Water Quality, Habitat and Biological Conditions of River Systems Affected by Pfiesteria or Pfiesteria-like Organisms on the Lower Eastern Shore of Maryland: 1997.


Preliminary monitoring work was initiated with evidence of increases in fish exhibiting lesions in the Pocomoke River during October 1996. Surveys intensified in spring 1997 and longitudinal stations, covering more than 40 miles from the tidal fresh waters near Snow Hill into the mesohaline Pocomoke Sound, were established in June 1997. On August 6th, only days after a colloquium reviewing the fish, water quality and potential human health affects being experienced by watermen on the Lower Pocomoke River, a fish kill of more than 10,000 fish was recorded. The fish kill was located just beyond the mouth of the river and continued for four days. This incident would be one of four fish kill or lesion events located in three Lower Eastern Shore watersheds during the summer.

Water Chemistry Associated with the Fish Kills.

Comparison of August-October 1995-1997 median concentrations of water quality parameters across the Chesapeake Bay tributaries and mainstem were made with median concentrations for fish kill and fish lesion sites. Fish kill and fish lesion sites include data from the closure zones during and after the fish kills or lesion events in 1997. Most carbon, nitrogen and phosphorus fractions measured at these areas almost always ranked among the highest concentrations for Baywide data comparisons taking into account the effect of salinity.

Medians for total and dissolved organic carbon in the fish kill areas of the three affected systems ranked highest in their respective salinity regimes (Chicamacomico River = oligohaline, Pocomoke River and Kings Creek = mesohaline). Some portion of the high carbon concentration was certainly due to the natural blackwater chemistry of this region but anthropogenic sources were also likely contributors to these high levels.

Median total nitrogen in the Chicamacomico River site was among the top 25% of sample sites of oligohaline stations while Pocomoke River and Kings Creek locations ranked in the top 10% of mesohaline stations. Dissolved organic nitrogen ranked all three rivers in the top 10% of their respective salinity regimes. Elevated levels of dissolved organic nitrogen are also consistent with a system receiving high organic nutrient inputs.

The medians for total phosphorus concentrations in the mesohaline Pocomoke River and Kings Creek fish kill locations were among the top 10% in the Bay; the Chicamacomico River lesion site was average (ranked at the 50th percentile) among oligohaline stations, most of which have elevated phosphorus levels. For dissolved organic phosphorus, the Chicamacomico River ranked second highest for oligohaline sites behind Back River which receives effluent from the Baltimore region. Kings Creek was among the top 5th percentile for dissolved organic phosphorus among mesohaline stations while the Pocomoke River ranked at the upper 20th percent of all mesohaline stations.

Water Quality Status and Trends.

Nitrogen concentration trends throughout the region between 1985 and 1996 are generally increasing while phosphorus trends generally fail to show any improvements. Some of the nitrogen trends may be linked, in part, to increased precipitation in recent years which has led to increased runoff from the watershed. The Pocomoke river, which has two long-term monitoring stations, shows increasing nitrogen trends in the Pocomoke Sound below Shelltown. Phosphorus concentrations are increasing at both the Pocomoke City and Pocomoke Sound stations although these trends are not statistically significant, and the trends not as strong as that for the nitrogen concentrations.

Results of sampling for Pfiesteria piscicida and similar species.

A total of 74 water samples were collected during suspected fish health events from the Pocomoke River (55 samples), Kings Creek (10 samples), and Chicamacomico River (9 samples) during August and September 1997. Samples were examined under light microscope to determine "presumptive counts" of Pfiesteria-like organisms. A subset of samples with positive counts were bioassayed using fish to evaluate toxicity. Samples from bioassays where fish mortality occurred, or from cultures grown on algal prey, were examined under scanning electron microscopy for final identification.

Dinoflagellate cell densities ranged from 0-900 cells/ml under light microscope assessments by North Carolina State University (NCSU) of samples collected during fish lesion or kill events. Samples collected from the Pocomoke River on August 6 and 7 ranged from 0-900 cells/ml with two samples (collected August 7) testing toxic to fish in bioassays. Cell densities from the late August (August 26 and 27) event on the Pocomoke River were 0-300 cells/ml. Samples from Kings Creek (September 10, 11, 16, and 20) ranged from 0-315 cells/ml, with two samples collected on September 11 testing toxic to fish. Dinoflagellate densities in Chicamacomico Creek samples (September 13-15) ranged from 80-315 cells/ml. A subset of the Chicamacomico Creek samples tested toxic to fish. It is important to note that many harmless dinoflagellate species appear identical to toxic species under light microscopy. Densities of toxic stage Pfiesteria piscicida greater than approximately 200 cells/ml are generally lethal to fish. Bioassays are necessary to confirm toxicity and scanning electron microscopy is necessary for species identification.

Scanning electron microscopy (SEM) revealed a complex of Pfiesteria piscicida and three similar dinoflagellates in association with the 1997 fish lesion and fish kill events on the Pocomoke River, Kings Creek, and Chicamacomico River. Pfiesteria piscicida was present in samples collected on the Pocomoke River (August 6-12), Kings Creek (September 11) and Chicamacomico River (September 13 and 15). Three additional species were confirmed on the Pocomoke River from samples collected on August 7; Gyrodinium galathium, an unnamed cryptoperidiniopsoid, and a third unnamed dinoflagellate referred to as "shepherd’s crook". Gyrodinium galathium was also present in samples collected from Kings Creek on September 16. The cryptoperidiniopsoid was also present in a sample collected September 15 on the Chicamacomico. All three of the non-Pfiesteria species are potentially toxic. Gyrodinium galathium and the cryptoperidiniopsoid have been collected in association with fish lesion outbreaks and kills in North Carolina and Florida (J. Landsberg, pers. comm.), and fish bioassays with water samples containing all three non-Pfiesteria species collected in Maryland have proven toxic (NCSU results). Although all three non-Pfiesteria species may be toxic, only the cryptoperidiniopsoid meets all the criteria developed by J. Burkholder to be considered part of the toxic Pfiesteria complex: attack behavior toward fish, ichtyotoxicity, complex life cycle including amoeboid stages, and kleptochloroplasty. Based on these results, it appears that a complex of at least four potentially toxic dinoflagellate species, including Pfiesteria piscicida, was present in August and September 1997 on the Pocomoke River, Kings Creek, and Chicamacomico River.


