COMPUTER LAB 8
EARTH SYSTEMS SCIENCE I
PG250, Fall 2010
HunterCollege
Lab 8. The Carbon Cycle
In this lab we do experiments with a global carbon cycle model using STELLA © software. Part 1 of the lab addresses carbon cycle changes that occurred between 1860 and 2000; part 2 addresses potential changes that might occur between the years 2000 and 2100.
Part 1. GLOBAL CARBON CYCLE: 1860-2000
(file name = pg250_lab8_carbon_cycle_1860_2000.STM)
Some preliminary information you will need to know
The global carbon cycle model that we are using in this lab was adapted from the model described in Chameides and Perdue, (Biogeochemical Cycles: A Computer-Interactive Study of Earth System Science and Global Change, Oxford University Press 1997).The fluxes between the stocks represent different processes in the carbon cycle, including biotic and non-biotic processes. All the fluxes are calculated according to rate constants, so that the flux out of each stock depends upon the amount of carbon in a stock. Figure 1 shows a generic STELLA model of such a flux.
Figure 1. conceptual model with a flux determined by a rate constant
The mathematical equation for the flux shown in figure 1 is described in equation 1.
Flux-out = stock * rate constant(1)
This means that the flux-out is determined by the magnitude of the stock multiplied by a constant number. For example, if the stock was the amount of water in a lake, and the flux-out was the flow of water out of the lake through a stream, a rate constant of 0.05/day means that 5% of the water in the lake flows out of the stream each day. Rate constants used in this fashion always have units (1/time), and time can be expressed in seconds, days, or any other time unit that makes sense. In the carbon cycle model used here, the time units are years, and rate constants have units (1/year). The units of the stocks are Gt-C (gigatons of carbon), and therefore the units of the fluxes are Gt-C/yr. A Gt is a billion (109)metric tons, and a metric ton is 1000 kg.To convert Gt to standard metric units use equation 2.
1 Gt-C= 1012 kg-C = 1015 g-C = 1 Pg-C (1 petagram)(2)
Atmospheric carbon dioxide concentrations are typically described in units ppm(v), or parts per million (by volume). For the atmospheric stock only, one can calculate the atmospheric carbon dioxide concentration (ppm(v)) from the atmospheric carbon stock (Gt-C) using equation 3.
1 ppm(v) CO2 = 2.13Gt-C(3)
The model
To familiarize yourself with this carbon cycle model, start with the terrestrial biosphere (figure 2). There are two stocks connected to the atmosphere stock: the living terrestrial biosphere and the dead terrestrial biosphere. The fluxes between the living biosphere and the atmosphere are gross productivity (by photosynthesis), and respiration. The flux from the living to the dead biosphere is called litterfall, and the flux from the dead biosphere to the atmosphere is called terrestrial decay. In addition to these natural processes, I added one more flux: def which represents global deforestation. Global deforestation is modeled as an extra flux of carbon from the living terrestrial biosphere to the atmosphere.The beta switch is related to the fact that the rate of productivity by the terrestrial biosphere is expected to increase as atmospheric CO2 levels increase. This is known as the CO2 fertilization effect. The magnitude of this effect is not well known, but we adapt the estimate used by Chameides and Perdue.
Figure 2. the terrestrial biosphere with deforestation and the CO2 fertilization effect
We also have exchanges of carbon between the atmosphere and the ocean surface (surf ocean) (figure 3). The flux from the atmosphere to the ocean is called atm2ocn, and the flux from the ocean to the atmosphere is called ocn2atm. Note the additional STELLA converter named revelle switch. This is named after a famous oceanographer named Roger Revelle, who found that the carbon flux from the atmosphere to the ocean is not a determined by a constant rate factor (in our model named a2orate), but that the rate itself is determined by the atmospheric concentration. This is similar to the beta switch mentioned earlier. Notice that these two rate factors have connectors drawn to them from stocks, while the other rate factors have no such connectors.
Figure 3. atmosphere – ocean carbon exchanges
The surface ocean is connected to the deep ocean by both physical and biological mechanisms (figure 4). The physical mechanisms involve ocean circulation, including upwelling and downwelling, referred to in the model as upwell and downwell. The biological mechanism involves the productivity of the marine organisms in the surface water (gross ocn prod), the rapid decay of some of those organisms which returns the carbon back into the surface water stock (ocn decay), as well as the sinking of those organisms which do not decay rapidly, but may eventually decayin the deep ocean waters (net ocn prod).
Figure 4. surface - deep ocean carbon exchanges
The much slower processes of sedimentation and weathering are also included in the oceanic portion of this carbon cycle model (figure 5). The fact that these processes are slower is reflected in the model by their very small rate constants compared to the rate constants for other fluxes. These processes are actually important only over very long time scales (thousands to millions of years), and are not really required when studying the carbon cycle over timescales of only 100-200 years such as in this lab. However, there is a reason they are included in this lab, which is discussed in the next paragraph.
