Managed Nitrogen Load Decrease Reduces Chlorophyll and Hypoxia in Warming Temperate Urban Estuary

Abstract

Many urban estuaries worldwide suffer from excess phytoplankton and hypoxia (low oxygen) due to high nutrient loads. A common water quality management strategy is to require wastewater treatment facility upgrades. This case study examines Narragansett Bay, a warming temperate mid-latitude urban estuary with seasonal periodic hypoxia, during June through September from 2005 to 2019. Within this period, numerous facilities were upgraded to nitrogen removal over several years. The response of the bay is more consistent with “textbook” expectations for reduced chlorophyll and hypoxia than what was seen in many other systems—despite its complex coastline geometry, numerous river inputs, and widely-distributed treatment facilities. River flow drives inter-annual variability with increased load, density stratification, chlorophyll, and hypoxia in wet years. Mean 2013-2019 bay-wide total nitrogen load was 34% less than the 2005-2012 mean, a reduction of about 106 kg yr-1, comparable to the range of flow-driven inter-annual variations. Chlorophyll Index and Hypoxia Index event-based metrics applied to high-frequency time series observations at eight sites quantify exceedances of severe and moderate thresholds. Relatively steady 33% and 16% Chlorophyll Index declines, for severe and moderate thresholds, occurred from about 2007 to 2019. The Hypoxia Index declined markedly by 2009 and 2014 for severe and moderate thresholds, respectively, and remained at or near zero from 2014 to 2019. The load reduction explains chlorophyll and hypoxia declines better than physical processes including river flow, stratification, tidal variations, winds, sea level differences, and temperatures. River flow about 55% higher than the 2005-2019 mean would increase non-treatment facility loads by an amount comparable to the managed load decrease, so future wet summers could partially reverse the improvements. Long-term trends include warming of about 0.5°C decade-1, which reduces oxygen saturation by 0.1 mg l-1 decade-1. This rate is likely a lower bound for temperature-driven oxygen decreases, because warming can also accelerate phytoplankton growth and bacterial consumption. Without warming, the managed load decrease would have curtailed hypoxia more effectively. Climate trends should be at least as important to future eutrophication as the managed load decline because, in addition to warming influences, long-term increases in river flow would increase load and stratification.