Nutrients in Waters of Maryland's Lower Eastern Shore

Recent concerns about the presence of toxic Pfiesteria-type organisms in rivers of the lower Eastern Shore have led to questions about nutrients in these rivers. Nutrients are suspected by scientists as one of the factors that may contribute to the proliferation of these organisms. This series of web pages is intended to answer some general questions about recently observed (average of 1994-1996) levels of nutrients in the tidal waters of this region of the State and how they may have changed over the last 12 years. The nutrients that are particularly important to these waters are nitrogen and phosphorus because of their role in stimulating algae. The data discussed below come largely from the State's ongoing Chesapeake Bay Monitoring Program which was initiated over the 1984 - 1985 time period and provides a comprehensive assessment of water quality and biological indicators at key sites around the Bay and it's tributaries.

Have nutrient levels been changing over the years?

Total Nitrogen Status and Trends in the Northern
Maryland Chesapeake Bay and Tributaries

Total Nitrogen Status and Trends in the Northern Maryland Chesapeake Bay and Tributaries

Total Nitrogen Status and Trends in the Southern
Maryland Chesapeake Bay and Tributaries

Total Nitrogen Status and Trends in the Southern Maryland Chesapeake Bay and Tributaries

legend for trend maps

No triangle on the above graphic indicates that there was NO significant trend for that region of the Bay.

Between 1985 and 1996, several tidal waters of the lower Eastern Shore (graphic on upper right) have shown significant increases in nitrogen levels. Some of this increase may be due to the high precipitation and therefore high river flow years of 1993, 1994 and 1996. These high flow years produce more runoff which carries nutrients into tidal regions.

Total Phosphorus Status and Trends in the Northern
Maryland Chesapeake Bay and Tributaries

Total Phosphorus Status and Trends in the Northern Maryland Chesapeake Bay and Tributaries

Total Phosphorus Status and Trends in the Southern
Maryland Chesapeake Bay and Tributaries

Total Phosphorus Status and Trends in the Southern Maryland Chesapeake Bay and Tributaries

legend for trend maps

No triangle on the above graphic indicates that there was NO significant trend for that region of the Bay.

Phosphorus levels in the Lower Eastern Shore region (graphic above) have generally remained stable over the 1985 to 1996 time period.


Water Quality and Algal Monitoring

This section of the Internet site describes the design of the Chemical/Physical Habitat Quality Monitoring Program. The program description is subdivided into sections covering spatial and temporal aspects and measured variables. The monitoring strategy described below addresses the immediate priority on the Pocomoke River. This strategy (number of stations, frequency of monitoring, etc.) will be modified as appropriate to the Manokin and Chicamacomico Rivers and other systems should fish health outbreaks occur, and to identify potential hot spots as resources allow. In all cases, habitat quality monitoring will build on the existing Chesapeake Bay Water Quality Monitoring Program (CBWQMP) and the Core/Trend Non-Tidal Water Quality Monitoring Program (CTWQMP) where possible, and will be conducted in coordination with fish health investigations by MD DNR AAHP.

The Pocomoke River has been identified as one of three "Intensive" systems to be monitored in Maryland's Chesapeake Bay watershed as part of the Mid-Atlantic Integrated Assessment (MAIA) Estuaries Monitoring Program. The goal of the MAIA project is to conduct monitoring which leads to producing a regional assessment of the condition of estuaries in the mid-Atlantic region and builds on existing monitoring activities. The MAIA project will evaluate Large Systems, Large Tidal Rivers, Small Systems and Intensive sampling in Albemarle-Pamlico Sound, Chesapeake Bay, the Coastal Bays and Delaware Estuary. The MAIA project will incorporate water quality nutrient monitoring protocols established under the CBWQMP. Nutrient samples will be analyzed at the same laboratory employed by CBWQMP. Sediment measurements will also be collected for benthic macro invertebrates, silt-clay content, sediment chemistry and sediment bioassay. Data scheduled to be collected from the Pocomoke River in September, 1997 for the MAIA project will be incorporated into the water quality and habitat data analysis in this work plan. The MAIA project will also provide summary data for characterization of the Pocomoke River with other systems within the Chesapeake Bay and the mid-Atlantic region.

map of water monitoring study area
The map above shows the study area of Maryland's Lower
Eastern Shore for water quality and algal monitoring.

