Collaborative Research: Thresholds and mechanisms of net ecosystem production (NEP) resilience following moderate disturbance: Why does one ecosystem recover and another one crash?

美国的森林是食物、纤维和能源的主要来源。它们在地球气候系统中发挥着基础性的作用，因为它们封存在植物中的生物质碳，否则这些碳可能会形成二氧化碳等大气温室气体的分子支柱。森林捕获大气二氧化碳和建立生物量的能力可能会随着年龄和干扰而发生重大变化。长期以来，科学家们一直在理论上认为，随着森林年龄的增长，生长和碳吸收会下降。然而，新的观察表明，在老化的森林中，低水平的干扰，如来自虫害、真菌病原体和极端天气的干扰，可能与直觉相反，维持甚至增加森林的碳固存和生长。这些高于预期的森林固碳率背后的机制尚不清楚。这项研究试图确定森林生长对干扰的恢复能力的基础机制及其阈值。研究人员还将评估不同的计算机模拟是否、如何以及为什么不能复制这种弹性。不同的计算机模拟对预测未来的森林碳储存、生长和产量至关重要。此外，他们将确定美国西部的常绿森林和东部的落叶林是否遵循独特的年龄-森林生长轨迹，这些森林具有不同的普遍干扰制度和气候。这个项目给社会、森林和土地管理者、小学教育工作者、大学生和森林科学家带来的好处是深远的。通过将生物信息的野外和模拟实验与北美森林的综合结合，这项研究将极大地推进我们对森林干扰的生态学思考，同时产生与生态系统和地球系统模拟直接相关和可获得的结果，供森林管理者努力在日益干扰的森林景观中最大限度地增加碳储存、生长和木材生产。该项目将为小学教师提供公开可用的教学材料，培训几名研究生和本科生，向科学家和土地管理人员提供公开和透明的数据和计算机代码来源，并在美国能源部的一个实验室和一个学术机构之间形成学生培训伙伴关系。随着美国中西部和东部的森林从早期森林演替向中期森林演替广泛推进，未来的陆地碳汇是不确定的。随着这种转变，早期演替的树冠优势种正在衰老，并让位于生物和结构更复杂的森林，这些森林越来越容易受到中等严重干扰的影响。最近的研究表明，这类森林的净初级生产力可能会以高于预期的速度维持，但目前的知识无法预测这种功能恢复能力的限度和背后的机制，因为目前的知识几乎完全来自对最近受到干扰的森林中严重的林分替代干扰动力学的研究。从相同的知识基础发展而来的生态系统和全球模型也很难再现适度干扰的影响。这项工作的三个核心研究目标是：1)确定支持净初级生产力对干扰的恢复能力的机制及其阈值；了解不同的森林模式是否、如何以及为什么未能复制这种净生产力恢复能力；以及3)阐明具有不同干扰机制和气候的温带落叶和针叶林是否遵循独特的年龄生产轨迹。该项目采用现场实验、模型测试和大规模数据合成三管齐下的方法，将人们对碳循环弹性的理解转变为对老化森林中一系列中等干扰强度的理解，阐明决定净初级生产弹性和衰退之间阈值的潜在机制。现场部分使用完全复制的干扰严重程度梯度，从0%到85%的落叶，系统地确定碳循环如何以及为什么随着干扰水平的上升而发生变化。PIS将采用一套碳和氮循环测量方法，重点研究树冠结构、叶片生理和树冠氮素再分配，以确定导致净初级生产力迅速恢复或在干扰后下降的机制。该项目的建模部分使用数据同化实验，在NSF支持的开源生态信息学工具箱中运行两个截然不同的生态生理模型，以确定模型假设未能模拟净初级生产力对干扰的弹性的最主要责任的过程，并反复告知下一个田间季节的采样优先级。最后，数据合成部分使用新获得的观测数据来描述干扰对北美温带森林年龄净生态系统生产轨迹的影响。

Forests of the United States are primary sources of food, fiber and energy. They play a fundamental role in the earth's climate system by sequestering in plant biomass carbon that might otherwise form the molecular backbone of atmospheric greenhouse gases such as carbon dioxide. Forests' capacity to capture atmospheric carbon dioxide and build biomass may change substantially with age and disturbance. Scientists have long theorized a decline in growth and carbon uptake as forests age. New observations, however, suggest that low levels of disturbance, such as those originating from insect pests, fungal pathogens, and extreme weather, in aging forests may, counter-intuitively, sustain or even increase forest carbon sequestration and growth. The mechanisms underlying these higher-than-expected rates of forest carbon sequestration are unknown. This study seeks to identify the mechanisms underpinning forest growth resilience to disturbance, and their thresholds. The researchers will also evaluate if, how, and why different computer simulations, critical to predicting future forest carbon storage and growth and yield, fail to replicate this resilience. Furthermore, they will determine whether evergreen forests in the western United States and deciduous forests in the East, with different prevailing disturbance regimes and climates, follow unique age-forest growth trajectories. The benefits of this project to society, forest and land managers, grade school educators, university students, and forest scientists are far-reaching. By combining biologically-informed field and simulation experiments with a synthesis of North American forests, this study will significantly advance our ecological thinking about forest disturbance, while producing results immediately relevant and accessible to ecosystem and earth system simulations, and to forest managers working to maximize carbon storage, growth, and timber production in increasingly disturbed forest landscapes. The project will produce openly available instructional materials for grade school teachers, train several graduate and undergraduate students, provide open and transparent sources of data and computer code to scientists and land managers, and form a student training partnership between a United States Department of Energy laboratory and an academic institution.  The future terrestrial carbon sink is uncertain as forests of the United States upper Midwest and east broadly advance from early to middle forest succession. With this transition, early successional canopy dominants are senescing and giving way to more biologically and structurally complex forests that are increasingly subject to moderate severity disturbance. Recent studies suggest that net primary production may be sustained in such forests at higher-than-expected rates, but the limits of and mechanisms behind such functional resilience cannot be predicted from present knowledge, which is derived almost entirely from studies of severe, stand-replacing disturbance dynamics in recently disturbed forests. Ecosystem and global models, developed from the same intellectual foundations, also have trouble reproducing the effects of moderate disturbances. The three core research objectives of this work are to: 1) identify mechanisms supporting net primary production resilience to disturbance, and their thresholds; understand if, how, and why different forest models fail to replicate this NPP resilience; and 3) elucidate whether temperate deciduous and coniferous forests, with different disturbance regimes and climates, follow unique age-production trajectories. The project uses a 3-pronged approach of field experiments, model testing, and large-scale data synthesis to transform understanding of how resilient the carbon cycle will be to a range of moderate disturbance intensities in aging forests, elucidating the underlying mechanisms that determine the threshold between net primary production resilience and decline. The field component uses a fully replicated gradient of disturbance severity, from 0 to 85 % defoliation, to systematically determine how and why the carbon cycle shifts in response to rising disturbance levels. The PIs will employ a suite of carbon and nitrogen cycling measurements, focusing on canopy structure, leaf physiology, and canopy nitrogen reallocation, to identify the mechanisms that cause rapid net primary production resilience or decline following disturbance. The modeling component of the project uses data assimilation experiments, running two very different ecophysiological models within an open source, NSF-supported ecoinformatics toolbox, to identify the processes most responsible for the models' hypothesized failure to simulate net primary production resilience to disturbance, and iteratively inform the next field season's sampling priorities. Finally, a data synthesis component uses newly available observations to characterize disturbance effects on age-net ecosystem production trajectories for North American's temperate forests.