Figure 5. surface - deep ocean carbon exchanges including sedimentation
The full carbon cycle model used in this lab is shown in figure 6. The only items in the full model not discussed previously are two fluxes: weathering ocn and ff. Weathering ocn, the weathering of organic ocean sediments back to the atmosphere, is another slow process that is only important over geological time scales. It occurs due to plate tectonics, when the ocean floor is eventually uplifted and exposed to the atmosphere to allow sediments that were previously laid down on the ocean floor to be weathered.Finally, we get to the reason why the slower processes of sedimentation and weathering are included in this model: the burning of fossil fuels, represented here by the ff flux. When we burn fossil fuels we are releasing energy that was stored in organic material by photosynthesis millions of years ago. At the same time, we are also releasing the carbon that was originally in the atmosphere, removed by photosynthesis millions of years ago, and has been stored underground ever since. In a sense, we are short-circuiting the very slow process of sedimentation and weathering. Thus, the reason we include these slow processes in the model is to allow us to include the removal of carbon from organic sediment stock through the burning of fossil fuels.
Figure 6. the full carbon cycle model used in this lab
Part 1 EXERCISES
1a. For each flux included in this carbon cycle model, briefly discuss how it functions and what you know about it from chapter 8 in Kump, Kasting, and Crane. (Expected length of answer 2-3 pages).
1b.Open up the STELLA Global Carbon Cycle model for 1860-2000 (pg250_lab8_carbon_cycle_1860_2000.STM). Look at the model layer and the equation layer to see that it is the same model described in this lab description. Also, look at the graphical output, and make sure you understand each item being graphed. Then, you will do a series of experiments. The first experiment will be a control run: to run the model in dynamic equilibrium representing the pre-industrial state of the carbon cycle. In the user interface level, you will find switches for deforestation and fossil fuel combustion. Turn both switches off, run the model, and look at the output graphs.Fill in the appropriate numbers in tables 1a, 1b, and 1c (below), and answer the following questions.
QUESTIONS: How much carbon is in the atmosphere at the beginning and at the end of the model experiment? How much is in the terrestrial, and how much is in the marine, biospheres? What has been thetotal input of carbon from deforestation and fossil fuel burning over this time period in this experiment? What is the total amount of carbon in the entire system? Do the values of atmospheric, land, or ocean carbon stocks change over time? Explain why or why not. Does the value of total carbon in the entire system change over time? Explain why or why not.
1c. Now we examine the effect of deforestation (actually, the total effect of global anthropogenic land cover changes). I put information into STELLA estimating the approximate annualnet input of carbon into the atmosphere from land cover changes since 1860.This information was obtained from the Carbon Dioxide Information Analysis Center (CDIAC; cdiac.esd.ornl.gov). Turn the deforestation switch on, and run the model again. Fill in the appropriate numbers in table 1 (below), and then answer the same questions from 1b. In addition, describe in words how the annual global emission rate of carbon into the atmosphere from land cover changes has varied since 1860 (make sure you look at the graph of annual emissions, not the total emissions to-date).
1d. Now we examine the effect of fossil fuel combustion. I put information into STELLA estimating the approximate annual input of carbon into the atmosphere from fossil fuel combustion (actually, the combined input from fossil fuel combustion, cement manufacturing, and gas flaring) since 1860, also taken from CDIAC. Turnthe deforestation switch off, turn the ff switch on, and run the model. Fill in the appropriate numbers in table 1 (below), and then answer the same questions from 1b. In addition, describe in words how the annual global emission rate of carbon into the atmosphere from fossil fuel burning has varied since 1860.
1e. Now, lets see what happens when we combine deforestation with fossil fuel emissions. Turn both switches on, and run the model. Fill in the appropriate numbers in table 1 (below), and then answer the same questions from 1b.How do the total emissions fromfossil fuel emissions compare to emissions from deforestation? Have their relative magnitudes changed over time? If so, how?
1f. Find the “missing sink.” One of the major outstanding questions related to the global carbon cycle is the fate of the carbon that was emitted by human activity. Not all of it stays in the atmosphere, and we are not sure where all of it goes!The “airborne fraction” is defined as the fraction of carbon emitted that remains in the atmosphere. Since the airborne fraction is less than one, some portion of the emitted CO2 ended up in stocks other than the atmosphere. What portion is this? Marine geochemists believe that over 100 Gt-C could have been absorbed by the ocean. Terrestrial ecologists believe that, at most, expansion and regrowth of temperate and boreal forests in the Northern Hemisphere may account for somewhat more than 100 Gt-C. Considering that emissions have been about 400 Gt-C, this leaves 100-200 Gt-C unaccounted for: this is the "missing sink."