A. Spatial Aspects

Spatial Locations: There are two types of spatial coverage that will be used to identify and characterize our chosen systems. The first type of coverage is a longitudinal coverage from upstream to downstream in a system. Location of long-term monitoring stations from the CBWQMP will be considered when station locations are chosen. This will allow 1) comparison of current habitat quality conditions to levels observed since 1985 and 2) the identification of nutrient loading sources to the tributary. The current Pocomoke River sampling (which included 20 longitudinal stations initially and since reduced to 14 stations) serves as an example (Figure 1 and Table 1).

map of pocomoke river monitoring stations

TABLE 1. Pocomoke River Longitudinal Water Quality Monitoring Station List

Station ID Lat/Long Description Depth (m)
MEE3.3 37 0 54.825
75 0 48.066
Near buoy W "A", midway between Oystershell Pt & Long Pt. 1.5
XAJ6642 37 0 56.428
75 0 45.735
1.9 miles SW of Tulls Pt, near State Line "D" 1.5
XAJ7164 37 0 57.037
75 0 43.615
1 mile SE of Tulls Pt., N of WOr buoy "G" 1.5
XAJ7384 37 0 57.353
75 0 41.507
At WOr buoy c"J" near the Muds. 2
XAK7010 37 0 56.895
75 0 38.923
At buoy "M" SSE of Williams Pt 1.5
POK0014 37 0 58.871
75 0 38.251
Off ramp at Shelltown 8-9
POK0037 37 0 59.681
75 0 37.380
0.9 miles above Wagram Creek 5
POK0057 38 0 00.661
75 0 37.901
0.5 nmi. Upstream silo at Cedar Hall 6
POK0087 38 0 02.375
75 0 39.649
Off ramp at town of Rehobeth 6
POK01118 38 0 03.122
75 0 38.175
2.8 miles below Puncheon Landing 8
POK0146 38 0 04.522
75 0 36.386
At Puncheon Landing 5-6
POK0164 38 0 04.291
75 0 34.588
At water tower below railroad bridge 1.8 miles from Puncheon Landing 8-6
POK0187 38 0 05.08
75 0 33.24
Upriver of Town Branch and launch ramp 8
POK0218 38 0 06.49
75 0 30.80
Mouth of Willow Grove Creek 5-6
POK0232 38 0 07.26
75 0 29.70
At Milburn Landing 6-7
POK0252 38 0 07.80
75 0 28.08
At Mattaponi Landing 7-8
POK0270 38 0 08.74
75 0 27.00
Mouth of Corkers Creek below Shad Landing 4-5
POK0296 38 0 09.52
75 0 25.36
Mouth of Nassawango Creek 4-5
POK0312 38 0 10.42
75 0 24.30
Off Snow Hill Sewage Treatment Plant 4
NAS0011 38 0 10.06
75 0 26.03
In Nassawango Creek at Nassawango Road Bridge 1.8
DIV0018 38 0 06.13
75 0 32.62
In Dividing Creek at Route 364 Bridge 2

The second spatial coverage will be used to focus on and better characterize smaller, hot spots in the river system that 1) have been identified from the longitudinal sampling as nutrient sources or locations that meet the habitat requirements for P. piscicida, or 2) are locations of high incidences of fish lesions or kills and may be event driven. An event may be defined as "normal rain event" of 1.0 inches or more of rain over a majority of the watershed over a 24 hour period.. This type of sampling will be limited to a smaller area and include near shore stations as well as stations in the main channel. In addition to standard water quality monitoring, water and sediment samples will be collected at selected intensive sites and evaluated for priority chemical contaminants. This will allow us to better characterize the habitat in a given region. Intensive, near shore sampling to date on the Pocomoke River has focused on the commercial pound and bank net locations near Shelltown where a majority of the fish with lesions have been found. Additional sites will be added based on the course of the fish lesion outbreaks and results from longitudinal habitat quality and P. piscicida monitoring.

map showing where water quality sampling occurs
Location of water quality sampling for Maryland's Lower Eastern Shore.

map showing where sediment sampling occurs
Location of sediment sampling for Maryland's Lower Eastern Shore.

B. Temporal Aspects

The frequency of water quality sampling will be once per month at a minimum, with increased frequency (twice per month and storm event based) at some locations if deemed valuable. All sampling schemes (longitudinal, near-shore intensive and non-tidal) will be carried through November and will resume next spring.

C. Measured Variables

The chemical/physical component of the CBWQMP measures a broad suite of physical and chemical variables. Several variables (conductivity, temperature, dissolved oxygen, pH, and Secchi depth) are measured in situ (Table 2). Temperature and conductivity are used to calculate salinity and density.

Table 2. Variables measured in situ, range of detection, methods and references.

Variable (units) Range (Lower-Upper Limit) Method and Reference
Temperature (0C) -5 to 45 Linear thermistor network: Hydrolab System Water Quality Instrumentation Manual (HSWQIM) 1984
Dissolved Oxygen (mg/L) 0-20 Au/Ag polarographic cell (Clark) HSWQIM
Conductivity (umhos/cm) 0-200000 Temperature compensated six electrode cell HSWQIM
pH (pH) 0-14 Glass electrode: Ag/AgCl reference electrode pair; HSWQIM
Secchi Depth (m) 0.1-50 20 cm. Diameter disk. Welch, P.S. 1948.

The remainder of the measured variables are determined in the laboratory. This includes nitrogen, phosphorus, carbon and silicon species, total suspended solids, chlorophyll, BOD, and fecal coliform. A complete list of variables, detection limits, methods and holding conditions can be found in Table 3. Dissolved and inorganic nutrient fractions, samples for particulate analyses and samples for chlorophyll/phaeophytin analyses were filtered in the field with Whatman GF/F filters.

