Coupling Biocrusts and Vegetation Dynamics to Improve Predictions of Dryland Change

世界范围内的环境变化已经将生态系统推向了极限，因此越来越需要更好地预测生态系统对这些外部压力的反应。先前的研究表明，在许多生态系统中发现的空间模式在接近临界点时以可预测的方式发生变化。这些空间格局可能预示着整个生态系统即将发生变化。旱地已被用作发展这一理论的关键系统。该理论预测，随着气候变得更加干燥，旱地植被的模式会从同质植被的光秃秃的缝隙转变为迷宫状或条纹状的植被覆盖，然后转变为斑点状的植被覆盖，最后灾难性地转变为光秃秃的状态。利用这种空间格局建立生态系统变化预警系统，需要准确描述重要反馈；否则，“错误警报”可能导致代价高昂的资源错配。不幸的是，目前的理论预测仅限于全球一小部分旱地。这可能是因为目前的模型忽略了生物土壤结皮或生物结皮。生物结皮是土壤颗粒和活微生物（如蓝藻、地衣、苔藓、细菌）之间的联系，这些微生物生活在土壤的上层几毫米内或上面。生物结皮可以覆盖比维管植物更大的表面积，在水、能量和营养的动态中起着关键作用。它们还改变了构成植被空间格局的核心过程。在预测气候驱动的旱地变化时，忽视生物结壳会带来很大的不确定性。目前的生态系统空间自组织理论只考虑一种同质的生物组合，在旱地的情况下，维管植物。将旱地概念化为一个完整的生物壳-维管植物复合体需要新的理论和模型来整合种间相互作用（如竞争、促进）和空间自组织。本研究以旱地为重点研究区域，发展了这样的理论和模型。该项目通过(1)在实验室中操纵降雨制度和跟踪生物结皮斑块的形成，以及(2)监测美国西南部的旱地来研究生物结皮斑块的动态。旱地格局形成模式将维管植物与生物结皮联系起来。将这些模型与先前不考虑生物结壳的模型进行比较，阐明了生物结壳在旱地恢复力中的作用，并预测了气候变化下生态系统状态（如生产力）和空间格局如何随着降雨制度的改变而变化。最后，多源遥感图像和深度学习方法指导大规模生物结壳制图。由此产生的模型预测了未来几十年气候变化对美国西南部旱地的影响。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Environmental changes worldwide have pushed ecosystems towards their limit, and there is a growing need to better predict ecosystem responses to these external pressures. Prior work suggests that spatial patterns found in many ecosystems change in a predictable way when approaching a tipping point. These spatial patterns can be harbingers of imminent whole ecosystem change. Drylands have been used as pivotal systems in developing such a theory. The theory predicts that as the climate becomes drier, patterns of dryland vegetation shift from bare gaps in homogenous vegetation to labyrinthine or striped vegetation cover, then to a spotty pattern, before catastrophically shifting to a bare state. Operationalizing an early warning system of ecosystem change using such spatial patterns requires accurate characterization of significant feedbacks; otherwise, “false alarms” could lead to costly resource misallocations. Unfortunately, predictions from current theory are limited to only a small area of drylands globally. This is likely because models currently ignore biological soil crusts, or biocrusts. Biocrusts are associations between soil particles and living microorganisms (e.g., cyanobacteria, lichens, mosses, bacteria) that live within or immediately on top of, the upper few millimeters of soils. Biocrusts can cover more surface area than vascular plants and play a critical role in the dynamics of water, energy, and nutrients. They also alter core processes that underlie vegetation spatial patterns. Neglecting biocrusts poses significant uncertainties in predicting dryland change driven by climate. Current theory of ecosystem spatial self-organization considers only one homogenous assemblage of organisms, in the case of drylands, vascular plants. Conceptualizing drylands as an integral biocrust-vascular plant complex requires new theories and models that integrate inter-specific interactions (e.g., competition, facilitation) and spatial self-organization. This research develops such theories and models, using drylands as a focal study area. The project investigates biocrust patch dynamics by (1) manipulating rainfall regimes in the lab and tracking biocrust patch formation and (2) monitoring drylands of the U.S. Southwest. Dryland pattern formation models link vascular plants with biocrusts. Comparing these models, with prior models that do not consider biocrusts, elucidates the role of biocrusts in dryland resilience and predicts how ecosystem state (e.g., productivity) and spatial patterns change with altered rainfall regimes under climate change. Lastly, multi-source remote sensing imagery and deep learning approaches guide large-scale biocrust mapping. The resultant models predict changes to drylands in the U.S. Southwest over the next few decades of anticipated climate change.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.