Determining On-site Variability of CO2 and H2O Fluxes Using Footprint and LiDAR Data

Understanding the exchange of energy and gases between the earth's surface and the lower atmosphere is essential for answering many questions related to, e.g., the global carbon budget, ecosystem functioning, air pollution mitigation, greenhouse gas emissions, weather forecasting, and projections of climate change. However, uncertainties in carbon dioxide (CO2) and water vapour (H2O) budgets limit our ability to reproduce and project these exchange processes. Exchange processes are usually analysed based on micrometeorological measurements from tall flux towers, thought to be representative of large area averages. A limitation of this approach is that the actual source areas of these fluxes are not always known and that the impact of land-surface heterogeneity (at small or large scale) on the fluxes is not yet completely understood. The micrometeorological measurements of the major carbon flux networks around the world, such as Ameriflux, Canadian Carbon Program, CarboEurope (in which the UK plays a prominent role) and Oz-Net, are essential to validate global estimates of CO2 sources and sinks, to develop and validate land surface models and to understand the sensitivity of CO2 fluxes under changing climate conditions. Unfortunately, flux tower measurements currently suffer from substantial uncertainty, which is primarily due to the indeterminate relationship of fluxes and their source areas; at present our current understanding can explain 60-80% of the variance of the fluxes. The overall goal of this project is to incorporate information on topography and structure of vegetation (tree height, canopy depth, and foliage density) in footprint estimates and thereby substantially reducing the potential errors in the calculation of the CO2 and H2O budgets. The selected forested sites consist of the very few long-term flux stations within the boreal forest biomes and represent the three dominant species of the boreal forest (jack pine, black spruce, aspen). The combination of these three forest stands will provide data that is sufficiently representative to allow for upscaling to the boreal forest biome scale. The boreal forest constitutes the world's second largest forested biome (after the tropical forest) and plays an important role in regulating the climate of the northern hemisphere and in the global carbon cycle. The footprint model developed by the PI and widely used by the international community will be applied on long-term data sets to estimate the size and location of the area containing the sources or sinks (footprint) of CO2 and H2O fluxes measured at the three sites. The footprints will account for, and depend on, atmospheric conditions, such as wind speed and boundary layer stability, and surface characteristics, e.g. roughness. This footprint model is one of very few models that are valid over a huge range of stratifications and receptor heights. The major improvement of the footprint model will incorporate three-dimensional information on the structure of the forest, derived from airborne scanning LiDAR measurements, leading to exceptionally detailed high temporal resolution source information. Unlike data from passive sensors, the unique LiDAR data set provides information from within the tree canopy. The results will be used to analyse impacts of structure of vegetation and small changes in elevation on the net CO2 and H2O fluxes. The new understanding will assist future studies of upscaling from flux towers to the spatially heterogeneous boreal forest landscape and will reduce the uncertainty in the modelling of carbon budgets at local, regional and continental scale. It will lead to a greater understanding of local structural effects on carbon sources and sinks and thus the dynamics of carbon cycling and to major improvements of the description of these exchange processes in land surface models. Hence, the new insights will help reducing uncertainty in projections of climate change.