Testimony to Senate Committee on the Environment and Public Works – 5 August 2002

Prof. Robert T. Heath, Ph.D.

Water Resources Research Institute and Dept. Biological Sciences

Kent State University, Kent OH 44242-0001

History of the problem of anoxia in Lake Erie: Anoxia in the bottom waters of Lake Erie has been observed since 1930 (Figure 1 from Bolsenga and Herdendorf 1993). Originally it was constrained to the Sandusky subbasin, the region of the lake north of Huron, between Sandusky and Lorain. As eutrophication of the lake increased in the 1960s and early 1970s the region of the lake that became anoxic in the summer spread to cover substantial portions of the sediments of the central basin of the lake.

Eutrophication of the lake was caused by excessive inputs of nutrients from human activities including sewage, industrial processes and agricultural fertilizers. High concentrations of nutrients in turn stimulated growth of noxious forms of phytoplankton (algae suspended in the water). These noxious phytoplankton (such as Microcystis) put compounds into the water that are distasteful and may be harmful to humans, thereby diminishing the quality of the water for fish and birds and for human consumption. These noxious phytoplankton also were inefficiently grazed by zooplankton, so the carbon fixed by photosynthesis of these phytoplankton was not moved efficiently through the base of the food web to higher organisms, such as fish and birds. Although these algae fixed large quantities of energy, it was wasted instead of supporting a healthy food chain. When phytoplankton died they sank to the lower reaches of the lake and were decomposed by bacteria that are natural components of the food web. The bacterial metabolic decomposition processes required oxygen, consuming all oxygen available, in turn leading to oxygen depletion in the lower waters of the lake. The oxygen in the lower waters is replaced only through circulation of the bottom waters with the oxygenated surface waters. Circulation is constrained because of the thermal stratification of the lake in the summer. Typically, complete re-circulation of the water column doesn’t occur until the autumn and the decline of thermal stratification.

Mandated constraint of inputs of nutrients to the lake in the 1980s succeeded in reversing the eutrophication process. The essential nutrient in the least relative supply was identified as phosphorus (P). Limiting the input of P to Lake Erie was seen as the most efficient means of limiting growth of noxious phytoplankton. As the concentrations of P in forms readily available to algae and bacteria declined, the abundance of noxious phytoplankton declined and were replaced by species of phytoplankton that were efficiently grazed and did not diminish water quality with noxious exudates.

The reclamation of Lake Erie’s water quality and its food web from the eutrophic conditions that existed 30 years ago is one of the major successes in large-scale ecosystem management. That success is now threatened by increases in phytoplankton production, return of some of the noxious phytoplankton species, and by an increase in the area of the lake covered by anoxia in the late summer. The cause of this is uncertain.

Figure 1. Anoxic regions of Lake Erie from 1930 – 1982. Shaded regions indicate anoxic regions detected in summer. From Bolsenga and Herdendorf 1993.

What is different this summer? For the past decade my research group has investigated the structure and function of the base of the food web, both under the influence of zebra mussels and in their absence. We have focused on the uptake and transport of carbon (C) and P because of the significance of these elements to the ecosystem function.

The base of the food web is comprised of phytoplankton, zooplankton (micro-crustaceans, rotifers and protists) and bacterioplankton. The bacterioplankton are a large number of species of non-pathogenic bacteria that are essential for performance of many ecosystem functions. The base of the food web is an interplay between growth of phytoplankton and bacterioplankton. The movement of energy and materials through the base of the food web can take two major pathways: 1) phytoplankton can be grazed directly by microcrustaceans or 2) dissolved photosynthate released by phytoplankton can support bacterial growth and a microbial food web. We have shown that the relative importance of these pathways is not constant in Lake Erie. The direct grazing pathway is most important in coastal regions of the lake and the microbial food web becomes relatively more important in offshore and oligotrophic regions. As part of our research we have studied several sites that include the portion of the lake that most frequently became anoxic, the Sandusky subbasin (SSB).

