Understanding Microbial Biofilms in Receiving Waters Impacted by Airport Deicing Activities (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

14 C H a P T E R 3 The findings of the review of available information led to the definition of four testable hypotheses that were proposed to and approved by the ACRP Project 02­32 panel. In this chap­ ter, each hypothesis is described as a hypothesis statement, a discussion of the knowledge gap to be addressed through the testing, the suggested mechanism underlying biofilm growth response in the context of the hypothesis, the practical implications of the hypothesis, the approach to testing the hypothesis, and the data required to accomplish testing. Subsequent chapters of this report present the materials and methods used to collect data (Chapter 4) and the results and conclusions of the testing (Chapter 5). 3.1 Hypothesis 1—Effect of Light on Biofilm Growth 3.1.1 Hypothesis Statement The amount of biofilm growth that occurs under given conditions of readily biodegradable dis­ solved organic matter (i.e., soluble BOD5) is directly proportional to the availability of sunlight. 3.1.2 Knowledge Gap Addressed Reports from airports with prolific biofilm conditions anecdotally report different levels of biofilm accumulation in otherwise apparently similar stream segments that experience different levels of light intensity or shading. Algae and bacteria are always components of stream biofilm communities. Photosynthetic algae are generally positively influenced by increasing light, and many bacteria can be adversely affected by light, which suggests that two opposing factors might be involved. This suggests that light may be an important controlling factor on stream biofilm growth. The limited information available in the literature is inconclusive with respect to this phenomenon. 3.1.3 Suggested Mechanism Biofilm mats contain algae, fungi, water, bacteria, extracellular biopolymers, and particulate matter, including inert biomass and organic particulate matter. Phototrophic organisms, such as algae, use sunlight for energy (by photosynthesis). Biofilm mats typically have a biomass depth that allows development of multiple zones of oxygen reduction, in which substantial anaerobic bacteria may exist. The presence of photosynthesis and anaerobic transformation processes pro­ motes a condition that may be described in one of two ways: • Bacterial films will require more readily biodegradable organic matter than algal biofilms do to produce equivalent biological mass. Testable Hypotheses

Testable Hypotheses 15 • Bacterial films will produce less biomass than algal mats when biochemically transforming equivalent readily biodegradable organic matter. In other words, ambient UV light may promote the development of biofilm mats that have more biomass and extent than would be the case under poor UV light conditions, under which the influence of algae is limited by light availability. 3.1.4 Practical Implications Streambeds that are protected from sunlight (by deep or turbid waters, or shading) may tolerate surface water runoff with a higher BOD5 concentration without exhibiting prolific biofilm growth. 3.1.5 Approach to Testing the Hypothesis The approach to testing involves field, laboratory, and modeling components that address the question: Is the amount of biofilm present under a given set of environmental and nutritional conditions proportional to the availability of sunlight? 3.1.6 Data Requirements Data required to address this hypothesis consist of measurements of biofilm growth under a variety of light conditions. Because organic carbon and UV light are likely rate­limiting factors in the growth of biofilms, availability of BOD5 and a gross indicator of sunlight availability are necessary. A measure of the amount of biofilm is required to ascertain relative abundance. Pref­ erably, quantitative biofilm data will be available to support testing for a statistically significant positive correlation between growth and sunlight exposure. However, qualitative indicators may provide insight in the absence of quantitative measurements. Field The field component of the testing approach requires identifying stream reaches that have the following: • Similar water quality characteristics • Similar physical characteristics • A stormwater runoff component at airports with deicing operations • Evidence of prolific biofilm growth in association with the deicing season • A segment exposed to direct sunlight that displays biofilm growth • One or more segments that are protected from exposure to direct sunlight by natural (e.g., tree canopy) or manmade (e.g., culvert) obstructions and display biofilm growth In practical terms, consecutive reaches in the same stream are most likely to satisfy these criteria. Required field data include water quality, light intensity, and biological parameters through­ out the deicing season and afterward until the biofilm accumulations have died off. Laboratory The required laboratory data consists of measurements of biofilm growth and relative com­ position by holding water quality, water chemistry, and environmental conditions constant, while varying light intensity within the range (400–700 nanometers) of photosynthetically active radiation (PAR).

