Collaborative Proposal: MSB-FRA: A macrosystems ecology framework for continental-scale prediction and understanding of lakes

合作提案：MSB-FRA：用于大陆尺度预测和湖泊理解的宏观系统生态学框架

基本信息

批准号：
1638679
负责人：
Kendra Cheruvelil
金额：
$ 239.27万
依托单位：
Michigan State University
依托单位国家：
美国
项目类别：
Continuing Grant
财政年份：
2016
资助国家：
美国
起止时间：
2016-10-15 至 2025-03-31
项目状态：
未结题

来源：
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1638679&HistoricalAwards=false
关键词：
Collaborative Proposal MSB FRA macrosystems

项目摘要

Lakes are recognized as hotspots for processing carbon, nitrogen, and phosphorus and thus are critical for understanding how human activities affect global cycles of these essential nutrients. However, to estimate the total contribution of lakes in the United States to these global cycles, they have to rely on measurements from a small number of well-studied lakes because scientists do not have the resources to study every lake all the time. The resulting extrapolations to estimate global cycles and predict future change have many uncertainties. Consequently, it is important to understand where and when information from small subsets of lakes can be accurately applied to the wide variety of lake types and landscape settings across the continental United States. To improve future extrapolation efforts and to understand the role of lakes in global nutrient cycles, this award will build an unprecedented database that combines nutrient measurements from existing government and university monitoring programs (for about 15,000 lakes) with lake and landscape characteristics from national publicly-available digital maps for all lakes in the continental United States (about 130,000 lakes). Using this novel and unprecedented database, three components will be studied that are needed to determine the contribution of lakes to continental nutrient cycles. First, lake nutrients will be studied jointly rather than individually to provide insights into the conditions in which cycles are linked or not, which will help to reduce uncertainty in continental estimates of lake nutrients. Second, as scientists expand their studies from a few lakes to the entire continent, the relationships between lake nutrients and their landscape controls can differ in strength and even direction among different regions, further contributing to uncertainties in continental understanding of lake nutrient cycles. Finally, compiling data on every lake increases the chance of discovering novel environmental conditions that have not previously been studied, yet may play important roles in continental-scale nutrient cycles. Through these important research activities, scientists will increase their confidence in estimating the effects of lakes on global cycles. This award contributes to the broader scientific community because the database will be made publicly-available in a timely manner to complement the National Ecological Observatory program and to developing open-source advanced computer tools for analyzing large datasets for this and other big-data studies. In addition, the diverse team (by gender, career-level, and discipline) will train and mentor early-career scientists in interdisciplinary, team-based, and data-intensive science to be leaders in addressing challenging questions such as how future land use intensification and changes in global climate will affect lakes and the services they provide. Ecosystems, such as lakes, are complex, heterogeneous, and strongly influenced by their ecological context?environmental or anthropogenic factors that operate at multiple scales. This complexity makes extrapolating site-level estimates of ecological services, state, and function challenging. The overarching goal of this research is to understand and predict patterns in the three major nutrients for all continental US lakes to inform estimates of lake contributions to continental and global cycles of nitrogen, phosphorus, and carbon. The proposed work will address three important phenomena that limit scientists? ability to extrapolate freshwater nutrients at continental scales. (1) Because cycles of nitrogen, phosphorus, and carbon in inland water interact with each other and are often affected by similar controls, they should be considered as linked, not isolated. (2) As studies expand to view the whole continent, interactions between driver variables at different scales (cross-scale interactions) also increase. (3) A hallmark of the Anthropocene is the rise of novelty in ecosystems--new environmental conditions or new combinations of conditions. Such novelty may confound extrapolation in unknown ways. The proposed research is an unprecedented effort that will: address these important phenomena, develop new continental-scale data products for aquatic macrosystems ecology, and contribute novel, data-intensive analytical methods from computer science and statistics. This award will answer five research questions related to the above phenomena using two approaches. First, funds will be used to build a large, integrated database of all lakes in the continental United States (called LAGOS-US) that includes measures of in situ nutrients collected from tens of thousands of lakes, and ecological-context metrics calculated for all 130,000 continental lakes using geographic information systems and remote sensing datasets. Second, analyses of the database will be conducted for each research question using existing and novel statistical and computer science analytical tools to improve macrosystems ecology knowledge of freshwater nutrients. This award will complement the National Ecological Observatory strengths by providing data for a broader range of aquatic ecosystems and by providing the ecological context for the six continental Observatory lake sites. This award will result in four major intellectual contributions to macrosystems ecology. (1) The identification of regions where coupling and decoupling of nutrients occur, leading to a more comprehensive understanding of relationships between ecological context drivers and linked nutrient cycles. (2) Increased understanding of the types and spatial structure of ecological contexts that are more likely to lead to cross-scale interactions. (3) The identification of the role that novelty in ecological context plays in continental-scale predictions. (4) The transformation of understanding of the ecological contexts that influence biogeochemical cycles at macroscales and lake contributions to these cycles. Given the likely prevalence of such phenomena in other macrosystems, the results will be transferable to other ecosystem types, and more broadly to macrosystems ecology.

