P-limitation is thought to be much more characteristic of freshwater systems than marine systems making the results of Sañudo-Wilhelmy et al. (2004, Nature 432: 897-901) of great relevance to our understanding of P dynamics in lakes.  We propose to coordinate an effort among limnologists to establish an unprecedented dataset of C:N:P measurements of seston from lakes and reservoirs representing a variety of waterbody types (morphology, geology, climate), geographic regions and designated uses.  Toward facilitating this objective, we have asked Dr. Robert Sterner to join our group. We have collaborated with Dr. Sterner for several years on studies of nutrient dynamics in Lake Superior. Dr. Sterner is an authority on the topic of ecological stoichiometry and recently coauthored an acclaimed book on this topic (2002, Ecological Stoichiometry). Dr. Sterner’s position as a leader in the stoichiometry community will greatly facilitate our ability to satisfy this objective.

We will also conduct measures of C:N:P stoichiometry on seston collected from Lake Erie as well as from inland lakes in the State of Ohio. To facilitate accomplishing this objective, we have asked Dr. Steven Wilhelm to join in this effort. Dr. Wilhelm coordinates MELEE (Microbial Ecology of the Lake Erie Ecosystem), has collaborated with us on the topic of P availability (2003, Aquat. Microb. Ecol. 32: 275-285) and has successfully leveraged ship time for our group on Lake Erie from Environment Canada during the past 8 years.

Major Objectives

1) to demonstrate the utility of the oxalate rinse for use with freshwater phytoplankton

2) to coordinate an international effort among limnologists to establish a dataset of C:N:P measurements of oxalate-rinsed seston from lakes and reservoirs representing a variety of waterbody types (morphology, geology, climate), geographic regions and designated uses

a) to directly contribute to this effort through collection of C:N:P data on seston collected from the Great Lakes (emphasis on Lakes Erie and Superior) as well as from inland lakes and reservoirs in the State of Ohio

Likewise, the World Health Organization invokes the Redfield Ratio as an indicator of ecosystem carrying capacity to support blooms of toxic cyanobacteria (WHO, 1999). For planning and management, it is important to be able to estimate which of the key resources (light, nitrogen or phosphorus) is likely to control phytoplankton biomass in a given system and examining the Redfield Ratio serves as an important approach for doing so.

The Redfield Ratio is also invoked in drafts of a Pan-European Eutrophication document applicable to both inland and coastal waters (Claussen et al., 2004) and likely to be adopted by the European Union Commission later this year.  The current draft adopts guidelines currently included in the Oslo-Paris Convention (OSPAR), a regional convention governing practices aimed at protecting the marine environment of the NE Atlantic Ocean. The OSPAR Convention includes a "Common Procedure for the Identification of the Eutrophication Status" where nutrient ratios are included as one of the key assessment parameters (OSPAR, 1997). In adopting the OSPAR "Common Procedure", the Steering Group charged with drafting the proposed legislation invoke the Redfield Ratio in interpreting the importance of dissolved nutrient ratios. 

In light of accumulating evidence that the Redfield Ratio is not universally applicable, it is clear that classical ideas of C:N:P ratios in the water column need revision, especially if we are to continue advocating the use of the Redfield Ratio as an environmental indicator of lake nutrient status. As such, this proposal is responsive to the "Ecosystems and Habitats" theme listed as "high-priority" in the 2005-2010 Ohio Sea Grant Strategic Plan.

Objective 1: We have used several model strains of phytoplankton currently available in our labs in order to assess the general utility of the oxalate rinse. These include eukaryotic (Chlorella sp., an isolate from Lake Superior) and prokaryotic (Synechococcus sp. PCC 7942 and Synechococcus sp. ARC11, an isolate from Lake Erie) representatives.  Our collaborator Steven Wilhelm (Tennessee) has been conducting parallel studies using Microcystis aeruginosa LE-3, a toxin-producing cyanobacterium isolated from Lake Erie. We are extending our assessment to several additional strains isolated from the Great Lakes and which are presently maintained in our labs. These isolates include several Synechococcus-like cyanophytes isolated by us from both lakes.

Batch cultures have been grown under phosphate-sufficient and phosphate-deficient conditions (N:P > 50:1) and cultures processed at various stages of growth. Culture growth has been monitored by in vivo chlorophyll fluorescence or by direct cell counts.  Physiological phosphorus deficiency has been monitored by measuring alkaline phosphatase activity (Sterner et al. 2004).

