Constraining the Sheffield dynamic global vegetation model using stream‐flow measurements in the United Kingdom

Abstract

The biospheric water and carbon cycles are intimately coupled, so simulating carbon fluxes by vegetation also requires modelling of the water fluxes, with each component influencing the other. Observations of river streamflow integrate information at the catchment scale and are widely available over a long period; they therefore provide an important source of information for validating or calibrating vegetation models. In this paper, we analyse the performance of the Sheffield dynamic global vegetation model (SDGVM) for predicting river streamflow and quantifying how this information helps to constrain carbon flux predictions.The SDGVM is run for 29 large catchments in the United Kingdom. Annual streamflow estimates are compared with long time-series observations. In 23 out of the 29 catchments, the bias between model and observations is less than 50 mm, equivalent to less than 10% of precipitation. In the remaining catchments, larger errors are because of combinations of unpredictable causes, in particular various human activities and measurement issues and, in two cases, unidentified causes.In one of the catchments, we assess to what extent a knowledge of annual streamflow can constrain model parameters and in turn constrain estimates of gross primary production (GPP). For this purpose, we assume the model parameters are uncertain and constrain them by the streamflow observations using the generalized likelihood uncertainty estimation method. Comparing the probability density function of GPP with and without constraint shows that streamflow effectively constrains GPP, mainly by setting a low probability to GPP values below about 1100 g C À1 m 2 yr À1 . In other words, streamflow observations allow the rejection of low values of GPP, so that the potential range of possible GPP values is almost halved.

The impact of alternative trait‐scaling hypotheses for the maximum photosynthetic carboxylation rate (Vcmax) on global gross primary production

“…The SDGVM was developed as a daily timestep, global biogeography and eco-physiology model Woodward & Lomas, 2004) to predict the primary biomes of Earth and their associated fluxes of C and water in response to global change. SDGVM has been described and extensively evaluated at site and global scales Cramer et al, 2001;Woodward & Lomas, 2004;Picard et al, 2005;Sitch et al, 2008;Beer et al, 2010;De Kauwe et al, 2013Friend et al, 2014;Walker et al, 2014b;Zaehle et al, 2014), so here we provide a brief description of the model and the process simulation methods relevant to this paper.…”

Evaluating the performance of the MODIS Leaf Area Index (LAI) product over a Mediterranean dryland planted forest

“…However, such low spatial resolution data, for which each pixel may be made up of several land cover types, might be of limited utility for semi-arid forested areas that are mostly sparse and heterogeneous (Sprintsin et al 2007), therefore requiring detailed and careful validation by regionor site-specific measurements. LAI is a key parameter in land process models, terrestrial carbon modelling, dynamic vegetation models, large-scale carbon budgets and large scale climate models (Prince et al 2001, Picard et al 2005. Due to its central role in the dissipation of radiant energy input its erroneous assessment will affect the results of the models implemented.…”

Efficient Emulation of Radiative Transfer Codes Using Gaussian Processes and Application to Land Surface Parameter Inferences

“…A critical property of the emulator is that its execution is much faster than running the full model, and for this reason they have been widely used in statistics [18][19][20][21]. We focus our attention in Gaussian process (GP) emulators, which have been widely used in a number of fields (e.g., the climate and vegetation modelling communities have used emulators for some time [22][23][24][25]). In Earth Observation, GPs have been used as fast surrogates to computationally expensive models in [26] for sensitivity analysis.…”