湖泊被认为是处理碳、氮和磷的热点，因此对于了解人类活动如何影响这些必需营养素的全球循环至关重要。然而，为了估计美国湖泊对这些全球循环的总贡献，他们必须依赖于少数经过充分研究的湖泊的测量结果，因为科学家没有资源一直研究每个湖泊。由此产生的估计全球周期和预测未来变化的推断存在许多不确定性。因此，了解何时何地可以将来自小型湖泊子集的信息准确地应用于美国大陆的各种湖泊类型和景观环境非常重要。为了改进未来的外推工作并了解湖泊在全球养分循环中的作用，该奖项将建立一个前所未有的数据库，该数据库将现有政府和大学监测项目（约 15,000 个湖泊）的养分测量结果与国家公共部门的湖泊和景观特征相结合。美国大陆所有湖泊（约 130,000 个湖泊）的可用数字地图。使用这个新颖且前所未有的数据库，将研究确定湖泊对大陆养分循环的贡献所需的三个组成部分。首先，将联合而不是单独研究湖泊养分，以深入了解循环相关或不相关的条件，这将有助于减少大陆对湖泊养分估计的不确定性。其次，随着科学家将研究范围从几个湖泊扩展到整个大陆，不同地区的湖泊养分与其景观控制之间的关系可能在强度甚至方向上有所不同，这进一步增加了大陆对湖泊养分循环理解的不确定性。最后，收集每个湖泊的数据增加了发现以前未曾研究过的新环境条件的机会，但可能在大陆规模的养分循环中发挥重要作用。通过这些重要的研究活动，科学家将增强估计湖泊对全球循环影响的信心。该奖项为更广泛的科学界做出了贡献，因为该数据库将及时公开，以补充国家生态观测站计划，并开发开源先进计算机工具来分析本研究和其他大数据研究的大型数据集。此外，多元化的团队（按性别、职业水平和学科）将培训和指导跨学科、基于团队和数据密集型科学的早期职业科学家，使他们成为解决具有挑战性的问题的领导者，例如未来的土地使用方式全球气候的加剧和变化将影响湖泊及其提供的服务。湖泊等生态系统是复杂的、异质的，并且受到其生态背景（在多个尺度上运行的环境或人为因素）的强烈影响。这种复杂性使得对生态服务、状态和功能进行现场水平的推断具有挑战性。这项研究的总体目标是了解和预测美国所有大陆湖泊三种主要营养物质的模式，以便估计湖泊对大陆和全球氮、磷和碳循环的贡献。拟议的工作将解决限制科学家的三个重要现象？推断大陆尺度淡水营养物质的能力。 (1) 由于内陆水中的氮、磷和碳循环相互作用，并且经常受到类似控制的影响，因此应将它们视为相互关联的而不是孤立的。（2）随着研究范围扩大到整个大陆，不同尺度的驱动变量之间的交互作用（跨尺度交互作用）也随之增加。 (3) 人类世的一个标志是生态系统中新奇事物的兴起——新的环境条件或新的条件组合。这种新颖性可能会以未知的方式混淆推断。拟议的研究是一项前所未有的努力，它将：解决这些重要现象，为水生宏观系统生态学开发新的大陆尺度数据产品，并从计算机科学和统计学中提供新颖的数据密集型分析方法。该奖项将使用两种方法回答与上述现象相关的五个研究问题。首先，资金将用于建立美国大陆所有湖泊的大型综合数据库（称为 LAGOS-US），其中包括从数万个湖泊收集的原位营养物质的测量值，以及为所有湖泊计算的生态背景指标。使用地理信息系统和遥感数据集研究 130,000 个大陆湖泊。其次，将使用现有的和新颖的统计和计算机科学分析工具对每个研究问题进行数据库分析，以提高淡水营养物的宏观系统生态学知识。该奖项将通过提供更广泛的水生生态系统的数据以及提供六个大陆观测站湖址的生态背景来补充国家生态观测站的优势。该奖项将为宏观系统生态学带来四项重大智力贡献。 (1) 识别养分耦合和解耦发生的区域，从而更全面地了解生态环境驱动因素和相关养分循环之间的关系。（2）增加对更有可能导致跨尺度相互作用的生态环境类型和空间结构的理解。 (3) 确定生态环境中的新颖性在大陆尺度预测中所起的作用。（4）对影响宏观尺度生物地球化学循环的生态环境以及湖泊对这些循环的贡献的理解的转变。鉴于此类现象在其他宏观系统中可能普遍存在，研究结果将可转移到其他生态系统类型，更广泛地转移到宏观系统生态学。