Here I describe in brief our findings from the past two years in the SSB and compare them with our findings from two cruises in July 2002. The observations that I present here were taken from a station in the SSB near the international boundary:

LAT 41^o 40’ LON 82^o 30’

Depth profiles of dissolved oxygen at this site are shown in Figure 2. Dissolved oxygen concentration was determined potentiometrically with a Hydrolab multi-parameter data sonde, calibrated within 24 hours of the observation. Oxygen depletion in the bottom waters at this station is not unique to this summer.

2002

2001

2000

Dissolved Oxygen Concentration (mg L^-1)

Figure 2. Depth profiles of oxygen concentrations during July 2000, 2001, 2002 at Sandusky Subbasin Station.

Following collection of physical variables on-site, water samples were collected and returned to the Biochemical Limnology Laboratory at Kent State University where we examined the status of the base of the food web. Our observations are summarized in Table 1. We also provide a comparison with the past two years and note observations that are statistically and scientifically significant in bold type.

Water transparency, estimated by the maximum depth at which a 20 cm white plate can be discerned – the Secchi depth – is significantly lower this summer than in the recent past indicating a significant decrease in the transparency of the water. We also observed a significant increase in chlorophyll content in samples over past years. This increase in chlorophyll supported photosynthesis. The “health” of the algae is indicated by the photosynthetic potential and the optimum photosynthetic rate, scaled for unit amount of chlorophyll. The observations indicate that the algae are growing actively; their photosynthetic capabilities do not appear to be limited by nutrient availability. Consistent with this is seen a significant increase in the amount of P in algal particles. When algae are in nutrient rich waters they store excess amounts of P in their tissues as insurance against nutrient limitation at a later time.

We also observe large amounts of P in bacterial particles. Bacteria growing actively increase their amount of P by increasing the amount of RNA, an essential biochemical necessary for protein synthesis and active growth. Active bacteria, in general, increase in size and rate of incorporation of dissolved carbon compounds. We observed that bacteria this year were significantly larger, and incorporated significantly more dissolved leucine (a dissolved biochemical compound we use to test their growth rate). Our observations are consistent with the view that bacterioplankton in Lake Erie are growing significantly faster than in years past – at least at the site and times we have investigated. Available-P, estimated both with a bioassay and by chemical means, does not appear to differ significantly this summer vs. previous summers. The amount of dissolved organic P (DOP) is significantly increased and the total P is significantly increased.

Our observations indicate that phytoplankton remain P-limited and susceptible to management plans devised around the assumption that they are P-limited. The phosphate turnover times of about 30 minutes indicate that the plankton community is P-limited but not severely so. If it were severely P-limited, many plankton would be capable of producing large amounts of alkaline phosphatase to obtain available P from certain DOP compounds. Alkaline phosphatase is detected fluorometrically by the hydrolysis of methyl-umbelliferyl phosphate (MUP). The rate of MUP hydrolysis can be used to detect P-limitation, high MUP rates indicate severe P-limitation. The rates of MUP hydrolysis were modest, indicating that the community is not severely P-limited.

These findings (based on VERY LIMITED OBSERVATIONS) are consistent with the view that phytoplankton and bacterioplankton – the base of the food web – are more abundant and active this year in Lake Erie than in the recent past. We observe significant increases in the amount of P as dissolved organic P, forms of P available for phytoplankton growth only under certain conditions. The sources of additional DOP and the stimulation of plankton growth are unknown and a matter of concern and conjecture.

Table 1. Observations in Sandusky Subbasin during July 2000, 2001, and 2002.

