Yuri Kim
Geog711 Problem set 3
10/09/2007
Daily to decadal dynamics of nitrogen cycling
1. Statement of the problem
What is the relationship between nitrogen deposition concentration and vegetation growth? How does nitrogen deposition concentration effect on soil carbon cycling and soil water condition?
2. Experimental setup
The purpose of this simulation is focusing on a role of nitrogen to the vegetation and soil water condition. Therefore, this simulation sets scenariosof increasing the values of nitrogen deposition under the around doubling CO2 concentration of current condition, which means limitless carbon supply condition, 600ppm.“CO2_FALG”in the ch8004.ini file should be set “0” in order to apply constant 600ppm to the whole simulation. The three levels of nitrogen deposition for the simulation are organized in Table 1.
The study site isChapel Hill, NC. Before running simulation, spin up process was done with 1.0m loamy soil—silt 50%, sand 30%, and clay 20%--to stabilize initial condition of the simulation. Simulation period is 500 years for the annual simulation and 25 years for the daily simulation. In order to simulate the future whose climate data do not exist, recycling of 25 year weather data—from 1980 to 2004—is inevitable. The climate conditions of 25 years are provided in “ch8004.met.”
The vegetation type of this simulation is Evergreen needle leaf (enf). One parameter varied in this simulation is “Canopy evaporation coefficient” in epc file; it changed from 0.0045 l/LAI/d to 0.001 l/LAI/d because of the unrealistic canopy evaporation—too high—with 0.0045 l/LAI/d. The 1m loamy soil—silt 50%, sand 30%, and clay 20%--is the soil for this simulation.
The dependent variables that can be produced with above all setting are; GPP, NPP, MR, HR, maximum LAI, evaporation, and outflow for the annual simulation; and GPP, NPP, MR, HR, soil water potential, soil water content, soil stores for carbon—e.g. soilC, soil1C, soil2C, soil3C, and soil4C)--, litter stores for carbon—LITRC, LITR1C, LITR2C, LITR3C, and LITR4—and so on for daily simulation.
Table 1. Simulation setup
Input file / Output file / Treatment VariableATM_CO2 (ppm) / NDEP (kgN/m2/yr)
0.0001N / ch8004.enf.0.0001n / ch8004.enf.0.0001n.(annout/dayout) / 600 / 0.0001
0.001N / ch8004.enf.0.001n / ch8004.enf.0.001n.(annout/dayout) / 600 / 0.001
0.005N / ch8004.enf.0.005n / ch8004.enf.0.005n.(annout/dayout) / 600 / 0.005
3. Results
1) Annual scale result
Figure 1. Annual maximum LAI of three nitrogen deposition level—0.0001N, 0.001N, and 0.005N—for 500 year simulation.
0.0001N (defaulting nitrogen deposition) shows decreasing LAI trend after a prominentLAI spike of 1985, whereas the LAI of higher N deposition value (0.005N) showed increasing and asymptotic trend. The LAI of 0.001N shows decreasing trend in the early stage of the simulation, but it starts to increase againfrom around 2180.The reason for this phenomenon could be that the nutrients in the soil was depleted in the earlier period of the simulation and regainedthe nutrients by decomposing of litter. Therefore, figure 1 reflects that nitrogen play a role of limiting factor for the vegetation growth.
2) Daily scale result
In order to consider nearly limitless supply of soil water, two wet years are chosen for the simulation. One is 1996 which is followed by wet year,and the other is 2003 followed by dry year (Table 2).
Table 2. Annual total precipitation for the daily scale result
Year / 1995 / 1996 / 2002 / 2003Precipitation (mm) / 1404 / 1417 / 1046 / 1405
Figure 2. GPP, NPP and soil water potential of 1996 and 2003.
Figure 2 show how GPP and NPP of three levels of nitrogen deposition scenarios are associated with soil water stress. High nitrogen content soil, 0.005N, generally produces more vegetation mass than 0.001N and 0.0001N. This causes more soil water stress environment in 0.005N case comparing with 0.001N and 0.0001N. Therefore, during a growing season, the period that NPP of 0.005N is lower than that of 0.001N and 0.0001N is generally accompanied by the period that soil water potential of 0.005N is much higher than that of other two cases.Around day 162 of 1996 and day 182 of 2003 are the good examples of showing impact of soil water stress on the vegetation growth. This result reflects soil water is a limiting factor for the vegetation growth even though soil is highly fertilized.
Figure 3. The relationship between litter1 carbon and LAI.
The LTR1C, which represents the fastest decay rated litter store for carbon, show higher value with lower nitrogen deposition scenario except for the red arrowed period of figure 3—from around day 125 to day 175 for 1996, and from around day 150 to day 200.
One of the possible reasons for the higher carbon content in litter1 pool with 0.005N scenario is soil moisture shortage. According to the soil water potential graph of figure 2, water stress of 0.005N is much severer than low nitrogen deposition scenarios during the red arrowed period. This makes the seasonal patterns of LAI different (figure 3).The figure 3 shows that LAI of 0.005N has some decreasing trend during that period, whereas LAI of 0.0001N keeps increasing. LITR1Ccan directly represent the amount of falling leaves. Therefore, it can be said that 0.005N lost leaves in this period, and this produces higher carbon content in litter1 pool.
Also this soil water stress prohibits soil decomposition of litter. Thus relatively large amount of carbon can be remained in 0.005N litter1 pool than 0.001N and 0.0001N litter1 pool.
4. Conclusion and Discussion
In conclusion, higher nitrogen deposition produces more vegetation mass than lower nitrogen deposition case under the same conditions of water and carbon supply. Also soil water is the ultimate limiting factor for the vegetation growth; even though limitless amount of nutrients can be supplied to the soil, growing vegetations exploit soil water, and this phenomenon results in stopping further vegetation growth.
One of the remained questionsof this simulation result is the relationship between soil carbon and litter carbon. Figure 4 shows that the rank of soil1 carbon amount is the order of 0.005N, 0.001N and 0.0001N, however, litter1 carbon shows generally opposite order. The logics or formulas about relationship between soil carbon and litter carbon in BIOME-BGC are required in order to solve this problem.
Figure 4. Soil1 carbon and litter1 carbon of year 1996 and 2003.