USGS Science⌕ Search

USGS · sir20065283

Ground-Water Nutrient Flux to Coastal Waters and Numerical Simulation of Wastewater Injection at Kihei, Maui, Hawaii

Abstract

Water sampling and numerical modeling were used to estimate ground-water nutrient fluxes in the Kihei area of Maui, where growth of macroalgae (seaweed) on coral reefs raises ecologic concerns and accumulation on beaches has caused odor and removal problems. Fluxes and model results are highly approximate, first-order estimates because very few wells were sampled and there are few field data to constrain model calibration. Ground-water recharge was estimated to be 22.6 Mgal/d (million gallons per day) within a 73-square-mile area having a coastline length of 8 miles or 13 km (kilometers). Nearly all of the recharge discharges at the coast because ground-water withdrawals are small. Another 3.0 Mgal/d of tertiary-treated wastewater effluent is injected into the regional aquifer at a County treatment plant midway along the coast and about a mile from shore. The injection plume is 0.93 miles wide (1.5 km) at the shore, as estimated from a three-dimensional numerical ground-water model. Wastewater injected beneath the brackish ground-water lens rises buoyantly and spreads out at the top of the lens, diverting and mixing with ambient ground water. Ground water discharging from the core of the injection plume is less than 5 years old and is about 60 percent effluent at the shore, according to the model. Dissolved nitrogen and phosphorus concentrations in treated effluent were 7.33 and 1.72 milligrams per liter, roughly 6 and 26 times background concentrations at an upgradient well. Background nitrogen and phosphorus fluxes carried by ground water are 7.7 and 0.44 kg/d-km (kilograms per day per kilometer of coast). Injected wastewater fluxes distributed across the plume width are 55 and 13 kg/d-km nitrogen and phosphorus, roughly 7 and 30 times background flux. However, not all of the injected load reaches coastal waters because nutrients are naturally attenuated in the oxygen-depleted effluent plume. Water from a downgradient well reflects this attenuation and provides a more conservative estimate of injection flux approaching the shore: 27 and 1.5 kg/d-km nitrogen and phosphorus, roughly one-half and one-ninth the injection-source estimates, and 3.5 and 3.4 times background flux. Effluent has 8 O and 2 H stable-isotope signatures that are distinct from local ground water, as well as 15 N and 11 B signatures diagnostic of domestic waste and laundry detergents, respectively. Pharmaceuticals and organic wastewater compounds also were present in effluent and the downgradient well. These isotopes and chemicals served as wastewater tracers in Kihei ground water and may be useful tracers in nearshore marine waters and aquifers elsewhere in Hawaii.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Charles D. Hunt. 2007. Ground-Water Nutrient Flux to Coastal Waters and Numerical Simulation of Wastewater Injection at Kihei, Maui, Hawaii. https://doi.org/10.3133/sir20065283

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

Magnitude and frequency of urban floods in Alabama and Mississippi using data through the 2024 water year

To improve flood-frequency estimates for urban streams in Alabama and Mississippi, the U.S Geological Survey, in cooperation with the Alabama and Mississippi Departments of Transportation, estimated streamflows corresponding to selected annual exceedance probabilities (AEPs) for 64 streamgages on urban streams in the two States—48 in Alabama and 16 in Mississippi. Using estimates from 42 of the 64 streamgages, regression equations were developed that can be used to predict streamflows corresponding to the selected AEPs at ungaged locations on urban streams. Annual peak-flow data through September 2024 were used in the at-site flood-frequency analyses. Drainage area and the percentage of developed land use in the drainage basins of the streamgages (from the 2011 National Land Cover Database) were identified as statistically significant explanatory variables. Pseudo coefficient of determination (pseudo-R2) values ranged from 81 percent for the 0.2-percent AEP (500-year flood) to 91 percent for the 4- and 2-percent AEPs (25- and 50-year floods, respectively). Standard errors of prediction ranged from 23.1 percent for the 4-percent AEP (25-year flood) to 36.1 percent for the 2-percent AEP (500-year flood).