项目成果

期刊论文数量（56）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

Small values in big data: The continuing need for appropriate metadata

大数据中的小价值：对适当元数据的持续需求

DOI：
10.1016/j.ecoinf.2018.03.002
发表时间：
2018-05
期刊：
Ecological Informatics
影响因子：
5.1
作者：
Stow, Craig A.;Webster, Katherine E.;Wagner, Tyler;Lottig, Noah;Soranno, Patricia A.;Cha, YoonKyung
通讯作者：
Cha, YoonKyung

LAGOS‐US RESERVOIR : A database classifying conterminous U.S. lakes 4 ha and larger as natural lakes or reservoir lakes

LAGOS–US RESERVOIR ：将美国 4 公顷及以上连续湖泊分类为天然湖泊或水库湖泊的数据库

DOI：
10.1002/lol2.10299
发表时间：
2023-01
期刊：
Limnology and Oceanography Letters
影响因子：
7.8
作者：
Rodriguez, Lauren K.;Polus, Sam M.;Matuszak, Danielle I.;Domka, Marcella R.;Hanly, Patrick J.;Wang, Qi;Soranno, Patricia A.;Cheruvelil, Kendra S.
通讯作者：
Cheruvelil, Kendra S.

Comparison of total nitrogen data from direct and Kjeldahl‐based approaches in integrated data sets

综合数据集中直接方法和基于凯氏定氮方法的总氮数据比较

DOI：
10.1002/lom3.10338
发表时间：
2019-10
期刊：
Limnology and Oceanography: Methods
影响因子：
0
作者：
Stanley, Emily H.;Rojas‐Salazar, Shirley;Lottig, Noah R.;Schliep, Erin M.;Filstrup, Christopher T.;Collins, Sarah M.
通讯作者：
Collins, Sarah M.

Learning Deep Neural Networks under Agnostic Corrupted Supervision

在不可知的腐败监督下学习深度神经网络