Item	Meaning	July 2000	July 2001	July 2002
Secchi Depth (cm)	Transparency	455±13	607±0.02	163±12
Chlorophyll a (ug/L)	Est. algal abundance	1.0±0	---	2.8±0.3
Popt	Opt. PS per unit chl.	2.93	---	2.12
Photosynthetic Potential	Est. of algal Physiological status	0.026	---	0.0096
Algal-PP (nM)	Algal mass est.	69±11	---	697±42
Bacterial-PP (nM)	Bacterial mass est.	54±4	---	488±50
Bacterial size (µm^3)	Bacterial Size	0.067±0.005	---	0.18±0.02
Bacterial # (´ 10^5 cells/ml)	Bacterial abundance	8.30±0.42	11.99±1.55	2.77±0.29
Bacterial Production (´ 10^-5 µgC/ml/hr)	Rate of bacterial production	12.1±3.5	14.1±1.0	148 ± 3
Bacterial Respiration (µgC/ml/hr)		---	---	281±4
Bacterial Growth Efficiency (%)		---	---	34
BacAP (nM)	Bio. est. P–available	5.6±1.4	---	3.3 ± 0.5
SRP (nM)	Chem. est. P-available	100±50	undetectable	118 ± 7
DOP (nM)	Dis. Organic P	590 ± 40	200 ± 40	6962 ± 150
TP (nM)	Total P	756±100	330±50	8265±242
P turnover time (min)	Est. of P-limitation	30.1±0.8	34.6±3.1	26.8±3.7
MUP (nM/hr)	Est. of P-limitation Via enzyme activity	43 ± 2	36 ± 2	24.2 ± 2.7

Bold: values taken in July 2002 that differ significantly from measurements made at similar times in 2000 and 2001 in same place of Lake Erie: Sandusky subbasin (SSB).

Implications for increased regions of anoxia: Given these observations, I believe a likely explanation for increased regions of anoxia in Lake Erie is increased production at the base of the food web. If these increased amounts of phytoplankton are incompletely grazed, they could sink to the lower regions of the lake on death and be decomposed by natural non-pathogenic bacteria that consume oxygen to depletion. I should like to emphasize that this explanation is not the only possible explanation; it is the one I regard as the most likely explanation. Natural geochemical and biological processes can also consume oxygen. Oxygen consumption by these natural processes is normally replenished by entrainment of oxygenated waters during storms. During unusually long periods of stagnation, oxygen can be depleted from bottom waters without extraordinary production occurring in the surface waters.

Possible Causes of increased phytoplankton and bacterioplankton growth: Because the phytoplankton appear to be P-limited (although weakly), I believe we need to examine possible sources of P and the processes by which it can be supplied at a rate to support increased phytoplankton growth.

External loading of P comes from the watershed but external to the lake. Such external sources can come from identifiable points (point-source loading: sewage treatment plants, combined sewer overflows, industrial effluents, etc.). Regulation of point source loading is strict and generally works well to control unwanted excessive inputs of P to the lake. Alternatively, non-point sources of P-loading from sources such as agricultural and residential runoff of fertilizers is not well regulated nor easily monitored because of its diffuse nature.

Internal loading of P is a term applied to processes that recycle P already in the lake from unavailable forms to available forms of P (e.g. inorganic orthophosphate). P is unavailable for growth of phytoplankton and bacteria when it is sorbed to sediments, when it is in dissolved organic P compounds (DOP), or when it is incorporated into living or dead organic particles. P sorbed to sediment surfaces can be released when the oxygen concentrations decline below 0.4 ppm. This means that when oxygen is depleted from waters immediately above the sediment surface, P can be released in a useful form by desorption, potentially further stimulating the growth of P-limited phytoplankton. P can also be released in useful forms through the action of certain enzymes capable of hydrolyzing specific DOP compounds (Francko and Heath 1979) or through photolysis of DOP compounds by UV light capable of penetrating several meters into clear lake water (Cotner and Heath 1990). High temperatures of the lake water can increase the activity of hydrolytic enzymes acting on enzyme-sensitive DOP; clear water and increased intensity of UV light can increase the rate of photolysis of UV-sensitive DOP.