Table 3. Variables measured in the laboratory, current detection limits, method and preservation techniques. (calc = calculated values and n.a. not analyzed for and not calculated)
Variable (units) Detection Limits Method/Reference Holding Time and Conditions
Silica, Filtered (mg/l as Si) 0.01 mg/l Technicon Industrial Systems, 1986 28 days at 4 oC
Total Organic Carbon
(mg/l as C)
calculated PC + DOC calculated
Dissolved Organic Carbon 0.24 mg/l Menzel and Vaccaro, 1964 28 days at -20 oC
Particulate Carbon (mg/l as C) 0.063 mg/l Leeman Labs, Inc., 1988 28 days at -20 oC
Total Suspended Solids (mg/l) 1.5 mg/l APHA, 1981
(sect. 209D p. 94., Gravimetric)
14 days at -20 oC
Total Dissolved Nitrogen, filtered (mg/l as N) 0.02 mg/l D'Elia et. al, 1977, Valderma, 1981; EPA, 1979. (Method 353.2) 28 days at -20 oC
Particulate Nitrogen
(mg/l as N)
0.0105 mg/l Leeman Labs, Inc., 1988 28 days at -20 oC
Ammonium, filtered
(mg/l as N)
0.003 mg/l EPA, 1979.
(Method 350.1; colorimetric automated phenate)
24 hrs. at 4 oC, unacidified; 28 days at 20 oC, pH <2 with H2SO4
Nitrate & Nitrite, filtered
(mg/l as N)
0.0002 mg/l EPA 1979. (Method 353.2; colrimetric, automated cadmium reduction; diazotation) 28 days at -20 oC
Nitrite, filtered (mg/l as N) 0.0002 mg/l EPA 1979. (Method 353.2; colrimetric; diazotation) 28 days at -20 oC
Total Phosphorus (mg/l as P) calculated TDP + PP calculated
Total Dissolved Phosphorus, filtered (mg/l as P) 0.001 mg/l EPA 1979.
(Method 365.4; colorimetric; automated ascorbic acid)
24 hrs at 4oC; 28 days at 4oC, pH <2 with H2SO4
Orthophosphate, filtered (mg/l as P) 0.0006 mg/l EPA 1979.
 (Method 365.1; colorimetric; automated ascorbic acid)
Filter immediately, 28 days at -20 oC
Particulate Phosphorus
(mg/l as P)
0.0012 mg/l Aspilla et al., 1976 28 days at -20 oC
Chlorophyll/Phaeophytin n/a Std. Methods 1985.
(Method 1002)
30 days at -20 oC
Biological Oxygen Demand 2.0 mg/l Std. Methods 1989.
 (Method 5210)
48 hours at 4 oC
Fecal Coliform 2.2 MPN Std. Methods 1989.
(Method 9221 D)
8 hours at 4 oC
Total Coliform 2.2 MPN Std. Methods 1989.
(Method 9221 D)
8 hours at 4 oC

Algal Monitoring

This section of the work plan describes the design of the water column and sediment monitoring for Pfiesteria piscicida and other algae. The program description is divided into station location, sampling frequency, sampling protocols, laboratory analysis, data management, and data analysis. The monitoring strategy described below addresses the immediate priority on the Pocomoke, Manokin, and Chicamacomico Rivers. As with water quality monitoring, this strategy will be modified as appropriate should fish lesion outbreaks or fish kills occur elsewhere, and to identify potential hot spots as resources allow.

A. Sample locations

Water column and sediment samples for algae will be focused at locations of reported fish lesions or kills, and at other sites in the affected tributary which appear likely for P. piscicida outbreaks (ex. shallow, brackish, poorly flushed, nutrient enriched embayments) based on consultations with phycologists. If resources allow, additional samples will be collected in Chesapeake Bay and coastal bay tributaries unaffected by fish health problems, yet which meet the conditions suitable for P. piscicida outbreaks. Sample collection and locations will be coordinated with the Maryland Department of the Environment fish kill team so the all samples are sent through DNR.

map showing where algal sampling occurs
The map to the left shows the area of Maryland's Lower Eastern Shore where
Pfiesteria piscicida
and fish health monitoring are taking place.

B. Sampling frequency

Frequency and timing of algal sampling will be driven by lesion outbreaks, fish kills, and consultations with phycologists. Pfiesteria piscicida is most active in its toxic forms at water temperatures near or above 25C, therefore water column sampling will be concentrated July through October, unless fish health problems occur at other times of the year. Attempts will be made to sample during calm conditions, as a well mixed water column presents difficulties for collecting P. piscicida cells. P. piscicida exists in its toxic forms only for brief periods of time (generally less than 24 hrs), therefore, should a lesion outbreak or fish kill be reported, every attempt will be made to sample immediately. This is particularly important as P. piscicida may naturally exists in very high concentrations in non-toxic forms. As a result, the mere presence of P. piscicida in the water column or sediment does not serve as conclusive evidence that it is the cause of past lesion outbreaks or kills.

C. Sampling protocols

Water column samples will be collected at the site of activity (generally the surface). At each site, two water samples will be collected. One bottle will be fixed with 0.2-0.4 ml acidic lugols solution per 100 ml for species identification. The second bottle will remain unfixed and will be used for bioassays. Sediment samples will be collected from the same locations as water column samples using a 9 inch Ponar dredge. The top 2-3 cm of sediment will scraped from the sample, mixed well, placed in a 1 quart jar. Additionally, complete water quality monitoring (see above section) will be carried out at each site sampled for P. piscicida. Md DNR field sheets will be completed at each sampling site with information on site number, date, time, location, and physical parameters.

D. Laboratory Analysis

All water column and sediment samples will be split between North Carolina State University (Dr. J. Burkholder) and Florida Department of Environmental Protection (Dr. K. Steidinger) for analysis. Preserved water column samples will be scanned under light microscopy for P. piscicida appearing cells. If presumptive P. piscicida cells are found, scanning electron microscopy will be conducted to confirm. Positive results will be reported as form (to distinguish toxic from non-toxic forms) and count (cells/ml). Unfixed water column samples will be used in algal and fish bioassays to test for toxicity. Sediment samples will be subjected to standard procedures to encourage encysted P. piscicida, if present, to enter the water column from which it can be identified. Some support will be provided to these labs to help improve analysis protocols and capabilities.