Oxalate rinsed and unrinsed cells have been processed for particulate phosphorus (POP) analysis using a protocol adopted by the Sterner lab (University of Minnesota).  Phosphorus is analyzed spectrophotometrically using a 10 cm path-length cell. Conventional carbon, hydrogen and nitrogen (CHN) analysis has been conducted on both rinsed and unrinsed cells to assess possible cell disruption or other biomass losses due to the oxalate rinse. To further assess the integrity of cells following rinsing, we have stained cells using Sytox Green (Molecular Probes, Inc.), a high-affinity nucleic acid stain that readily penetrates cells with compromised plasma membranes and yet cannot gain access to live cells.

Objective 2: In late 2005, we made a request to colleagues to process seston for determination of C:N:P using both conventional procedures and the new oxalate rinse.  We distributed a standardized protocol for the oxalate rinse procedure to individuals interested in contributing to this dataset.

We requested that collaborators in this project compile a series of core physio-chemical measurements at their field sites. These were to include parameters provided by conventional conductivity-temperature-depth (CTD) survey as well as basic water chemistry (TP, SRP, TN, ammonium, nitrate + nitrite). Taxonomic analysis at each site was also be encouraged. We also encouraged collaborators to gather complimentary indices of the phytoplankton nutrient status at each site (e.g. alkaline phosphatase assay, enrichment bioassays).

Objective 2a will be addressed as part of scheduled field work on Lakes Erie and Superior and through collaboration with colleagues working on the other Great Lakes.

At each station to be occupied as part of our work in the Great Lakes, a CTD cast preceded sampling. Water samples were collected from discrete depths using a metal-clean in situ pumping system (Field and Sherrell, 2003; DeBruyn et al., 2004; Sterner et al., 2004) plumbed through a laminar flow hood (Lake Superior: RV Blue Heron) or into a Class 100 clean van (Lake Erie: CCGS Limnos) where both filtered (0.2 or 0.45 µm Calyx Polypropylene capsule filter, GE Osmonics) and unfiltered water was collected. Alternatively, water was collected from discrete depths using Niskin bottles (Lake Erie: RV Lake Guardian, CCGS Griffon) or a Van Dorn sampler (Lake Erie/Maumee Bay: RV Mayflyer). Water collected in this manner was processed for POP and CHN analysis as described above for culture material.

At several locations and where time permited, we conducted size-fractionated analysis of phosphorus quotas. Samples were collected on 47 mm diameter polycarbonate filters in series (20, 5, 2 and 0.2 micron) prior to processing of individual filters as described above. We extended this approach on several occasions to collect net phytoplankton by vertical net tow using either a 20 micron mesh-size hand net or a 1 m diameter, 80 micron mesh-size net towed using a winch. In Lake Superior, we have used this approach to concentrate large-sized diatoms which are usually poorly represented in samples collected using our pump system.

At each site, a suite of basic water chemistry measurements (TP, SRP, TN, ammonium, nitrate + nitrite) was made. On Lake Erie, in association with sampling on the CCGS Limnos and CCGS Griffon, analyses were made by the National Laboratory for Environmental Testing using standardized techniques (NLET 1994; DeBruyn et al., 2004).  For sampling conducted on the RV Lake Guardian, nutrients were analyzed by EPA-GLNPO while in Lake Superior (RV Blue Heron), analyses were conducted by the lab of Dr. R. Sterner using a Lachat nutrient analyzer. Likewise, dissolved nutrient samples from Maumee Bay (RV Mayflyer) were sent to the Sterner lab for analysis. At most sites, physiological phosphorus deficiency of field samples was monitored by measuring alkaline phosphatase activity (Sterner et al. 2004).