16 understanding microbial Biofilms in Receiving Waters Impacted by airport Deicing activities Modeling Modeling analyses will be used to complement and add a depth of understanding to observa­ tions made in the field and laboratory by providing a systematic, mathematical reasoning to the observations. The modeled biofilm will consist of the following defined categories: heterotrophic organisms (which consume chemical oxygen demand [COD] and oxygen and are designated as XH), phototrophic organisms (which consume inorganic carbon and oxygen and are designated as XP), autotrophic organisms (which consume ammonia and oxygen and are designated as XN), and inert biomass. The modeling tool will allow the impact that light and dark conditions, and varying concentrations of COD (SS) has on the active biofilm mass (i.e., XH + XP + XN) to be evaluated. The results will then be compared with field and laboratory observations. 3.2 Hypothesis 2—Potential for Phosphorus to Limit Biofilm Growth 3.2.1 Hypothesis Statement Stream biofilms will exhibit phosphorus limitation when the concentration of PO4 is in the range 0.005 to 0.025 mg/L or less. 3.2.2 Knowledge Gap Addressed Phosphorus commonly limits algal and bacterial abundance in aquatic systems, but there is no information available on the potential for phosphorus to limit the growth of biofilms in streams receiving deicer runoff. 3.2.3 Suggested Mechanism Mohamed et al. (1998) conducted experiments to determine the limiting nutrient in associa­ tion with high soluble BOD5 loads from pulp and paper mill effluent in the Fraser River, Brit­ ish Columbia, Canada. In phosphorus­limited systems, it was determined adding phosphorus increased biofilm growth and reduced the amount of extracellular polymeric substances in the biofilm. A high concentration of extracellular polymeric substances is indicative of nutrient­ deficient biofilms (Romani and Sabater 2000). The threshold of phosphorus availability is dependent on water conditions. Lock and John (1979) and Blenkinsopp and Lock (1994) found that biofilms were influenced by phosphorus availability under flow pattern and storm flow conditions, respectively. As a result, a phosphorus­deficient flowing water body may have a different biofilm response to a given BOD5 concentration than a similar system that has excess phosphorus. Approximate macronutrient requirements of 12 grams (g) nitrogen and 2.3 g phosphorus are needed per 100 g of cell biomass to avoid being the rate­limiting macronutrient (Metcalf and Eddy 2003). Thus, extent of cell colonization on stream surfaces may affect the condition of phosphorus limitation (i.e., as organic carbon is added to a stream system and cell growth occurs, phosphorus may transition from a non­limiting state to a limiting state as demand increases). 3.2.4 Practical Implications Phosphorus deficient streams may exhibit less prolific biofilm accumulations under a given BOD loading than streams in which biofilm growth is limited by organic carbon or some other factor. Therefore, biofilm suppression may be a side benefit of stormwater P­reduction efforts, or other factors that promote P deficiency.

Testable Hypotheses 17 3.2.5 Approach to Testing the Hypothesis The approach to testing focuses on laboratory and modeling components. Field investiga­ tions are not practical because they require stream segments that are similar in terms of physical and hydrological characteristics, deicer loading and concentrations, and other environmental factors; but differ in terms of being phosphorus limited, or having excessive phosphorus. It is unlikely that such segments can be identified. The laboratory experimental design examines the response of biofilms to phosphorus limitation in controlled systems designed to simulate related stream conditions and exposed to variable phosphorus and organic carbon (amongst other variables) levels. Existing math­ ematical models will then be used to confirm and provide insights into the laboratory observations. The resulting data will support testing for correlation between growth and PO4 concentration. 3.2.6 Data Requirements The primary data requirement for testing this hypothesis is measurement of biofilm growth in the presence and absence of phosphorus limiting conditions. Preferably, the bio­ film data will be quantitative and support testing for a statistical significance of apparent limitation. However, qualitative indicators may provide insight in the absence of quantitative expressions. Laboratory Laboratory data are required to provide an understanding of the bacterial capability to alter their stoichiometric phosphorus requirement, and a basis for evaluating the potential for phos­ phorus to be a rate­limiting macronutrient against a substantial population shift. The required data consist of measurements of biofilm growth and relative composition in laboratory reactors under conditions that generally replicate field conditions, but with phosphorus present in vary­ ing concentrations. Modeling Similar to the model applied in testing Hypothesis 1, modeling analyses will be conducted to complement and add a depth of understanding to observations made in the laboratory by providing a systematic, mathematical reasoning to the observations. The modeling tool will allow the impact that varying PO4 concentrations on the active biofilm mass (i.e., XH + XP + XN) to be evaluated. The evaluation will then be compared with field and laboratory observations. 3.3 Hypothesis 3—Impact of Physical Stream Characteristics on Biofilm Growth 3.3.1 Hypothesis Statement The physical characteristics of receiving streams influence the extent of biofilm accumulation for a given water chemistry condition as follows: • Category 1: Shallow, turbulent, well­mixed channels promote biofilm growth and require ambient BOD5 concentrations less than 50 mg/L to avoid prolific biofilm growth. – streambed surface area (ABED) to bulk liquid volume (VB) ratio (ABED:VB) greater than 100 square meters per cubic meter (m2/m3) (1 cm depth)