Results and Conclusions

The toxic dinoflagellate Pfiesteria piscicida or a very similar species has been confirmed as the primary cause of the August 6 - 9 and August 26 - September 3 fish kills and fish lesion events on the lower Pocomoke River. At least 24 life stages of this species have been identified, all but 4 of which are completely harmless to organisms larger than other algae and bacteria. Only under certain environmental conditions will this organism briefly assume one of its toxic forms which may result in fish lesions and kills. Based on what we know of the biology of this organism and the habitat information that has been collected on the Pocomoke River, a hypothesis has been proposed to explain why toxic outbreaks of Pfiesteria-like organisms have recently occurred on the lower Pocomoke.

Shallower depths, more surface area, slower currents and increased light availability allow algae to utilize nutrients and bloom. Nutrients not utilized by algae in upper river because of depth, and limited light availability from suspended sediments and dissolved organic matter.

Pocomoke River Hypothesis
graphic explaining DNR's hypothesis concerning the Pocomoke River
Click on the part of the image above to see the corresponding data .
[Dissolved Oxygen, Nutrients, Chlorophyll a, Fish Kill/Pfiesteria]

The above is a hypothesis or model to explain why a Pfiesteria-like organism may have caused fish kills and lesions in the lower Pocomoke River. Maryland scientists are currently working with the Pocomoke River Technical Advisory Committee to develop experimental and monitoring techniques to test this hypothesis. Habitat conditions in other rivers and embayments that have experienced toxic outbreaks of Pfiesteria-like organisms will be compared to those in the Pocomoke to find common factors that trigger these outbreaks.

The Pocomoke River has some unique characteristics. It is deep from bank to bank, free flowing, and a blackwater system for most of its length. These factors prevent a significant algal community from developing in the mainstem of the River and, as a result, nutrients that enter the river from runoff and point sources are transported largely intact to the Sound. In the vicinity of Shelltown, the River widens quickly into a broad, shallow, and slow moving embayment, allowing algal communities to bloom. The resulting habitat in this area (warm, moderate salinity, calm, poorly flushed, nutrient enriched, and large algal community) is ideal for large populations of Pfiesteria to grow in its non-toxic forms. The final piece of the puzzle required for Pfiesteria to transform from its harmless to its toxic forms is large concentrations of fish; primarily menhaden. During the latter part of the summer, large schools of young menhaden begin to congregate in the lower portions of Chesapeake Bay tributaries. Plankton feeding menhaden probably find the Shelltown region attractive because of its high algae levels. The very low dissolved oxygen levels above Shelltown in the summer (a result of the small algal community and large bacterial community in the mainstem of the River) may further concentrate fish in the Shelltown region by both driving resident fish out of the River and by blocking upstream movement of fish from the bay. All factors are now in place for Pfiesteria to assume a toxic form and cause lesion outbreaks and kills. Why did this happen now and not in previous years? 1996 was an extremely wet year (rainfall in the watershed was well above the 81 year average), and it is hypothesized that the resulting runoff contributed more nutrients to the River than previously.

Pocomoke River Dissolved Oxygen

dissolved oxygen plot showing fish kill, rain event and longitudinal sampling

The line chart above shows dissolved oxygen values from the Pocomoke River in the month of August.  The decrease in dissolved oxygen that is highlighted inside the black circle is the DO Barrier that is shown on the Pocomoke River Hypothesis diagram.   This barrier may prevent fish from moving upstream away from the predation of the Pfiesteria-like organism.

Pocomoke River Nutrients

dissolved inorganic line plot showing results from longitudinal cruises
dissolved inorganic phosphorus line plot showing results from longitudinal cruises

The two line plots above show results from the longitudinal sampling cruises that DNR has been running since June 17, 1997. The plots (dissolved inorganic nitrogen, upper, and dissolved inorganic phosphorus, lower) are the fractions of nitrogen and phosphorus that are most readily available for use by phytoplankton. The areas highlighted by the black rectangles show that as phytoplankton. take up the nutrients from the water column the concentrations of DIN and DIP decrease. Some of the decreases in DIN and DIP are due to dilution with higher salinity, lower nutrient concentration waters from the Chesapeake Bay. Salinity dilution plots, however show a high loss of DIN and DIP in the Shelltown region.

Pocomoke River Nutrients

salinity dilution plot for DIN

salinity dilution plot for DIP

The two line plots above show results from the longitudinal sampling cruises that DNR has been running since June 17, 1997. The line plots show how nutrient concentration changes with changes in salinity (down the Pocomoke River from just above Pocomoke City to Pocomoke Sound). The general concave shape of the plots indicate a net loss of the dissolved inorganic nitrogen and phosphorus due to uptake by plankton (algae).

Pocomoke River Chlorophyll a

chlorophylla line plot from longitudinal pocomoke river surveys

The line plot to the left shows the chlorophyll a data obtained through DNR's longitudinal, fish kill and rain event sampling of the Pocomoke River beginning on June 17, 1997. Chlorophyll a is a pigment found in algae and is used to show their abundance in the water column. The area highlighted by the circle shows the algae blooms that are feeding off of the nutrients flowing down the Pocomoke River. These blooms attract the Pfiesteria-like organism and fish like menhaden that feed on the algae.

Summary of Results of Monitoring for
Chemical Contamination in the
Lower Pocomoke River


In response to the recent problems with toxic, Pfiesteria-like organisms in the lower Pocomoke River, State, Federal and University scientists have been sampling water, sediments and fish tissue for possible contamination by pesticides, metals and PCB compounds. Sampling is continuing; however, as of April 16, 1998, no significant levels of contamination have been identified. The results of these analyses are summarized below.

Surface Water

Water samples were taken in the Pocomoke River on August 7, 1997 during a major fish kill and again on August 18 following a second fish kill. Additional samples were taken in the river and in two small tributaries on September 29-30 following a rain event. Contaminants have not been observed in surface water at levels known to cause lesions in fish or which suggest any association between the occurrence of contamination in the Pocomoke Watershed and the fish kills which occurred in the Pocomoke River in 1997.

