Accuracy of the Coupled Carbon Model Hadcm3lc at Predicting Future Carbon Dioxide and TemperatureEssay Preview: Accuracy of the Coupled Carbon Model Hadcm3lc at Predicting Future Carbon Dioxide and TemperatureReport this essayAccuracy of the coupled carbon model HadCM3LC at predicting future carbon dioxide and temperatureIn the last 200 hundred years global atmospheric carbon dioxide levels have increased from 280 ppm (Pearman et al 1986) to approximately 390 ppm today (Mauna Loa, 2011) although not the only GHG released by anthropogenic sources, there is a strong correlation between the rise in CO2 and the rise in temperature of 0.7 K in the last 100 years (NASA, 2011). This change has been driven by several factors most importantly the combustion of fossil fuels. With the increasing use of fossil fuels as a source of energy and the reduction in the Earths potential to remove this air-borne GHG; there has been increasing concern as to the effect this will have on the planet. This has led to the formation of the International Panel for Climate Change (IPCC) and the creation of Global Circulation Models (GCM) to attempt to predict the future climate and temperature to inform politicians and policy makers. The models are often a simplified simulation and attempt to calculate the rise in CO2 and the corresponding increase in temperature along with the effect this will have on the reservoirs in the future. They use certain parameters usually concentrating on fluxes of CO2 but due to lack of knowledge they can often neglect important factors. Feedback systems occur when the reservoirs capacity changes and can dramatically affect the models outcome and are often difficult to accurately model due to a lack of knowledge.
One particular model HadCM3LC; outlined in nature 10 years ago (Cox et al, 2000), concentrated on soil in the future as both a sink and a source. In this article the focus of the model will be the possible under-estimation of carbon uptake by terrestrial and oceanic systems, the positive feedback of carbon into the atmosphere and the lack of accuracy of the model.
The model ran from 1860 to 2100 integrating 3 models to simulate the fluxes between the four main pools of carbon. It combined the third Hadley centre ocean-atmosphere model (HadCM3); an ocean-carbon cycle model (HadOCC) and a global vegetation model (TRIFFID). The HadCM3 is a widely used model with a high resolution and good match between atmosphere and air components. It requires no flux adjustments and can be run for large periods of time with little surface climate drift. The HadOCC model is concerned with the inorganic diffusion of carbon and flux using the biological pump from the atmosphere into the oceans and the disposition of carbon within the deep ocean as carbonate; it is a dynamic model and requires flux updates to maintain accuracy. The TRIFFID model is based on the carbon fluxes from biosphere and soil into the atmosphere; it looks at 5 different types of vegetation, using the stomatal-conductance based on models from C3 and C4 plants. It dynamically updates the vegetation land cover every 10 days as the carbon dioxide and climate force the evolution of the landscape changing the uptake and release of CO2. It takes into consideration competition, agricultural regions, litter fall and large-scale disturbances; transferring them to the soil pool and converting it back to atmospheric CO2 via microorganism decomposition with a doubling of soil respiration for every increase of 10K.
The HadCM3LC model had a predicted temperature rise from present day levels of between 5.5 – 8K by 2100 and a CO2 concentration ranging from 730 – 980 ppmv in the simulations based on a business as usual scenario leading into runaway global warming. Unlike most of the previous models, this one incorporates an atmospheric-land model with a more data and complex formulae. It overestimated the atmospheric CO2 and temperature for the year 2000 by 10 years placing it almost 20 ppm higher than the recorded value.
In global carbon cycling the 4 dominate pools with the largest fluxes between them are the hydrosphere, atmosphere, biosphere and soil; currently the estimated fluxes (figure 1) are relatively easy to calculate due to in-field experimentation and measurements; these fluxes are largely dependent upon multiple factors which are harder to successfully quantify and are therefore uncertain in the future.