Alabama, Mississippi↗

Phosphorus concentrations and loads in two Mississippi Alluvial Plain watersheds following nutrient reduction efforts during 2010–22

The Mississippi Alluvial Plain ecoregion in Mississippi was historically part of the largest continuous wetland in North America but was clearcut and converted to row crop agriculture beginning in the late 1800s. To reduce the effects of agriculture on surface waters in the Mississippi Alluvial Plain, nutrient reduction best management practices have been implemented throughout the region, particularly in two small watersheds: Harris Bayou watershed (174.04 square kilometers) and Porter Bayou watershed (267.15 square kilometers). Since 2009, the U.S. Geological Survey, in cooperation with the Mississippi Department of Environmental Quality, has collected water-quality samples monthly and during high-flow runoff events, at three locations within each watershed to monitor and assess nutrient trends following best management practice implementation. Here, we have analyzed long-term and seasonal trends in discharge, total nitrogen (TN), and total phosphorus (TP) concentrations and loads using a combination of quantile-Kendall analysis, quantile regression, generalized additive models, and weighted regression on time, discharge, and season. In Harris Bayou watershed, discharge values at a ~0.4–0.5 daily nonexceedance probability increased by 15 percent per year at the uppermost monitoring site. Additionally, TN concentrations decreased throughout the watershed over time, particularly in the spring and winter months. Reductions in TP were less apparent and only observed at one sampling location in Harris Bayou watershed. Nutrient concentrations in Porter Bayou watershed remained consistent over time and increased at one sampling location. However, despite discrepancies in concentration trends, flow-normalized TN and TP loads consistently decreased over time in both watersheds. At the mouth of both watersheds, peak TN and TP loads decreased between 37.7 and 56.9 percent. Collectively, these results provide evidence that best management practices may help reduce the effect of agricultural practices in the region, particularly in Harris Bayou watershed. Additionally, management efforts resulted in greater reductions in TN concentrations than in TP concentrations.

Scientific Investigations Report↗

Streamflow, water quality, and energy development characteristics and water-quality trends for selected sites in the Green River and Muddy Creek drainage basins, southwestern Wyoming, water years 2006–16

Energy resource development in southwestern Wyoming has been rapidly expanding since the early 2000s to include innovative technologies for extracting oil and gas and enhancing renewable energy. In response to concerns about the potential negative effects of energy development on natural resources, the Wyoming Landscape Conservation Initiative was created to implement a long-term, science-based program of assessing, conserving, and enhancing fish and wildlife habitats while facilitating responsible energy development through local collaborations and partnerships. The U.S. Geological Survey, in cooperation with the Bureau of Land Management, began monitoring streamflow and water quality as part of this initiative in 2006 at four U.S. Geological Survey streamgages within the Green River and Muddy Creek drainage basins to characterize the effects of energy development on streamflow and water quality in these areas. To quantify changes in water quality metrics, trend analyses for the 2006–16 water years were calculated and related to drainage basin characteristics including energy development (wells per square mile). Effects of energy development on water quality varied by watershed and the variability was driven primarily by the runoff efficiency and the geochemistry of the drainage basin. The New Fork River water chemistry was a calcium-bicarbonate type with low ionic strength and showed no statistically significant trends in water quality despite increased energy development. The Green River experienced moderate energy development, with significant upward trends in magnesium and alkalinity, indicating low sensitivity to development. Muddy Creek near Dad had minimal energy development and complex hydrogeochemical characteristics, showing mixed trends in water quality constituents. Muddy Creek near Baggs experienced moderate energy development, with significant upward trends in several water quality indicators indicating a response to the land disturbance associated with development. Overall, geochemistry and runoff efficiency influenced the vulnerability of water quality to energy development activities, highlighting the importance of drainage basin physical characteristics on vulnerability to development.

Scientific Investigations Report↗