Sediment Samples

Historical sediment sample data from 1986 through the present have been reviewed and on July 25, 1997, samples were taken at 5 sites in the Lower Pocomoke River between Cedar Hall and Pocomoke Sound. All contaminant concentrations were lower than, or similar to, other tributaries and were lower than levels reported as posing harm to aquatic life.

As part of the Chesapeake Bay Tributary Monitoring Program, sediments have been tested in Pocomoke Sound (1986-93) and the Pocomoke River (1989-93). A permethrin concentration of 34.1 ppb was observed in the Pocomoke River in 1991. Subsequent data for the same site in 1992 and 1993 were 1.3 ppb and 8.8 ppb, which are both low levels relative to other sites in the Bay. In samples of sediments taken at 5 sites in the Lower Pocomoke on July 25, 1997, metals, pesticides and organic contaminants were found to be similar or low compared to other sites in the Bay.

Fish Tissue

Historical data for fish tissue has been reviewed and samples of a wide variety of commonly consumed fish were taken from the Lower Pocomoke River on June 10, 1997 and again on September 18, 1997 (results still pending for most recent collection). Levels for most toxic substances were below laboratory detection limits. A few metals were detected at levels well within human health standards.

Prepared by: Maryland Department of the Environment –TARSA, April 16, 1998

For further information, contact Richard Eskin at 410-631-3906.

Pfiesteria Investigations
Summary of Fish Health Components


Pocomoke River fishermen reported finding deep sores on many of the fish they caught in the river and Pocomoke Sound, first in October 1996, and again in April 1997. Biologists from DNR, in conjunction with outside experts, began to investigate the Pocomoke observations in October 1996, first, by contacting the Pocomoke River watermen and sampling the affected fish. The samples were examined microscopically by a DNR pathologist and tissue samples were forwarded to an MDA laboratory to test for bacteria. At the same time, samples of water and sediment from the lower Pocomoke River were sent to the Florida Marine Resources Institute to be tested for the presence and abundance of Pfiesteria piscicida,- the toxic dinoflagellate or "killer algae". The same types of samples were repeated in April and May 1997. Also in spring 1997, biologists collected fish from the Pocomoke with trawl nets in a preliminary study of the incidence and distribution of lesions in the fish populations in the river.

In June and July 1997, DNR began a more intensive study of fish populations and water quality in the Pocomoke, in cooperation with watershed studies by MDE and MDA, human health surveillance by DHMH and experimental studies conducted by researchers from the University of Maryland.

Dead or dying menhaden and prolonged incidences of lesions were observed in the lower Pocomoke River on two occasions in August 1997. Similar events occurred in Kings Creek in Somerset County and the Chicamacomico River in Dorchester County, September 1997. As many as 30,000 fish, mostly Atlantic menhaden, may have been affected in the four events. The presence of Pfiesteria piscicida has been confirmed in the Pocomoke and Chicamacomico. Other species of toxic, Pfiesteria-like organisms also may have been involved in causing erratic behavior and mortality in fish during these events.

Fish Health Findings

Fish populations

Limited trawl net sampling of the Pocomoke was conducted by DNR in April and May 1997. From 0-10% of the fish, depending on species, showed external abnormalities-the highest incidence was in white perch (up to 10%). At the same time, however, higher frequencies of abnormalities were reported by commercial watermen from fish caught in pound nets, fyke nets and bank traps.

A more rigorous trawl net sampling program was initiated in June 1997. Fish were collected from randomly selected stations from outer Pocomoke Sound upstream to Snow Hill. Of 7068 fish captured in trawls from June-November, 74 (about 1%) had external abnormalities. However, of 57 menhaden caught in the trawl samples, 90% had lesions or deep sores; most of these menhaden were caught in August samples.

Personnel from the Fisheries Cooperative Unit at University of Maryland Eastern Shore, under contract to DNR, supplied observers to document catches from commercial watermen's nets. From early July through October 1997, observers recorded 12,860 fish, including 34 species from pound nets and bank traps operated by commercial watermen in the lower Pocomoke River, Pocomoke Sound, and adjacent tributaries. Sixteen percent (2,044) of these fish had skin abnormalities- 12% of the fish had abnormalities classified as abrasions, 3.4% had lesions or sores, and 2 fish (0.01 %) had symptoms of lymphocystis, a fairly common viral infection of fish.

Dead and dying menhaden were observed in Kings Creek in September 1997. Cast net sampling on September 8 produced 306 menhaden with 100% lesions. Kings Creek was closed on September 10 and reopened on October 18. Sampling was conducted on 37 days and continued through November 26. Sample gears included fyke net, cast net, gill net, bottom trawl and midwater trawl. A total catch of 8223 fish of all species showed 1057 (13%) skin anomalies. An anomaly could be a scrape, lesion or reddening of the fins or skin. For the most part, menhaden anomalies were the round lesions associated with Pfeisteria exposure. Menhaden comprised 67% (5521 fish) of this catch and 19% of them had anomalies. Percent anomalies on days with a reasonable sample size (greater than 50 fish) ranged from to 100% on September 8 to 0% on September 26. Percent anomalies remained below 10% after October 12.

In the Chicarnacomico River on September 14, thousands of menhaden (some with lesions) were observed swimming erratically near the surface. The river was shut down immediately and reopened on November 1. Sampling was conducted on 45 days from September 15 through November 26. Gears included cast net, gill net, fyke net, and'mid-water trawl. The collections produced 2867 total fish of all species. Of this total catch, 56% of the 1063 menhaden collected had some sort of anomaly. Percent daily anomalies ranged as high as 61% (October 11).

