OCE-PRF Predicting OMZ and seasonal microbial activity from community structure using machine learning and novel measurements of ATP turnover

点击翻译按钮获取中文摘要

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).Marine microbes, including single-celled bacteria, archaea, and eukarya, are the main drivers of carbon and energy transformations in the ocean. The efficiency of the microbial loop at transforming dissolved nutrients into biomass and up the food web is defined by the individual activities of the microbial community. Unfortunately, current methods for characterizing microbial activities in the marine environment are biased, with most direct measurements focusing on heterotrophic microbes that respire carbon dioxide in oxygenated waters. These measurements exclude a large diversity of microbes that play major roles in ocean nutrient cycles, including photosynthesizes and chemotrophs as well as microbes in low oxygen, anoxic waters. This project will measure the activity of microbial organisms through novel applications of the adenylate energy system, which is used by all organisms and in almost all metabolic pathways. This work will focus on sites in the California Current Ecosystem with a wide range of oceanographic conditions and variable microbial communities. These locations include low oxygen canyons representative of oxygen minimum zones (OMZs), upwelling locations reflecting eastern boundary currents, and temporally dynamic coastal sites. Within the planned sampling, support will be allocated for undergraduate students to join shipboard efforts to help with seawater sampling and gain experience conducting oceanographic research. The datasets produced from these sampling efforts will be used to develop predictive models of microbial metabolism (activity) that can be applied to other oceanographic regions, based on community structure, abundance, and geochemical parameters across time and environmental concentrations.The highly productive California Current Ecosystem is subjected to seasonal successions of phytoplankton communities, with diatom and dinoflagellate dominance coinciding with upwelling. The ability of these phytoplankton groups to convert nutrients into biomass have impacts on primary productivity, biogeochemical cycles, and the microbial loop. Hypoxic environments shift the microbial community towards anaerobes and chemotrophs, which have unique metabolisms and differential efficiencies converting acquired carbon sources into biomass and energy. Therefore, the microbial loop efficiency in OMZs is likely to be directly impacted by microbial community structure, but precisely defining these differences in microbial activity has been hampered by biased sampling techniques. This project will address the hypotheses that microbial activity, in terms of energy storage, metabolic rates, and growth rates will be directly liked to and predicted by phytoplankton community structure and abundances in seasonal upwelling systems. Additionally, microbial activities in OMZs are linked directly to the implementation of anaerobic metabolic strategies and can thus be predicted by changes in community structure. Microbial activity will be characterized using high throughput applications of novel measures of the energy charge and adenosine triphosphate (ATP) turnover rates while the microbial community structure will be defined with 16S and 18S rRNA gene amplicon sequencing. Two dedicated, single day cruises to the La Jolla and Scripps Canyons located within the California Bight, and containing known OMZs will be conducted. Additionally, microbial activity measurements will be incorporated into an on-going time-series that is sampled for microbial community structure twice-a-week at the Scripps Institute of Oceanography pier in La Jolla, CA. Measurements from the cruises and time-series will be used to develop predictive models of microbial metabolism based on community structure, abundance, and geochemical parameters across time and within known OMZs using self-organizing maps (SOMs) and random forest regression. Therefore, this project will produce predictive models that can be generalized to broader ocean regions in addition to datasets with unprecedented resolution of natural microbial activity across time and environmental conditions.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.