Carbon and Nitrogen Accumulation in a Savanna Landscape : Field and Modeling Perspectives
Carbon and Nitrogen Accumulation in a Savanna Landscape : Field and Modeling Perspectives
复制标题
DOI：
--
复制
发表时间：
2004
期刊：
影响因子：
--
通讯作者：
S. Archer;T. Boutton;C. McMurtry
中科院分区：
文献类型：
--
作者：
S. Archer;T. Boutton;C. McMurtry
间接评估表明，在过去100年中，世界旱地木本植物的增加可能对碳固存产生了重大影响。然而，这些评估的特点是高度的不确定性。一个链接的演替-植被地球化学模型最近被用来估计在过去的150年在北美大平原南部网站的木本植物的增殖伴随着生态系统的碳和氮库的变化。在这里，我们使用历史航空照片（1950年，1976年和1990年）和现场数据评估建模方法的有效性。土壤有机碳（SOC）和全氮（TN）库在木本植物入侵的土壤中相对于残留草地群落富集，其富集程度与木本斑块年龄呈线性函数关系。然而，木本斑块年龄仅解释了21-57%的SOC和44-68%的TN的方差，这表明树木占用时间以外的因素在起作用。从1950年至1990年，对地上植物碳库变化的实地航空照片评估一直较低（约100万美元）。23 ~ 38%）。一些差异可能反映了这样一个事实，即景观响应，而个别方面的碳封存率，而建模方法代表“平均”景观。模拟的长期地上植物碳固存率（360公斤碳公顷-1y-1）落在广泛的范围内观察到的短期（14至26年）率（32至933公斤碳公顷-1y-1）观察到的三个景观。TN累积速率的实地和基于模型的估计值之间的差异远大于SOC。此外，库存的植物群落之间的差异的迹象各不相同。我们的数据表明，预测和理解SOC和TN的变化后，植物群落生活型组成的变化不能可靠地实现使用简单的，经验的关系。相关联的演替地球化学模型是能够产生的植物和土壤碳储量的估计与基于现场的估计一致，但没有产生土壤总氮的变化相比，基于现场的评估。可能需要纳入物种、地形土壤和土地利用历史对SOC和TN的影响，以提高动态、机械模拟模型的性能。
Indirect assessments suggest that increases in woody plants in the world’s drylands during the past 100 y may have had a significant impact on carbon sequestration. However, these assessments are characterized by a high degree of uncertainty. A linked succession-biogeochemistry model has recently been used to estimate changes in ecosystem carbon and nitrogen pools accompanying the proliferation of woody plants over the past 150 y at a southern Great Plains site in North America. Here, we evaluate the validity of that modeling approach using historical aerial photos (1950, 1976, and 1990) and field data. Field sampling indicated soil organic carbon (SOC) and total nitrogen (TN) pools on sites invaded by woody plants become enriched relative to remnant grassland communities, the extent being a linear function of woody patch age. However, woody patch age explained only 21–57% of the variance in SOC and 44–68% of the variance in TN, suggesting factors other than time of tree occupation are at play. Field-aerial photo assessment of changes in aboveground plant carbon pools from 1950–1990 were consistently lower (ca. 23 to 38%) than those predicted by the linked succession-biogeochemistry model. Some of the discrepancy likely reflects the fact that landscapes responded rather individualistically with regard to rates of C sequestration, whereas the modeling approach represents “an-average” landscape. The modeled longterm rate of aboveground plant C sequestration (360 kg C ha–1y–1) fell well within the broad range of observed short-term (14 to 26 y) rates (32 to 933 kg C ha–1y–1) observed on three landscapes. Discrepancies between fieldand model-based estimates of TN accumulation rates were much greater than those for SOC. In addition, the sign of the differences varied among the plant communities inventoried. Our data suggest that predicting and understanding changes in SOC and TN following shifts in plant community life form composition cannot be reliably achieved using simple, empirical relationships. The linked successionbiogeochemistry model was able to generate estimates of plant and soil C storage consistent with field-based estimates, but did not generate changes in soil TN comparable to field-based assessments. Inclusion of species, topoedaphic, and land use history effects on SOC and TN may be required to improve the performance of dynamic, mechanistic simulation models.