Fish pathology

Pathology examinations were conducted on a total of 203 fish from areas where there were high incidences of external abnormalities from October 1996-September 1997. These areas included Pocomoke River and Sound, Kings Creek and Chicamacomico River. Of this total, 163 fish (10 species) were affected with skin abnormalities; 40 healthy fish (5 species) were examined as references. Sub-samples of these fish were analyzed for bacterial infections. Lesions in the affected fish for the most part fell into two categories: 1) those with granulomas (a wound-healing response) and evidence of fungal infections in the dermal (exposed) tissues; and 2) those without granulomas and no evidence of fungal infections. Both types of lesions were chronic and active, meaning that they had been present long enough for the fish to show a defensive response, and that they did not show significant healing. Type 2 lesions were found in most of the fish examined from April-August 1997, except for menhaden taken from the Pocomoke during the two events in August. Type I lesions were found in most of the fish collected from the Pocomoke during the two events, as well as from the Kings Creek event in September and the Chicamacomico event in October. Fish with lesions taken from the Pocomoke in October 1996 (3 white perch and 2 white catfish) showed the Type 2 lesions.

Several bacterial species were isolated from tissues of the affected fish. These were all species found naturally in estuaries, and were very likely opportunistic infections, with no particular significance.

Experimental studies

These studies were designed to observe closely what happens to healthy fish when exposed to Pocomoke River water. A system of on-shore tanks was set up to hold fish in flowing river water. Exposed fish (white perch) will be observed in 1998 to determine if and when lesions develop. Samples from the exposed fish will be analyzed for indicators of stress, immune function, and other health measures. Healthy fish also will be placed in cages deployed in the river for specified periods, and subjected to the same analyses. Pilot experiments were conducted in the fall of 1997 to validate the methods, especially to make sure that fish were not damaged by handling or being maintained in the tanks and cages. Full scale experiments will take place during 1998. The pilot studies were successful in determining optimum sample sizes, methods for collecting blood and tissues, and eliminating the enclosures as sources of error.


The overall incidence of severe lesions and sores in the entire Pocomoke River fish populations from fall of 1996 through fall of 1997 was low, but some species, especially menhaden, had high rates of abnormalities. The incidence of reported (but not verified) abnormalities also was higher in the commercial catch than in independent trawl samples.

Observations from fish sampling programs in many areas of the Bay also indicated very low incidences of affected fish.

Two different kinds of lesions (Type I and Type 2) were identified by microscopic examination of fish tissues from the Pocomoke and other affected areas. Because of these differences and the fact that the occurrence of the two types was separated in time, it is likely that more than one cause was at work. One type of lesion (Type 1) was more closely associated with menhaden mortality events than Type 2. Additional observations during 1998, especially the experimental exposures of fish to Pocomoke River water, will improve our knowledge of the nature and causes of the lesions.

A protocol was adopted in 1997 to monitor the occurrence and extent of suspected Pfiesteria related fish sores and human health implications.

When a significant fish kill in progress was confirmed and the fish exhibited Pfiesteria-like sores (Type 1) or a significant number of fish were acting erratically and 20% or more of one species exhibited fresh Pfiesteria-fike sores or an increase in the number of fish with Pfiesteria-like sores, the area was closed until the conditions that initiated the closure had ceased to exist for 14 days.

Adoption of this protocol resulted in an advisory being issued for the Pocomoke in early August and the Pocomoke being closed for 37 days in 1997; the Chicamacomico was closed 5O days- and Kings Creek was closed 39 days.

For more Information contact Cindy Driscoll from Fisheries at the Oxford Cooperative Laboratories at 410-226-5193

Pocomoke Watershed Pollution Assessment

Background and Summary of Results to Date


The Pocomoke River watershed is a combined watershed that includes the drainages of the Pocomoke, Manokin and Big Annemessex Rivers. The Pocomoke River watershed drains approximately 433 sq. mi. of Worcester, Somerset, and Wicomico Counties in Maryland, 38 sq. mi. in Delaware, and 17 sq. mi. in Virginia. The primary tributaries of the Pocomoke River are Dividing (62 sq.mi.) and Nassawango (69 sq.mi.) Creeks. The Manokin and Big Annemessex Rivers drain 163 sq. mi. in Somerset and Worcester Counties, Maryland.

Landuse in the watershed in Maryland is roughly 33% agriculture, 38% forest, 10% wetlands, 2% developed land and 3% water. The largest communities in the watershed are Pocomoke City, Princess Anne, Crisfield and Snow Hill.

Agriculture in the watershed is mixed cash grain, vegetable crops and livestock. Corn, soybeans, and small grains are the predominate crops. Poultry production is the main type of livestock agriculture. The annual poultry population is estimated at approximately 22,400,000. The manure produced by all of the livestock in the watershed is estimated to be in excess of 329 million pounds.

Surface Water Quality Results

Water quality monitoring conducted by MDE and DNR indicates that nutrient concentrations in the fresh water portion of the Pocomoke River system are similar from Snow Hill to the Delaware state line. Total nitrogen concentrations are consistently above 1 mg/l at all sampling sites above Snow Hill. Total phosphorus concentrations are consistently above .1 mg/l.

Concentrations vary with flow and season. Nitrogen concentrations are higher during the winter months when ground water recharge is occurring. Phosphorus concentrations are higher during storm events in the summer.

Nutrient loads and yields have been calculated for two small subwatersheds in the Upper Pocomoke. Compared to average yields in the literature, the nutrient yields from the small watersheds being monitored in the Upper Pocomoke are high (Frink 1991).

