Pennsylvania Numeric Threshold Criteria Development for Lakes and Ponds

Princeton Hydro was contracted by the Pennsylvania Department of Environmental Protection (PADEP) to assist in the development of nutrient thresholds for lakes, reservoirs, and ponds within the Commonwealth of Pennsylvania. This project was conducted through the Borough of Harveys Lake, Luzerne County, PA, since Princeton Hydro has worked with the Borough to implement a variety of watershed measures to improve the water quality of Harveys Lake. Our efforts working with the Borough have resulted in a substantial reduction in the total phosphorus (TP) concentrations, which allowed PADEP to take Harveys Lake off the impaired list. Thus, Harveys Lake was used as a “template” of a restored lake for the Nutrient Criteria analysis.

Nutrient concentrations vary greatly between waterbodies based on numerous factors including lake size, morphology, lake origin, geographic location, surrounding land use, soil type, flushing rate, other hydrologic processes, etc. PADEP provided Princeton Hydro with their state lakes water quality database, which spanned from 1985 to 2017. The variables originally used in the analysis were total phosphorus (TP), total nitrogen (TN), Secchi depth and chlorophyll-a (Chl-a). Eventually, TN was dropped as being a nutrient of concern for this analysis since most of the waterbodies around the state are phosphorus limited and since increases in phosphorus loading and concentrations typically lead to more cyanobacteria relative to more favorable types of algae (e.g. green algae, diatoms). Thus, the focus on the Nutrient Threshold analysis was determined to be TP.

In order to categorize these waterbodies in a logical manner, Princeton Hydro ran two separate permutation multivariate analyses of variance (perMANOVA) on a select features to determine the most effective grouping of the state’s waterbodies. The first analysis included a 4-way segmentation based on lake formation (natural vs. artificial) and depth (≤3.5 m, >3.5 m), and was labeled as “lake type”. The second analysis included a 3-way segmentation based on major ecoregion (Glaciated and Southwest Appalachian Plateau; Pocono, Central Mountains, and Laurel Highlands; and Piedmont and Lowlands) and was labeled “ecoregion”. Later, a more refined Principal Component Analysis was completed, which revealed a strong relationship between TP and Chl-a. It also showed that deeper lakes (maximum depth of at least 3.5 meters), particularly glacial lakes that thermally stratify, are more sensitive to TP relative to shallow waterbodies. Thus, a mean growing season TP concentration criterion for deeper lakes was recommended at 0.02 mg/L, while the mean growing season TP criterion for shallow lakes was recommended to be 0.03 mg/L.