 |
Carbon Assimilation and Modelling of the European Land Surfaces |
 |
 |
|
Where are the current carbon sources and sinks located on the land and how do European sinks compare with other large continental areas? This question is most often addressed using atmospheric inversion techniques, which provide estimates of net surface-atmosphere exchanges of CO2 using an optimisation process. The atmospheric CO2 concentrations from a number of measurement sites constrain the source/sink areas in the presence of estimated atmospheric transport and known emissions. Since the typical atmospheric gradient of CO2 from North to South pole is only 3 ppm, the accuracy of the network is critical. Bousquet et al. (1999) estimate a sink of 0.5+/-0.6 Gt C/yr for Northern America, a sink of 0.3 +/-0.8Gt C/yr for Europe and 1.3 Gt C/yr for Siberia. By contrast, Fan et al (1998) conclude that a Northern hemisphere sink of 1.7 Gt C/yr may exist entirely over North America. The difference between these two studies is partly caused by the inclusion of more measurement sites in the Bousquet et al study. Although inversion studies can provide useful estimates of the net land-atmosphere CO2 exchange they are unable to separate this into contributions from accidental effects and land management. A key motivation for CAMELS is to produce a system which can use the atmospheric CO2 measurements along with remote-sensing products to constrain mechanistically-based models of the land carbon cycle, and which can therefore be used to isolate the causes of sources and sinks. In the table below we present the current best estimates of European sink strength obtained by various methods:
All methods indicate a land carbon sink but there is a factor 5 between the highest and lowest estimates. A likely reason for these discrepencies is that the various methods use different scaling techniques to transform information obtained at one spatial and temporal scale to another. This motivates the CAMELS first Objective : to provide a consistent estimate of the European land carbon sink by making intelligent use of all of the existing data-sources, which provide both small-scale constraints (e.g. flux measurements, inventory estimates) and large-scale constraints (e.g. atmospheric CO2 measurements, remote sensing products). Why do these sources and sinks exist, i.e. what are the relative contributions of CO2 fertilisation, nitrogen deposition, climate variability, land management and land-use change? This is a key scientific question because it relates to the causes of changes in net land carbon storage. It also has great policy relevance since the Kyoto protocol only allows nations to claim carbon credits for carbon accumulation due to direct land-use management. Contributions to the land carbon sink due to other accidental environmental changes such as CO2 fertilisation, nitrogen deposition and climate variability need to be quantified and excluded. This itself is a far from a trivial matter since the cause of the land carbon sink in each location is not known. Small scale manipulative experiments broadly show increased photosynthesis (and water-use efficiency) under enhanced CO2, but the applicability of this to real resource-limited ecosystems is uncertain. For example, a recent analysis of tree-rings in 20,000 forest plots in the eastern USA showed no evidence of increased growth rates due to rising CO2 (Caspersen et al., 2000). Some have suggested that the Northern hemisphere sink is instead due to nitrogen deposition which acts to enhance growth in nitrogen-limited systems. However, recent manipulative experiments also cast doubt on this mechanism (Nadelhoffer et al., 1999). The sensitivity of the terrestrial carbon cycle to climate change and climate variability (Dai and Fung, 1995, Bousquet et al., 1999) make it especially difficult to isolate the causes of land carbon uptake from measurement alone. This motivates the CAMELS second Objective : to elucidate the mechanisms responsible for the contemporary land carbon sink (in Europe and elsewhere), isolating the contribution due to direct land management. How could we make optimal use of existing data sources and the latest models to produce operational estimates of the European land carbon sink? As outlined above, there are a number of independent data sources which provide valuable information on the terrestrial carbon cycle. These include flux-measurements (e.g. from the CarboEurope cluster of projects), forestry carbon inventories, atmospheric CO2 concentration measurements (as used in inversion studies) and satellite products (e.g. the fraction of absorbed photosynthetically active radiation, fAPAR, which controls the rate of photosynthesis). At the same time terrestrial ecosystem models (TEMs) have been developed based on the physiological and ecological processes underlying land-atmosphere CO2 exchange. Observational estimates of the carbon sink differ partly because the different data sources are indicative of different time and space scales, while TEM estimates differ largely because they contain internal parameters and stores (e.g. leaf nitrogen, leaf area index, respiring soil carbon) which have not been properly constrained by the observations. To produce a best estimate of carbon uptake we need to make use of the all of the constraints implied by the different data sources, as well as the physiological and ecological constraints embodied in the TEMs. In otherwords, we need to simultaneously use the observations to constrain the internal parameters of the TEMs, whilst using the TEMs to interpolate the observations to produce useful large-scale estimates of the carbon sink and its causes. This is essentially a Data Assimilation (DA) problem, requiring a system similar to those used to initialise weather forecast models. This motivates CAMELS third Objective : to develop a carbon cycle data assimilation system (CCDAS) which optimally combines data and models to produce operational estimates of the European carbon sink and its constituent contributions ent contributions |
