New Geocarb Model

These files provide a new version of the GEOCARB model that allows you to simulate a CO2 emissions trajectory (e.g., from burning fossil fuels).

To run this, you need to have python version 3.5 or higher installed, and you need the python numpy and pandas packages installed. You also need the R reticulate package, but this should install automatically the first time you run the script.

To use this package:

• First execute source("new_geocarb.R") to load the script.
• Then execute load_geocarb() to load the python GEOCARB library.
• Then execute run_geocarb (details below).
  • You can specify two or more periods with different emissions profiles.
  • The first period is always the spinup period. By default it's 5 million years long, and we keep the last 2 million years.
  • The second and subsequent periods are the "simulation" periods, where we change the degassing rate, human CO2 emissions, etc.
• For any parameter that you don't want to change, you can just specify a number.
• For any parameter that you do want to change from one period to another, you pass a vector or list whose length is the number of periods.
• For instance, to have three periods, with a degassing rate of 7.5 in the spinup, 10.0 in period 1, and 12.5 in period 2, you would specify degassing = c(7.5, 10.0, 12.5).
• You can specify human emissions of CO2 that changes over time with the co2_emissions parameter. This should be a list with one element for each period.
  • For any period with no human emissions, that list element should just be the number 0.
  • For any period with constant emissions, that list element should just be a number representing the emissions, in billion tons per year.
  • For any period with changing emissions, that list element should be a vector or list whose length is the number of time steps in that period (for a 100 year period with a 5 year time step, this would be length of 20).
  • To specify three periods with no emissions in the spinup, a steady increase in the first simulation period, from 1 billion tons per year in the first year, up to 100 billion tons in the 100th year, and no emissions in the third period, you'd write, co2_emissions = list(0, seq(1, 100, 1), 0)
• You specify the periods with two parameters:
  • periods is a list or vector of numbers representing the number of years in each period. If you want to use three periods, with a 5 million year spin-up, a 100-year first simulation period, and a 2 million year second simulation period, you'd say, periods = c(5E6, 100, 2E6).
  • time_steps is a list or vector of numbers representing the number of years per time-step for each period. If you want to use three periods with a 50-year time step for the spinup, a 1 year time step for the first simulation period, and a 50 year period for the second simulation period, you'd say, time_steps = c(50, 1, 50). 50 years is a good time step when you're simulating millions of years, but if you're simulating a short period with rapid changes, you might want to use a shorter time-step.
• To read in the results of the GEOCARB run, you use the function read_geocarb to read in the file you specified when you ran the model with run_geocarb.

source("new_geocarb.R")

load_geocarb() # This loads the python geocarb library.

run_geocarb("geocarb_test_run.csv", # file to save the results in.
            co2_spike = 0,    # to put a single spike of 1000 GT CO2
                              # at the beginning of the first simulation
                              # period, you'd say
                              # co2_spike = c(0, 1000, 0)
            co2_emissions = list(
                0,                # no emissions in spinup
                seq(1, 100, 1),   # ramp up emissions for 100 years
                0                 # emissions stop abruptly at
                ),                # beginning of 2nd sim period.
             periods = c(5E6, 100, 2E6),  # 3 periods
             time_steps = c(50, 1, 50),   # time steps for the three periods.
             degas = 7.5,        # constant degassing rate for all three periods.
             plants = TRUE, land_area = 1,
             delta_t_2x = 3.0,   # Climate sensitivity, in degrees C per doubling of CO2.
             million_years_ago = 0,
             mean_latitude_continents = 30,
             start_recording = -2E6)  # Start recording data when
                                      # the spinup gets to -2 million years.

gc = read_geocarb("geocarb_test_run.csv")