应用营养循环模型（NuCM）模拟酸沉降对重庆铁山坪马尾松林生态系统的影响

【摘要】：Being a part of work in Integrated Monitoring Program on Acidification of Chinese Terrestrial Systems (IMPACTS), the Nutrient Cycling Model (NuCM) was used to simulate the dynamics of nutrient cycling in a Masson pine (Pinus massoniana) forest ecosystems at Tieshanping site, Chongqing city. In addition, the modeling work made predictions of future nutrient cycling in the investigated forest ecosystems affected by several acid deposition scenarios. Two years of observed data (from 2001 to 2002) were used to make the model calibration. Nearly all the biogeochemical processes which affecte nutrient cycling in forest ecosystem were simulated. These processes include: water flux in the vegetation and the soil, dry deposition, foliar leaching, nutrient uptake, nutrient translocation, litterfall, litter decay, tree biomass increment, mineralization of organic matter, cation exchange (for Ca2+, Mg2+, K+, Na+, NH4+, H+, and Al3+), anion adsorption (for SO42-, H2PO4-, and organic ions), mineral weathering, aluminum hydrolysis, solid-liquid-gas phase equilibrium, and root respiration.The results of the model calibration showed that water flux in the vegetation and the soil were well simulated, even the annual precipitation varied very much during the two simulation years. Key parameters should be considered here were evapotranspiration rate and soil hydraulic conductivity.Dry deposition was estimated based on observed wet deposition and throughfall to make a good simulation on chemical fluxes purely in throughfall. Since no measured data on foliar leaching and nutrient translocation, the parameters used for throughfall and litterfall calibration were calibrated. Satisfied results on throughfall calibration were achieved, while Ca and S fluxes in litterfall were underestimated. Litter decay and organic matter mineralization were well simulated by adjusting relative parameters.Tree biomass increment was estimated according to published literatures, the parameters used for biomass simulation include estimated tree growth rate and measured chemical component in leaves, but estimated values in boles and roots.Cation exchange was well simulated but NH4 because of a limitation in the model. S adsorption calibration was satisfied, while P adsorption was failed to calibrate also due to the model limitation.There was no available measured data on mineral weathering and aluminum hydrolysis, all the parameters for these processes calibration were estimated. Satisfied results were obtained in these two processes calibration, that the simulated values were in good agreement with the observed values of soil water chemical concentrations. Root respiration was not considered in the model calibration work.Sensitivity analysis results showed that vegetation growth rate, mineral weathering rate and organic matter decay rate were the most important parameters when using NuCM at the Masson pine forest ecosystem at Tieshanping site, Chongqing city. Additional field measurements may be necessary in order to better quantify their values. In addition, the sensitivity analysis indicated that for parameters which had little impact on the simulation results (e.g. carbonate equilibrium data, atmospheric CO2 levels, soil temperature, etc.), resources need not to be expended to quantify these parameter values for future efforts.After the model had been calibrated for such a specific forest stand, the nutrient cycling model (NuCM) was used to simulate various scenarios. Four scenarios on different atmospheric S and N deposition loads were simulated. The scenarios were: 1) keep current deposition rate constant for 50 years; 2) reduce S deposition by 50% from the first simulation year but keep others constant for 50 years; 3) reduce S deposition by 50% but increase NH4 deposition by 100% from the first simulation year, and keep others constant for 50 years; and 4) reduce S deposition by 50% but double NOx deposition from the first simulation year, and keep others constant for 50 years.The results of scenarios simulation showed that reducing S deposition would increase vegetation growth at Plot B, but decrease it at Plot C and have no effect on Plot L. The reason was uncertain. Increasing N deposition increased vegetation growth in all the plots. Reducing S deposition increased soil exchangeable Ca, Mg and K, while increasing N had a contrary effect. Ca, Mg and K leaching were decreased while S deposition reduced. Increasing N deposition had little effect on cations leaching, but increased N leaching as well as N accumulation in soil pools.NuCM was well calibrated using the two years of observed data at Tieshanping site. However, there were a lot of uncertain parameters, and they were estimated according to literatures or calibrated by the model. For example, the parameters for foliar leaching, nutrient translocation and mineral weathering were deficient, and these were very important values when making throughfall and soil solution cation concentration calibration. In addition, soil solution chemistry variation during the simulation years was often failed to simulate. A lot of processes determined nutrients dynamics in soil solution. It was extremely difficult to accurately determine the relative parameters, such as S adsorption, Al hydrolysis, vegetation uptake, etc.