By: Natalie Douglas
Photosynthesis is a biological process that removes carbon (in the form of carbon dioxide) from the atmosphere and is therefore a key process in determining the amount of climate change. So, how do we measure it so that we can use it in climate modelling? The answer is, in short, we don’t.
Photosynthesis is the process by which green plants absorb carbon dioxide (CO2) and water and use sunlight to synthesise the nutrients required to sustain themselves. Since plants absorb CO2, and generate oxygen as a by-product, the rate at which they do so is a fundamental atmospheric process and plays a critical role in climate change. In climate science, we refer to this rate as Gross Primary Productivity or GPP. It is typically measured in kgm-2s-1 which is kilograms of carbon per square metre per second. But why do we need to know this? Climate models, also known as General Circulation Models (GCMs), divide the Earth’s surface into three-dimensional grid cells that typically have a horizontal spatial resolution of 100km by 150km at mid-latitudes. Using supercomputers, a set of mathematical equations that govern ocean, atmosphere and land processes are solved and the results passed between neighbouring cells to model the exchange of matter (such as carbon) and energy over time . Fundamental to their solution are what we call initial conditions (the state of the climate variables at the start of the model run) and boundary conditions (the state of the required variables at the land surface). Due to the sheer complexity of the processes involved, we require another type of model to provide the latter – land surface models.
It isn’t possible to simply measure photosynthesis; an instrument that quantifies the amount of carbon a plant absorbs from the atmosphere doesn’t actually exist. There are, however, eddy covariance towers that are capable of measuring carbon fluxes at a given location. The locations of these towers are sparse but do provide good estimates for the fluxes at a given location. If it were possible to provide eddy covariance fluxes at all grid locations, say at their centres, this would suffice for a GCM, but since this is completely infeasible, we have the need for land surface models. The Joint UK Land Environment Simulator, or JULES, is the UK’s land surface component of the Met Office’s Unified Model used for both weather and climate applications , . Before JULES can model carbon fluxes it requires an ensemble of information including surface type, particulars on weather and soil, model parameter values, and its own initial conditions. A module within JULES is then able to calculate the carbon uptake at the surface boundary of the grid cell based on the number of leaves within the grid, the differences in CO2 concentrations between the leaf surface and the atmosphere, and several limiting factors such as light availability and soil moisture . Figure 1 shows a representation of the monthly average of GPP for June 2017 as modelled by JULES.
Earth Observation (EO) plays a crucial role in developing current climate research. There are numerous satellites in space capturing various characteristics of the Earth’s surface at regular intervals and at different spatial resolutions. Scientists cleverly transform this data, using mathematics, into the required variables. For example, NASA’s MODIS (MODerate resolution Imaging Spectroradiometer) satellites measure light in various wavelengths and a team of scientists convert this data into an 8-day GPP product . Neither models nor EO data are 100% accurate when it comes to determining the variables required for land surface and climate change models and so much of today’s research focuses in combining both sets of information in a method called Data Assimilation (DA). Using mathematics again, DA methods take both model estimates and observations as well as information regarding their uncertainty to find an optimal guess of the ‘true’ state of the variables. These methods allow us to get a better picture of the current and future states of our planet.