What is known is that the carbon cycle is driven by photosynthesis; the inputs into the biosphere are counteracted by respiration, in the natural environment without intervention these two processes are found in equilibrium. However due to the imbalance caused by anthropogenic release of carbon dioxide into the atmosphere, the equilibrium has changed in an attempt to remove CO2 leading to a net increase in the sequestered carbon in the soil, ocean and biosphere causing them to become sinks for carbon. The main anthropogenic sources of CO2 are fossil fuel combustion, cement production and change in land use. The largest source is the combustion of fossil fuels for energy with an increasing population and improving living standard, this is only set to increase unless other renewable energy sources or less GHG emitting ones become more widely used. Currently of the anthropogenic carbon emitted, 50% is absorbed by the land and oceans; although the ocean and land are removing carbon from the atmosphere, the anthropogenic input of CO2 is greater than the removal leading to a net accumulation in the atmosphere of 4.5 billion metric tonnes of carbon annually as of 2000 (Marland et al 2007). There is also a natural lag in the uptake of carbon dioxide by the different pools particularly the ocean as it can take millennium for the carbon within it to cycle. Due to its properties as both a GHG and natural acid, the increase can alter the environment which can dramatically perturb the fluxes and reservoirs; when they get perturbed enough, a threshold is reached resulting in a shift from carbon sink to carbon source further increasing carbon release into the atmosphere and creating a positive feedback possibly leading to runaway global warming although this is deemed unlikely (Houghton, 2005)
Figure 1: Carbon cycling model of major fluxes taken from NASA websiteThe processes involved in the HadCM3 model are driven by predominantly inorganic and physical processes which are well understood; mean temperature and climate is relatively easy to model, this model even incorporates ENSO events and the subsequent La Ninas. The flux of CO2 is a function of the partial pressure of CO2 in the atmosphere and surface ocean (Freidling et al, 2006) however an increase in temperature will likely reduce the carbon dioxide uptake due to the slowing of the thermohaline circulation slowing the removal of carbon dioxide into the deep ocean (Sarmiento, 1996) increasing the partial pressure in the surface waters although the future extent of this is not certain.
In Figure 1, there is a significant change in the spatial distribution of the CO2 release due to decreasing sea surface temperatures, this is related to the slow removal of CO2 from the deep ocean (Freidling et al, 2006) and increased warming of the northern and southern oceans caused by the cooling of the Mediterranean Sea. In contrast, the increase in surface pressure due to increased CO2 has led to the warming of the North Atlantic (Freidling et al, 2006). At mid-winter last year (14 March 2015), during which the ocean circulation slowed at 20%–20% and ocean acidity was at about 15.8%, the rate of warming in 2015 was around 2.5 °C per year compared to previous years, this was a decrease of 0.9% but it was a significant fall compared to previous years. We can see, when we looked across the last few summers to see how the effect of the ice sheet was affected by warming, that the reduction in the global circulation rate, which is the measure of the temperature rise, was a major factor in the increased CO2 uptake (Table 1). During the last ice sheet reduction which was in a cooling phase, the total change in the sea surface temperatures was 10.3 ± 0.9 °C relative to the normal and 15.8 ± 0.8 °C at the same time since the year 2000 (Figure 1B). Since the warming of the South Atlantic was almost all caused by an increase in Antarctic sea ice, the change was probably not as big or intense as expected.
Data used to make model and model simulation of climate change can be found in Sarmiento et al (1996), who also show a shift in model time since 1992 (see Sarmiento et al (1996: 2310.7) for a global average). The authors note that since 1990 the global average is higher than that of prior centuries (Sarmiento et al (1996: 2310.7). Therefore, the time taken to account for this global trend of warming was around 2310.7 to 2318.9 years ago before the cooling phase of the ice sheet was achieved and the rate of warming was almost exactly the same in 1995. This is consistent with a shift in the model to account for the rate of warming in 2013 when the surface warming rate was only 3.65–4.35 °C in the spring and is now 3.3–4.25 °C in the spring time.
As expected in the model, however, there was a notable increase in heat content after the ice sheet thaw (Sarmiento et al (1996), 2312.7 = 6°C in 2015). This was due primarily to an increased mass of solar irradiance and the cooling of the South Arctic ice sheet due to an increase in the mass of cold convection. This was expected due to the slowing of the cooling of the South Atlantic (Sarmiento et al (1996),
Soil is the most contentious source of debate with evidence only in the last