Average TN and TP Yields (lbs/acre/year)
Watershed TN Yield TP Yield
White Marsh Run 8.21 1.15
Piney Creek 11.31 1.50
Upper Pocomoke 20.14 2.20
German Branch 14.56 0.71
Ave. For Ag Land1 11.77 1.63

1. From Frink et.al.

Ground Water Quality

The surficial aquifer on the Pocomoke watershed has a high silt and clay. The sediments are poorly drained because of poorly incised steams and flat water-table gradients. In much of this area, drainage ditches have been constructed, which will act to lower the water table. In the area closer to the Chesapeake Bay in the Pocomoke River Basin, the water table is generally shallow and comprised of poorly-drained, low-permeability sediment. The majority of ground-water in surficial aquifers discharges to adjacent streams and is termed "base flow". The percentage of total stream flow contributed by base flow ranges from about 42 to 74% (Bachman and others, in press). These data are based on using hydrograph separation techniques at 20 sites with stream flow data.

Nitrate concentrations in the surficial aquifers underlying the Pocomoke River basin generally range from <0.1 to 35 milligrams per liter, with about 10 percent of the samples above the maximum contaminant level of 10 mg/L set by U.S. Environmental Protection agency. The median concentration was 1.1 milligram per liter. These values are based on data collected from 1976 to 1990 (Hamilton and others, p. 45).

The nitrate concentrations appear to be related to land use and the type of sediments in the aquifer and soils. The nitrate concentrations were elevated in agricultural and residential areas in zones where the aquifer consists of sandy deposits (Hamilton and others, p. 65). The sandy composition of the aquifer promotes aerobic conditions under which nitrate is stable. In agricultural and residential areas underlain by clay and silt deposits and anaerobic water, effects of human activities are less evident. The silt and clay deposits can cause lower nitrate concentrations for several reasons (Hamilton and others, p. 65). The anaerobic conditions promote denitrification. The high amounts of clay and silt can inhibit downward movement of fertilizers from the land surface into the water table and have abundant exchange sites for ammonia, which reduces the amount of nitrogen available to reach the water table.

The age of the ground-water can also affect the nitrate concentrations. Knowing the apparent age of ground water can help explain nitrogen concentrations and provide an estimate of how long a management practice may take to be effective. Based on existing data the apparent ground-water ages in the surficial Coastal Plain aquifers range from about 5 to 20 years (Focazio and others, 1998, in press).

Watershed Modeling Results

The Chesapeake Bay Program Phase IV Watershed Model (WSM) calculates point and nonpoint source pollutant inputs to Bay watershed streams and the subsequent transport of these pollutants to the Bay. The Pocomoke and Manokin Watersheds are represented by a single segment (Segment 430) in the Watershed Model. Thus one set of model inputs and outputs is used to represent these watersheds.

Land use information in the Watershed Model is based on the Chesapeake Bay Program Land Use coverage (CBPLU). The CBPLU is a combination of EMAP, CCAP, and GIRAS land use data. The Pocomoke watershed extends into Virginia (6%) and Delaware (7%), but is largely contained in Maryland (87%). The watershed in Maryland is roughly 33% agriculture, 48% forest and wetlands, 2% developed land, and 3% water. Agricultural lands are further split into conventional till, conservation till, hayland and pasture base on the 1992 Agricultural census and the Conservation Tillage Information Center.

Nutrient inputs to the model include point sources, septic systems, atmospheric deposition, and manure and chemical fertilizer applied to agricultural lands. Manure inputs to the model are based largely on the 1992 Agricultural Census. Manure is applied to agricultural lands at a rate (137 pounds N per tilled land acre and 40 pounds P per acre in the Pocomoke Watershed) that ensures all manure generated is used. An input rate of 70 pounds N per acre and 20 pounds P per acre is assumed for chemical fertilizer application to tilled land. Atmospheric N is deposited at a rate of 10 pounds per acre to all lands.

The fate of these inputs are modeled (for example, plant uptake accounts for 97 pounds N per tilled acre) by the Watershed Model resulting in land use loading rates to tidal waters given below:

Hayland (lbs/ac) Pasture
Nitrogen 31.5 25 3.7 8.8 1.8 16.5
Phosphorus 2.6 1.8 0.4 0.3 0.01 0.4

These loading rates applied to the Watershed Model land use acres result in a load to tidal waters, ignoring Best Management Practice implementation, of 13.4 million pounds of N and 0.16 million pounds of P.

For Further Information about Watershed Monitoring Activities please contact John McCoy at 410-260-8803 or jmccoy@dnr.state.md.us.


Bachman, L.J., 1984a, The Columbia aquifer of the eastern Shore of Maryland, part 1--hydrogeology: Maryland Geological Survey of Investigations Report 40.

Bachman, L.J., and others 1998, Relation of hydrogeomorphic setting, ground-water discharge, and base-flow nitrate loads from non-tidal streams in the Chesapeake Bay Watershed, USGS Report of Investigations, in press

Focazio, M.J. and others, 1998, Preliminary Estimates of Residence Times and Apparent Ages of Ground Water in the Chesapeake Bay Watershed and Water-Quality Data from a Survey of Springs, USGS Report of Investigations, in press.

Frink, C.R., 1991. Estimating Nutrient Exports to Estuaries. J. Environ. Qual. 20:717-724.

Hamilton, P.A. and others, 1993, Water-Quality Assessment of the Delmarva Peninsula--Effects of Agricultural Activities on, Distribution of, Nitrate and Other Inorganic Constituents in the Surficial Aquifer, USGS Open File Report 93-40

 Pfiesteria Links:  FAQ | Monitoring | Publications | Contact Us
Pfiesteria piscicida | Pfiesteria shumwayae

Search Maryland DNR

Search www.dnr.state.md.us

Restoration and Protection | Bay Grasses | Harmful Algae | Bay Monitoring
Bay Life Guide | Bay Education

Return to the Maryland DNR Home Page.
Your opinion counts! Take a