Heat Flux-Radiometric Surface Temperature Relationship Over Sparse Vegetation: Parameterizing B-1

In the bulk formulation of vegetation-atmosphere transfer, theparameter B^-1 is needed for evaluating the sensible heatflux from radiometric surface temperature. An excess resistance,expressed as a function of B^-1 (r_rB^-1/u_*, whereu_* is the friction velocity), shouldbe added to the aerodynamic resistance calculated between thelevel of apparent sink of momentum and the reference height.Over sparse vegetation, B^-1 (and consequently the excessresistance) can be very large and variable. A one-dimensionaltwo-layer model of the canopy-atmosphere interaction is used toinvestigate the behaviour of this fitting parameter and to derive anoperational parameterization in terms of structural and viewingcharacteristics. Besides canopy structural characteristics andradiometer viewing angle, input variables include weather data,stomatal and substrate resistances. B^-1 varies with almostall the input variables; however, the leaf area index (LAI) andthe view angle of the radiometer appear as the most significantfactors of variation. Using a set of weather data and componentresistances randomly generated between fixed limits, `average'curves representing B^-1 as a function of LAI for differentview angles are inferred from the model and polynomial expressionsare fitted to the simulated curves. This set of parameterizations isobtained from ranges of input data wide enough to be representativeof a large variety of experimental conditions. It is successfully testedagainst other parameterizations, using both simulated data andmeasurements made over contrasted surfaces in Niger, France andCalifornia. As the formulations proposed depend on the range ofvalues prescribed in the simulation process for each input data, theyare modifiable and adjustable to any experimental conditions.