In both terrestrial and aquatic ecosystems, the continuous cycles of production and decomposition of organic matter are the principal processes that determine the balance of organic carbon, nutrients, carbon dioxide, and DO in the biosphere. Plants (autotrophs) use solar energy, carbon dioxide, and inorganic nutrients to produce new organic matter and, in the process, produce DO by photosynthesis. Bacteria and animals (heterotrophs) use the organic matter as an energy source (food) for respiration and decomposition, and in these processes, consume DO, liberate carbon dioxide, and recycle organic matter back into the ecosystem as simpler inorganic nutrients. Water quality problems, such as depleted levels of DO, nutrient enrichment, and eutrophi-cation (overproduction of aquatic plants), occur when the aquatic cycle of production and decomposition of organic matter becomes unbalanced from excessive amounts of anthropogenic inputs of organic carbon and inorganic nutrients from wastewater discharges and land use-influenced watershed runoff.
DO is the most meaningful and direct signal relating municipal and industrial discharges of organic matter to downstream water quality responses over a wide range of temporal and spatial scales. In addition to DO's significance as a measure of aquatic ecosystem health, there are several other practical reasons for choosing DO as the signal for assessing changes in water quality, including the following:
• Historical records go as far back as the early twentieth century for many major waterbodies. New York City, for example, began monitoring DO in New York Harbor in 1909 and records exist for the Upper Mississippi River beginning in 1926, for the Potomac estuary in 1938, and for the Willamette River in 1929 (see Wolman, 1971).
• Basic testing procedures for measuring DO have introduced few biases over the past 90 years, thereby providing the analytical consistency needed for comparing historical and modern data (Wolman, 1971).
This section provides background on sources of DO data, the relationship between BOD loading, downstream DO levels, and the two key physical conditions (high temperature and low flow) that create "worst-case" DO conditions. As will be explained, DO data collected under worst-case conditions inherently contain the sharpest signal of the point source discharge/downstream DO relationship.
Key to this analysis is the existence of DO data with which a before- and after-CWA comparison can be made. Fortunately, systematic water pollution surveillance of many of the nation's waterways began in 1957 in response to the 1956 Amendments to the Federal Water Pollution Control Act. Figure 3-1 is a map, developed by Gun-nerson (1966), displaying minimum DO concentrations throughout the United States using data collected from 1957 through 1965. It illustrates both the spatial extent of historical data and the poor DO conditions found in many of the nation's waterways in the late 1950s and early 1960s.
These and more recent water quality data collected by state, federal, and local agencies are in USEPA's STORET database and are available for a before- and after-CWA comparison (Gunnerson, 1966; Ackerman et al., 1970; Wolman, 1971; USEPA, 1974). Currently, the system holds over 150 million testing results from more than 735,000 sampling stations, about 4.6 million of which are DO observations recorded from 1941 to 1995 (Figure 3-2). The challenge was to figure out how to mine STORET's mountain of DO data and create before- and after-CWA data sets that inherently contain the best response "signal" linking point source discharges with downstream DO. This task is not unlike panning for gold. What was needed was a series of screens to divert away all the "rubble and debris" (noisy data), leaving a clean set of "nuggets" (signal data). Using a systematic comparison of before- and after-CWA water quality data sets, the national policy for technology- and water quality-based effluent controls can be considered a success if downstream waterways with poor water quality before the CWA can be shown to have improved significantly after the CWA.
"Worst-Case" Conditions as a Screening Tool
The first step in developing the before- and after-CWA data sets was to analyze the relationship between point source BOD loading and downstream DO levels. As the reader will see in Section B, the rules subsequently adopted and applied to screen out
noisy data were based on eliminating hydrologic and other physical factors that interfered with, or confounded, the point source discharge/downstream DO signal (see Hines et al., 1976). As it turned out, the DO data that contained the strongest signal were the data collected under conditions that yielded the lowest DO levels (high water temperature and low flow). The purpose of this subsection is to explain the physical processes and spatial characteristics that make worst-case conditions the appropriate screening tool for developing the before- and after-CWA data sets.
Worst-Case Conditions from a Temporal Perspective In an unpolluted stream, DO concentrations in most of the water column are typically at or near saturation. Saturation, however, varies inversely with water temperature and elevation. At typical winter water temperatures of about 10° C, the solubility of oxygen is about 11.3 mg/L at sea level. At a higher summer temperature of 25° C, the solubility is only about 8.2 mg/L. This high water temperature-low solubility relationship makes hot weather an especially critical period for aquatic organism survival. Higher water temperatures mean a lower reserve of oxygen is available to buffer against any additional oxygen demands made by wastewater effluent discharges.
Wastewater effluent typically has an oxygen deficit (a DO concentration below saturation). Therefore, its initial entry into a waterway causes an immediate drop in stream DO near the outfall. The effluent becomes diluted as it mixes with the stream water and flows down the channel. The BOD of the stream water thus becomes the discharge-weighted average BOD of the effluent and the stream above the discharge. The volume of streamflow (the dilution factor), therefore, is a critical variable in de-
Was this article helpful?