Meteorology

It is important to adequately account for land surface processes within simulation models integrating hydrological, meteorological and climate processes. This book is an exclusive description of land surface hydrological processes, their observation and representation in models. The most valuable issues included in this book are process parameterization in the models and the assimilation of observations into models.
Land surface models are becoming more and more complex, although only a limited set of
their parameters has a physical basis and allows estimation based on observational data. One of the major difficulties in coupled climate and hydrological modelling is related to a proper parameterization of processes occurring at different spatial and temporal scales. For example, it is often necessary to describe the cumulative effect of small-scale processes at larger scales, e.g. soil infiltration as a pore-scale process and transpiration occurring at the leaf scale have to be represented in a mesoscale land surface model. Scaling is a very important issue in hydrological
modelling, because the relationships between soil moisture and évapotranspiration, and
soil moisture and streamflow are strongly nonlinear due to the interaction of numerous atmospheric, ecological, hydrological and anthropogenic forces influencing water fluxes and finally defining the spatial variability of water balance components and energy budget. Therefore discussion of scaling issues in several papers included in this book is very valuable.
The availability of spatially distributed data is usually the most critical factor in land surface modelling. Observations of atmospheric and land surface parameters include those that serve as input parameters to models and those that serve for validation purposes. The same parameter, e.g. temperature at a reference height, can serve as an input parameter, or as a validation parameter, depending on the model type. Recently, remote sensing has started to provide spatially distributed data, which are essential for further model development and validation, though in situ measurements still remain the main source of information. Two main sets of data are now provided globally by means of remote sensing: land surface temperature and different vegetation indices. The land surface temperature at diurnal resolution can be further used as an indicator of soil moisture dynamics and for partitioning between sensible and latent heat. The fraction of green, photosynthesizing vegetation can also be used as an indicator for the partitioning of energy fluxes.
The assimilation of observations into models, when model parameters and simulated
variables are adjusted in order to improve agreement between simulated and observed values, is of great importance. It should be distinguished from a direct model forcing based on observed data. However, the development of tools and methods for data assimilation is in its early stage. Examples of such methods are the use of satellite estimates of surface skin temperature to adjust simulated water content in soil, or the use of satellite derived vegetation indices for adjusting relevant model parameters or state variables describing vegetation cover. It is expected that the assimilation of observations into models will improve agreement between simulated and observed variables, and model performance in general.
This book consists of three sections, discussing three general areas of importance:
observations, modelling, and integration of observations and modelling.
Section 1 includes four papers on observations of atmospheric and land surface parameters of the hydrological cycle: from partitioning of net radiation and measuring water vapour to canopy microclimate and soil hydraulic properties. Kustas et al. discuss the use of satellite data and observations from weather stations for partitioning of net radiation into latent and sensible heat fluxes. They compare two modelling schemes accounting for soil and vegetation contributions to the mass and energy exchanges: a more rigorous one and a simple one, and demonstrate the adequacy of both for calculation of heat fluxes, where the second modelling scheme is more computationally efficient. Eichinger et al. analyse measurements of mean
water vapour profiles in the atmospheric surface layer, and demonstrate the existence of three sub-layers. The data support an assumption that the similarity function for water vapour is similar to that for temperature in the dynamic and dynamic-convective sub-layers. Katul et al investigate distributions and strengths of scalar sources and sinks of water vapour, carbon and heat in a canopy volume, considering forward and inverse methods based on foliage properties and measured mean scalar concentration distribution, respectively. Cuenca & Kelly investigate spatial and temporal variability of soil moisture and soil hydraulic properties (soil water
retention function and unsaturated hydraulic conductivity) from large-scale experimental data, which are crucial for the parameterization of SVAT schemes.
Section 2 includes five papers on advances in land surface modelling. Bastidas et al.
explore the use of observations (ground temperature and surface soil moisture) to parameterize the land surface model BATS by optimization in order to improve simulation of heat fluxes returned to the atmosphere. Duan et al. address issues concerning a priori parameterestimation procedures used in current land surface models, with particular emphasis on runoffrelated parameters. Koster et al. compare land surface water budgets generated by four atmospheric GCMs in relation to the precipitation and net radiation forcing simulated by each model. Chen et al. review progress in the coupling of advanced land surface models with atmospheric mesoscale models, considering the problems of soil moisture initialization, parameterization of surface vegetation and soil characteristics, and the sub-grid variability in topography, soil moisture, snow cover and vegetation characteristics. Maurer et al. compare water balance components from a mesoscale model with observed precipitation and simulated
évapotranspiration and surface energy fluxes from a macroscale hydrological model.
Section 3 includes five papers on integration of observation and modelling. Mohr et al.
explore the effect of sub-grid variability of soil moisture on the simulation of hydrological processes in a mesoscale watershed using a land surface model. Knorr & Lakshmi study assimilation of satellite-based data into a coupled land surface and vegetation model aimed at increased accuracy of simulated surface temperature, using two assimilation techniques.
Woods et al. discuss spatial variability in hydrology and sources of variability in streamflow for a temperate area in New Zealand, and compare results from a satellite-based model with field data. Lawford reviews the advances made in extensive field campaigns carried out under the ongoing GEWEX Continental-scale International Project (GCIP) in integrating observations and models and using them for improved understanding of various hydrometeorological processes. Piechota et al. investigate the hydrological implications of the El Nino Southern Oscillation (ENSO) for making long-range streamflow forecasts in eastern Australia and the western United States, where the effect of ENSO on hydrology is the strongest.
This book introduces a modern understanding of hydrology as an integration of
observations and modelling. This is necessary, because (a) observations without generalized description in the models are of a limited use and often not helpful for solving research and application problems, and (b) the models cannot be developed and validated without the observation base. The book will be of interest for modellers, experimentalists, and those working in the field of data assimilation.

Valentina Krysanova
Potsdam Institute for Climate Impact Research (PIK)
Potsdam, Germany