18 understanding microbial Biofilms in Receiving Waters Impacted by airport Deicing activities • Category 2: Relatively straight channels with moderate depth and flow promote moderate biofilm growth and require ambient BOD5 concentrations be maintained in the range 50 to 100 mg/L or lower to avoid prolific biofilm growth. – streambed surface area (ABED) to bulk liquid volume (VB) ratio (ABED:VB) in the range 25 to 75 m2/m3 (1.3 to 4 cm depth) • Category 3: Deep, slow­moving channels deter biofilm growth, and can experience ambient BOD5 concentrations greater than 100 mg/L without prolific biofilm growth. – streambed surface area (ABED) to bulk liquid volume (VB) ratio (ABED:VB) less than 10 m 2/m3 (10 cm depth) 3.3.2 Knowledge Gap Addressed There are anecdotal reports from airports that biofilm growth associated with deicing runoff varies with the physical characteristics within affected stream segments, but there is no clear understanding of the relationship between stream geomorphology and susceptibility to prolific biofilm growth. 3.3.3 Suggested Mechanism Available literature and understanding of biofilm processes suggest that several stream physical characteristics have the potential to influence biofilm proliferation within a given level of substrate2 and macronutrient availability. Shallow, fast­moving areas will tend to be well­oxygenated, have a low cross­sectional area to perimeter ratio which exposes attached biofilms to a large proportion of the dissolved nutrients in the water column, and causes higher shear stresses on the biofilm. Increased shear has been shown to influence biofilm structure, resulting in denser and more tenaciously attached biofilms (Buckingham­Meyer, Goeres, and Hamilton, 2007). In contrast, deeper, slow­moving stream segments tend to be less well­oxygenated, have a smaller proportion of the water column in close contact with the stream bottom, and are associated with less dense biofilms with a thicker mass transfer boundary layer above the biofilm, resulting in slower diffusive transport of nutrients into the biofilm (Battin et al. 2003). 3.3.4 Practical Implications Knowledge of how physical stream channel factors affect the abundance or absence of prolific biofilm accumulations will give airport operators insight as to why prolific biofilms occur in some stream segments and not others, as well as in designing stormwater conveyance channels and other infrastructure with features less likely to promote prolific biofilm growth. 3.3.5 Approach to Testing the Hypothesis Testing of this hypothesis relies primarily on collection and analysis of field data, supple­ mented by use of modeling tools to examine relative influence of factors through sensitivity analyses. Preferably, the biofilm data will be quantitative and support testing for a statistical significance of stream channel factors. However, qualitative indicators may provide insight in the absence of quantitative expressions. 2For the purposes of this research project, the following biochemical definition of substrate is used: the substance acted upon by an enzyme.

Testable Hypotheses 19 3.3.6 Data Requirements Required data consists of biofilm growth measurements in adjacent stream segments with sim­ ilar deicer and other pollutant loading characteristics but with different physical configurations. 3.4 Hypothesis 4—Identifying Nutrients Potentially Limiting to Biofilm Growth 3.4.1 Hypothesis Statement Ratios of bulk­phase substrate and macronutrient concentrations can be used to identify the rate­limiting substrate for biofilm growth in a stream. 3.4.2 Knowledge Gap Addressed There is currently no general indicator of a stream’s biochemical susceptibility to prolific biofilm growth as a function of limiting macronutrient (i.e., phosphorus and nitrogen) avail­ ability. This gap may be filled by evaluating the growth of biofilms under different conditions of relative abundance of substrate and macronutrient availability to determine which of these factors is likely to limit biofilm growth. 3.4.3 Suggested Mechanism Biofilm growth in flowing stream environments is limited by a single terminal substrate (the electron donor or electron acceptor) or a macronutrient. The idea of a limiting nutrient or sub­ strate is well established in the biodegradation literature. In systems with low levels of organic carbon, the organic carbon electron donor is often the limiting factor. Where organic carbon is more abundant, another factor, such as one of the macronutrients (N or P), may be limiting. Streams receiving deicer fluid runoff are periodically dosed with higher levels of organic carbon and therefore may alternate between conditions of organic carbon limitation versus limitation by some other nutrient. 3.4.4 Practical Implications Identifying the rate­limiting substrate or macronutrient under different relative conditions will provide a basis for the airport operator to judge a stream’s likely predisposition to exhibit more or less prolific biofilm growth under a given set of physical conditions and BOD loading. This information will provide an understanding of why biofilms are more prolific in one stream than in another, and insight into nutrient control strategies for affecting the potential for biofilm growth. 3.4.5 Approach to Testing the Hypothesis The effect of macronutrient availability will be evaluated through laboratory growth experi­ ments, observing biofilm growth under various conditions of relative availability of phosphorus and nitrogen. Field collection of phosphorus samples in conjunction with monitoring to sup­ port Hypothesis 2 will provide an indication of nutrient limiting conditions in streams receiving deicing runoff and experiencing various levels of biofilm growth. Similar to the model used for testing Hypothesis 2, modeling analyses will be used to comple­ ment and add a depth of understanding to observations made in the field and laboratory. The

20 understanding microbial Biofilms in Receiving Waters Impacted by airport Deicing activities modeling tool will allow the impact that varying ammonium concentrations has on the active biofilm mass (i.e., XH + XP + XN) to be evaluated. The results will then be compared with field and laboratory observations. 3.4.6 Data Requirements Data requirements for testing this hypothesis consist of laboratory experiments and applica­ tion of modeling tools. Laboratory experiments are required to assess the interactions of organic carbon and essential macronutrients. Changes in measured biofilm thickness allow assessment of the effects of varying each of the macronutrient parameters individually.