Lake Erie: To facilitate comprehensive sampling on Lake Erie, we collaborated with several individuals or agencies. Dr. Steven Wilhelm of the University of Tennessee coordinated our research efforts on the Canadian Coast Guard ships (CCGS Limnos and CCGS Griffon). Additional sampling on Lake Erie was facilitated through EPA-GLNPO (RV Lake Guardian) and through cooperation with the University of Toledo's Lake Erie Center.  Dr. Thomas Bridgeman, a UT faculty member at the LEC, has provided assistance for us in the past (Vincent et al., 2004; Rinta-Kanto et al., 2005) collecting water samples as part of his bi-monthly water quality surveys of western Lake Erie and Maumee Bay. Dates of specific research cruises thus far completed:

RV Lake Guardian: 7-9 April, 2006; CCGS Limnos: 8-18 August, 2006; CCGS Griffon: 21-23 February, 2007; RV Mayflyer: 2005; 7 July, 19 September, 4 October; 2007; 12 June, 27 August 

Lake Superior: Sampling on Lake Superior has been facilitated through our participation as co-PIs (along with Dr. R. Sterner) on the NSF-funded project, "The Nitrifying of Lake Superior and its Intersections with the P and Fe cycles". The project was active through 2007 with major field seasons scheduled for 2005 and 2006 with three 4-day research cruises scheduled during each of these years. Dates of specific research cruises where Sea Grant objectives were addressed:

RV Blue Heron: 27-31 August, 2005; 8-12 August, 2006

Inland Lakes: Regional sampling has been conducted by the PI and his group at various locations in Ohio and NE Indiana. By no means was this sampling considered comprehensive; rather, our goal was to sample several representative inland lakes and reservoirs that span the trophic continuum.  Our choices were guided in part by results compiled by the Citizens Lake Awareness and Monitoring (CLAM) program, sponsored by the Ohio Lake Management Society (http://dipin.kent.edu/CLAM/clampage.htm). CLAM provides an opportunity for lake stakeholders to take an active role in learning about aquatic ecology, lake and stream water quality, and watershed management.  As part of the program, each year CLAM releases an annual report indexing the trophic state of a subset of Ohio's lakes and reservoirs (i.e. those "adopted" by CLAM volunteers). The most recent annual report issued by CLAM was for 2001 (Carlson and Smith, 2001). Of the 38 lakes for which data were contained in this report, 12 lakes were classified as hypereutrophic (Secchi depth < 20 inches), 25 lakes were classified as eutrophic (Secchi depth of 20-78 inches) and only a single lake, Lake Buckhorn located in Holmes County, was classified as mesotrophic. In each case, the trophic status was confirmed by lake color analysis (Custar Color Strip) indicating algal growth and not suspended sediment, as the primary factor behind light attenuation.

As a representative eutrophic lake, we have sampled Nettle Lake, located in Williams County, in the NW corner of Ohio. This lake had a Secchi depth reading of ca. 30 inches in the 2001 report.  Efforts to sample Nettle Lake have been coordinated through CLAM volunteer Jim Short with sampling conducted on 7 May, 2007.

With no oligotrophic inland lakes reported for Ohio in the 2001 CLAM report, we have extended our sampling into nearby Indiana to sample several marl lakes whose oligotrophic status is the result of the co-precipitation or absorption of ortho- or hydrogen phosphate to calcium carbonate as groundwater, supersaturated with dissolved calcium carbonate, enters the lake bottom (Otsuki and Wetzel, 1972). Several of these lakes were the subject of study (productivity and water chemistry) several decades ago (Wetzel 1966; Wetzel, 1973). Through coordination with Indiana DNR, we have thus far sampled Crooked Lake and Little Crooked Lake (both sampled by Wetzel in the 1960's) as well as Big Cedar Lake. Sampling was conducted on 22 May, 2007.

To date, sampling a variety of freshwater lakes varying in their P status, we have consistently observed a partial allocation of P to the cell surface. Great Lakes seston show between 5-15% of cellular P allocated to the cell surface, regardless of physiological P status (alkaline phosphatase activity) whereas our limited sampling of inland lakes indicates that even higher levels of P can be absorbed to the cell surface. Thus our results are consistent with those shown for marine environments in terms of the cellular partitioning of P.

Our work investigating P cycling in lakes has also lead to some novel findings regarding the forms of P available for uptake by phytoplankton.  Specifically, samples collected as part of this Sea Grant-sponsored project have been analyzed for evidence of the capacity by cyanobacteria to utilize phosphonates, novel C-P bond compounds which include the popular herbicide Roundup. All samples analyzed thus far (inland lakes and Lake Erie)  show molecular evidence for the ability to utilize phosphonates. This may carry important implications for the potential use of such compounds as a P source by cyanobacteria in Lake Erie and Ohio's inland lakes.