Historical and projected carbon balance of mature black spruce ecosystems across North America: the role of carbon-nitrogen interactions
Historical and projected carbon balance of mature black spruce ecosystems across North America: the role of carbon-nitrogen interactions
复制标题
DOI：
10.1023/a:1019673420225
复制
发表时间：
2002-05-01
期刊：
影响因子：
4.9
通讯作者：
Massheder, JM
中科院分区：
文献类型：
--
作者：
Clein, JS;McGuire, AD;Massheder, JM
通过应用陆地生态系统模型 (TEM) 模型参数化（其中 C-N 动态耦合或非耦合），评估了北美黑云杉生态系统中碳 (C) 和氮 (N) 相互作用对大气 CO2 封存的作用。首先，通过利用当地气候数据模拟北部研究区（北部站点）和南部研究区（南部站点）黑云杉生态系统北方生态系统大气研究（BOREAS）涡相关塔站点的 C 动态，评估了为阿拉斯加博南扎溪长期生态研究站点黑云杉生态系统动态而开发的参数化的性能。我们将 1994 年至 1997 年模拟的每月生长季节（5 月至 9 月）初级生产总值（GPP）、生态系统总呼吸（RESP）和生态系统净生产（NEP）估计值与两个地点的现有实地估计值进行了比较。在北部站点，耦合和非耦合模拟的 GPP 和 RESP 月生长季估计与基于现场的估计高度相关（耦合：R-2 = 0.77，GPP 和 RESP 为 0.88；非耦合：R-2 = 0.67，GPP 和 RESP 为 0.92）。尽管 NEP 的模拟季节模式总体上与基于实地的数据相匹配，但与基于实地的 GPP 和 RESP 之间的相关性相比，基于实地的和模拟的每月生长季 NEP 之间的相关性较低（耦合和非耦合模拟的 R-2 = 0.40、0.00）。耦合参数化模拟的年度 NEP 在三年内有两年处于基于实地估计的不确定性范围内。另一方面，通过非耦合参数化模拟的年度 NEP 仅在三年中有一次落在基于现场的不确定性范围内。在南部站点，模拟 NEP 通常与基于实地的 NEP 估计相匹配，并且每月生长季节基于实地的 NEP 与模拟 NEP 之间的相关性（对于耦合和非耦合模拟，分别为 R-2 = 0.36、0.20）与北部站点的相关性相似。为了评估氮动态在北美黑云杉生态系统碳平衡中的作用，我们利用 TEM 的耦合和非耦合参数化，以 0.5 度分辨率（纬度 x 经度）的全球气候学模拟了 1900 年至 2100 年的历史和预测碳动态。根据对北部地点的分析，出现了几种一致的模式。与耦合参数化相比，非耦合参数化模拟的净初级生产（NPP）年际变化更大，这导致参数化之间的NEP年际变化存在显着差异。在非耦合模拟中，NPP 和异养呼吸之间的差异更大，导致预测期间有更多的碳封存。这些响应是耦合和非耦合参数化对二氧化碳和气候变化的根本不同响应的结果。在北美黑云杉生态系统中，非耦合参数化模拟的碳储量年代际变化范围远大于耦合参数化。对碳动态年代际响应的空间变异性分析表明，耦合和非耦合参数化模拟的碳通量对气候具有不同的敏感性，并且通量的气候敏感性在模拟的时间范围内变化。本研究的结果表明，可以通过以下方式减少不确定性：（1）析因研究重点阐明碳和氮相互作用在成熟黑云杉生态系统对大气二氧化碳和气候操纵的响应中的作用， (2) 在北美成熟黑云杉生态系统范围内建立一个连续、长期的碳动态测量网络，(3) 网络中的辅助测量，以阐明碳和氮相互作用在二氧化碳与大气交换中的作用。
The role of carbon (C) and nitrogen (N) interactions on sequestration of atmospheric CO2 in black spruce ecosystems across North America was evaluated with the Terrestrial Ecosystem Model (TEM) by applying parameterizations of the model in which C-N dynamics were either coupled or uncoupled. First, the performance of the parameterizations, which were developed for the dynamics of black spruce ecosystems at the Bonanza Creek Long-Term Ecological Research site in Alaska, were evaluated by simulating C dynamics at eddy correlation tower sites in the Boreal Ecosystem Atmosphere Study (BOREAS) for black spruce ecosystems in the northern study area (northern site) and the southern study area (southern site) with local climate data. We compared simulated monthly growing season (May to September) estimates of gross primary production (GPP), total ecosystem respiration (RESP), and net ecosystem production (NEP) from 1994 to 1997 to available field-based estimates at both sites. At the northern site, monthly growing season estimates of GPP and RESP for the coupled and uncoupled simulations were highly correlated with the field-based estimates (coupled: R-2= 0.77, 0.88 for GPP and RESP; uncoupled: R-2 = 0.67, 0.92 for GPP and RESP). Although the simulated seasonal pattern of NEP generally matched the field-based data, the correlations between field-based and simulated monthly growing season NEP were lower (R-2 = 0.40, 0.00 for coupled and uncoupled simulations, respectively) in comparison to the correlations between field-based and simulated GPP and RESP. The annual NEP simulated by the coupled parameterization fell within the uncertainty of field-based estimates in two of three years. On the other hand, annual NEP simulated by the uncoupled parameterization only fell within the field-based uncertainty in one of three years. At the southern site, simulated NEP generally matched field-based NEP estimates, and the correlation between monthly growing season field-based and simulated NEP (R-2 = 0.36, 0.20 for coupled and uncoupled simulations, respectively) was similar to the correlations at the northern site. To evaluate the role of N dynamics in C balance of black spruce ecosystems across North America, we simulated historical and projected C dynamics from 1900 to 2100 with a global-based climatology at 0.5degrees resolution (latitude x longitude) with both the coupled and uncoupled parameterizations of TEM. From analyses at the northern site, several consistent patterns emerge. There was greater inter-annual variability in net primary production (NPP) simulated by the uncoupled parameterization as compared to the coupled parameterization, which led to substantial differences in inter-annual variability in NEP between the parameterizations. The divergence between NPP and heterotrophic respiration was greater in the uncoupled simulation, resulting in more C sequestration during the projected period. These responses were the result of fundamentally different responses of the coupled and uncoupled parameterizations to changes in CO2 and climate. Across North American black spruce ecosystems, the range of simulated decadal changes in C storage was substantially greater for the uncoupled parameterization than for the coupled parameterization. Analysis of the spatial variability in decadal responses of C dynamics revealed that C fluxes simulated by the coupled and uncoupled parameterizations have different sensitivities to climate and that the climate sensitivities of the fluxes change over the temporal scope of the simulations.The results of this study suggest that uncertainties can be reduced through (1) factorial studies focused on elucidating the role of C and N interactions in the response of mature black spruce ecosystems to manipulations of atmospheric CO2 and climate, (2) establishment of a network of continuous, long-term measurements of C dynamics across the range of mature black spruce ecosystems in North America, and (3) ancillary measurements in the network to elucidate the role of C and N interactions in exchange of CO2 with the atmosphere.