Using numbers from table 1, calculate how much each stock has changed, calculate the airborne fraction, and enter the values intable 2. (Don’t forget: we must consider what was removed from the terrestrial biosphere by deforestation. This calculation is explained in the table.) According to the realistic scenario (which includes both deforestation and fossil fuel emissions), where does our model say the CO2 goes? According to our model, how much went into the oceans, and how much into the terrestrial biosphere? Is the missing sink on land, in the ocean, or both? (Remember, the “missing” part is the part that the ecologists have not found yet.)
experiment / Total def emission / Total ff emission / Total anthro emissionno def, ff
def only
ff only
def & ff
Table 1a. Total emissions from deforestation and fossil fuel burning, 1860-2000. All values in Gt-C
experiment / Atm C 1860 (Gt-C) / Atm C 2000 (Gt-C) / Atm C 1860ppm(v) CO2 / Atm C 2000 ppm(v) CO2 / Atm C(Gt-C) / Atm C
ppm(v) CO2
no def, ff
def only
ff only
def & ff
Table 1b. Beginning and ending values of atmospheric carbon stocks, and the change () between 1860 and 2000.
experiment / Land C 1860 (Gt-C) / Land C 2000(Gt-C) / Ocn C 1860 (Gt-C) / Ocn C 2000 (Gt-C)no def, ff
def only
ff only
def & ff
Table 1c. Beginning and ending values of the land and ocean carbon stocks.
experiment / Total anthro input(Gt-C) / Change in atm C
(Gt-C) / Airborne fraction (column2 / column1) / Change in land Cplus total def emission (Gt-C) / Change in ocean C (Gt-C)
no def, ff
def only
ff only
def & ff
Table 2. Changes in values of carbon stocks 1860-2000. All values except airborne fraction in Gt-C; airborne fraction is a fraction (a unitless number).
1g. EXTRA CREDIT Using the concepts and the equation discussed in critical thinking problem chapter 3 #6 in Kump, Kasting, and Crane, estimate the change in the outgoing longwave flux associated with the increased greenhouse effect in each of the four experiments we made in this lab, and use the zero-dimensional global radiation balance model to estimate the change in global equilibrium mean surface temperature associated with each scenario (assuming no additional feedbacks). Explain how you did your calculations, and fill in the values in table 3.
experiment / Change in LW-out (W/m2) / Change in Equilibrium Temp. (K)no def, ff
def only
ff only
def & ff
Table 3. Extra credit: estimated changes in LW-out flux and global mean surface temperature.
Part 2. GLOBAL CARBON CYCLE: 1860-2100
(file name = pg250_lab8_carbon_cycle_1860_2100.STM)
We will use the same model to examine what might happen to atmospheric carbon dioxide levels under different scenarios of future emissions. We know approximately what emissions have been in the past (we already used those numbers in part 1 of this lab). Now, we use a version of the model that is set up to run from 1860 all the way to 2100, and allows the user to choose from four different scenarios of global carbon emissions for the 21st century. Open the model. In the user interface level there are two switches and a slider. The two switches allow the user to turn on/off deforestation and fossil fuel emission, as in the previous model. If you like, you can make some runs playing with these switches, but for our purposes we will leave them in the “on” position. The slider controls ff_future_switch, which allows the user to choose amongst four different scenarios of fossil fuel emission during the years 2000 through 2001.
Part 2 EXERCISES
2a. Set the ff_future_switch to 1, which chooses scenario 1, and run the model. Graph 4 shows the annual global emission rates from deforestation and fossil fuel emissions. QUESTIONS:Describe in words what happens to fossil fuel emissions during the 21st century under this scenario, and what are the final emission rates (in the year 2100) compared to the year 2000.Describe in words how carbon storage in the atmosphere, on land, and in the ocean, each change during the 21st century for this scenario? Fill in the appropriate values in table 4.
2b. Set the ff_future_switch to 2, which chooses scenario 2, and run the model. Answer the same questions as in 2a.
2c. Set the ff_future_switch to 3, which chooses scenario 3, and run the model. Answer the same questions as in 2a.
2d. Set the ff_future_switch to 4, which chooses scenario 4, and run the model. Answer the same questions as in 2a.
2e. Discuss the results from 2a-2d with regards to the response time of the system. In particular, discuss what these results tell you about the potential effectiveness of different policies to curb global carbon emissions, and what kind of changes in atmospheric carbon levels you believe we can realistically expect during the 21st century.
experiment / Fossil fuel emission scenario (brief description) / Temporal change in atm C(brief description) / Change in atm C 2000-2100
(Gt-C) / Change in atm C 2000-2100
ppm(v)
Scenario 1
Scenario 2
Scenario 3
Scenario 4
Table 4. Scenarios of fossil fuel emissions for 2000-2100. Columns 1 and 2 briefly describe the trend in emission and atmospheric concentration, respectively. Columns 3 and 4 show the difference in atmospheric carbon levels between 2000 and 2100 expressed in units of Gt-C and ppm(v), respectively.
ESS I, Fall 2010, Hunter College, lab 8, p. 1 of 9