Organisms grazing on particulate organic matter (e.g. living or dead tissue material) release P in available and unavailable dissolved forms. Increased grazing activities by zebra mussels and their congener, quagga mussels, may be a source of increased P-availability. My research over the past several years has shown that zebra mussels release sufficient available P to relieve phytoplankton in surrounding waters from P-limitation (Heath et al. 1995). Their effect on the whole lake community remains unclear (Heath et al. 2000), although they may exacerbate blooms of the nuisance cyanobacterium, Microcystis (Culver et al. 1999).

Conclusions and Specific Recommendations: It is not clear why the zone of anoxia has apparently begun to expand after at least a decade of being confined to small regions of the central basin of Lake Erie. With apparent increases in phytoplankton abundance, it is tempting to reminisce about the causes of large regions of anoxia observed during the 1960s and early 1970s. Anoxia of those days resulted from eutrophication due to excessive external loading of available P from point sources. Because of the great restrictions on point-source P-loading, it is unlikely that the current problems arise in the same way. The role of other sources needs to be investigated. It is unclear whether external non-point source loading from urban and agricultural sites or internal loading due to zebra- and quagga mussels or a combination of external and internal sources are capable of causing the problems currently observed.

I don’t believe we need new research of the issues involved as much as we need new ways of placing current research into a more useful context.

(1) Research on ecosystem level effects of P-loading and the possible effects of dreissenid mussels need to be placed into comprehensive models useful for ecosystem management. The “Great Lakes Modeling Summit: Focus on Lake Erie” (IJC 2000) is an excellent point of departure for this purpose.

(2) Scientific research on the Great Lakes needs to move beyond its current ad hoc status by incorporating continuous comprehensive monitoring activities at levels far expanded beyond current efforts.

(3) The Great Lakes need to be valued as national (indeed, an international) treasures rather than being viewed as regional resources alone. Issues besetting the Great Lakes need to be addressed in innovative bi-national ecosystem research, monitoring and management programs.

References:

Bolsenga, S. J. and C. E. Herdendorf. 1993. Lake Erie and Lake St. Clair Handbook. Wayne State Univ. Press. (Detroit). x + 467 pp.

Cotner, J. B., Jr. and R. T. Heath. 1990. Iron redox effects on photosensitive phosphorus release from dissolved humic materials. Limnol and Oceanogr. 35: 1175 - 1181.

Culver, D. A, D.B. Baker, R.P. Richards, A.M. Beeton, T.H. Johengen, G.A Leshkevich, H.A. Vanderploeg, J.W.Budd, W. W. Carmichael, R.T. Heath, C. E. Wickstrom, H.J. MacIsaac, and L.Wu. 1999. "Toxicity, ecological impact, monitoring, causes and public awareness of Microcystis blooms in Lake Erie." Final Report to the Lake Erie Commission. 57 pp + 12 tables + 78 figures. 31 March 1999.

Francko, D.A. and R.T. Heath. 1979. Functionally distinct classes of complex phosphorus compounds in lake water. Limnol and Oceanogr. 24 (3):463-473.

Heath, R.T. 1986. Dissolved organic phosphorus compounds: do they satisfy planktonic phosphate demand in summer? Can. J. Fisheries Aquat. Sci. 43 (2):343-350.

Heath, R.T., G. Fahnenstiel, W.S.Gardner, J.F. Cavaletto and S-J Hwang. 1995. Ecosystem level effects of zebra mussels (Dreissena polymorpha): An enclosure experiment in Saginaw Bay. J. Great Lakes Research 21: 501 - 516.

Heath, R.T., X. Gao, H. Wang, and V. Mattson. 2000. Influence of zebra mussels on phytoplankton photosynthesis in Lake Erie. Ohio Journal of Science 100: A-43.

Int. Joint Com. 2000. Great Lakes Modeling Summit: Focus on Lake Erie. ISBN 1-894280-17-2.

Acknowledgements: This work was supported by grants from the Lake Erie Protection Fund (98-09) and Ohio Sea Grant (R/ZM-25). I thank Dr. Xueqing Gao and Tracey Trzebuckowski for help in preparing this report and Carla Skytta and Laurie White for technical assistance.