Water Quality Assessment Program in the Indian River Lagoon, Florida:
II. Redesigning a Monitoring Network

Gilbert C. Sigua, Joel S. Steward, Janice D. Miller, and Wendy A. Tweedale

Environmental Sciences Division, St. Johns River Water Management District (SJRWMD)

P.O. Box 1429, Palatka, FL 32178

 

Abstract

The protection and maintenance of good water quality related to the re-establishment of a healthy "seagrass ecosystem" in the Indian River Lagoon (IRL) can best be accomplished through re-examination and refinement of the water quality monitoring network (WQMN) to meet current needs. Modifications to the IRL-WQMN were intended to focus its mission toward a) providing answers to specific questions related to the long-term management of seagrass and the water quality of its environment, b) increasing the statistical power of the data collected, c) increasing the effectiveness of staff and laboratory resources, and d) collecting complementary data for the calibration of the Pollution Load Reduction Model. A selective reduction in the number of sampling stations was proposed to eliminate statistically unnecessary sampling, resulting in a more efficient sampling effort in the IRL system. The stations that are retained will continue to provide information on the long-term spatial and temporal trends of water quality and help discern the covariant or causal link between seagrass coverage (distribution/density) and water quality. The proposed modifications also included the following changes: 1) selection of existing sampling stations based on their proximity to seagrass measurement transects; 2) an increase in the sampling frequency; 3) inclusion of near-bottom nutrient samples; 4) measurement of the organic fractions of total suspended solids; and 5) centralization of laboratory analyses to reduce potential analytical errors. By streamlining the WQMN as proposed, staff and laboratory resources will be used more effectively and place less budgetary demand on the participating agencies. It will be a more cost-effective and efficient monitoring tool to measure the water quality of the seagrass environment.

Introduction

The IRL-WQMN was established in 1988 as a coordinated multi-agency project spanning the entire length (~ 248 km) of the IRL system (Figure 1). Water quality monitoring in the IRL system consisted of sampling at regular intervals (monthly) for a suite of parameters agreed upon by the different participating agencies. The active participants of the network are the St. Johns River Water Management District (SJRWMD), South Florida Water Management District (SFWMD), Volusia County, Brevard County, Indian River County, and NASA-Dynamac. These agencies collectively managed a total of 150 stations (nearly one station per Lagoon 1.6 km). The IRL-WQMN had the task to generate information on the physical and chemical conditions of the IRL and to infer the Lagoon’s well-being or biological integrity. The IRL-WQMN is an invaluable management tool (Steward et al 1994), with a mission to:

This five-point mission of the IRL-WQMN is facilitated by close coordination of the participating WQMN agencies to ensure proper, quality-assured operation of the network, and the periodic evaluation and modification of the network as necessary to meet current resource assessment needs.

Indian River Lagoon System: An Overview

The IRL, from Ponce DeLeon Inlet to Jupiter Inlet, is a biogeographic transition zone, rich in habitats and species, and with the highest species diversity of any estuary in North America. This system is comprised of three interconnected estuarine lagoons, the Mosquito Lagoon (ML), the Banana River Lagoon (BRL), and the Indian River Lagoon (IRLB-Indian River Lagoon-Brevard County; IRLIR-Indian River Lagoon-Indian River County). The Lagoon system receives inputs of salt water from the ocean through inlets and freshwater from direct precipitation, groundwater seepage, surface runoff, as well as discharges from creeks and streams (non point sources) and wastewater treatment plants (point sources). Generally, little flushing action exists at the northern end of the estuary as tidal influence in that area is small and overwhelmed by wind. In areas close to the inlets, tidal elevations and currents are more pronounced, and, thus flushing is improved.

Mosquito Lagoon. Mosquito Lagoon is a large, shallow estuarine system along the east central coast of Florida in Volusia and Brevard counties (Figure 1). The northern end of the lagoon connects to the Atlantic Ocean through Ponce DeLeon Inlet near New Smyrna Beach. The 57-km long watershed of ML, is bounded on the east by the barrier island and on the east, west, and south by dune ridges (Higman 1994).

Banana River Lagoon. Banana River Lagoon is located in Brevard County. The 52-km long watershed of the BRL lies east of Merritt Island and west of the barrier islands. These barrier islands are composed of relict beach ridges formed by the action of wind and ocean waves (Brown et al 1962). The western watershed boundary is the Kennedy Parkway until the parkway turns west, at which point the boundary follows a dune ridge south (Steward and VanArman 1987). The prominent physical feature of this drainage area is Cape Canaveral, which is located on the barrier island. South of the Cape is Canaveral Barge Canal, a navigational channel which connects the IRLB and the BRL with Port Canaveral and the Atlantic Ocean.

Indian River Lagoon. The Indian River Lagoon is about 648 sq. km in area. It has four connections to the Atlantic Ocean, namely: Sebastian Inlet, Fort Pierce Inlet, St. Lucie Inlet, and Jupiter Inlet via Hobe Sound. Circulation and flushing in the IRLB and IRLIR are greatly influenced by freshwater inflows, inlets, and winds. Principal freshwater sources for the IRLB and IRLIR are natural streams, direct land runoff, and a number of wastewater treatment plants. The major streams are located south of Merritt Island and include Eau Gallie River, Crane Creek, Turkey Creek, Sebastian River, and St. Lucie River.

Modifications to the Existing IRL-WQMN

Specific resource management questions were developed to re-focus the WQMN within the context of its mission and enable conversion of data into meaningful information regarding the interrelationship of water quality, light, and seagrass requirements. The WQMN participant agencies have offered the following set of questions whose answers should be attempted by the network:

These questions provided guidance in the redesign of the WQMN with respect to what variables need to be measured, spatial and temporal sampling specifications, and analytical procedures.

Regular data analysis and presentation of monitoring results should be as routine as the monitoring itself. They are important as a means to evaluate the WQMN’s performance—its ability to answer resource management questions and provide accountability to the public. Knowing the type of information to be presented determines the methods of data analysis and presentation. Green (1979) emphasized the importance of developing testable hypotheses during the design phase of environmental studies. The development of testable hypotheses and the selection of statistical methods are the first steps in evaluating the expected performance of the WQMN.

Redesign of the IRL-WQMN was proposed in 1995 and was adopted in 1996 for implementation by the different participating agencies. The modifications to the WQMN will be in effect for up to three years before re-evaluation. Modifications to the IRL-WQMN are the following:

Sampling Methods And Redesign Protocols

The coordination of a multi-agency monitoring network in a consistent manner demands standardized methods, procedures, and equipment (Sigua et al 1996). This standardization is necessary to produce reliable and comparable data. This section identifies and describes sampling methods that were adopted by the IRL-WQMN participating agencies.

Number and Locations of Sampling Stations. The SJRWMD proposed a selective reduction (from 150 to 23 stations) in the number of sampling stations to eliminate statistically unnecessary sampling, resulting in a more efficient sampling effort in the IRL system (Table 1). The stations that were retained continue to provide information on the long-term spatial and temporal trends of the water quality and help discern the covariant or causal link between seagrass coverage (distribution/density) and water quality.

The decision to retain or delete stations was initiated with a sequence of statistical analyses of water quality data that served to group neighboring stations representing a Lagoon segment of relatively homogeneous water quality. Then, within each segment, one or two representative stations were selected based on proximity to seagrass monitoring transects and to prominent watershed impacts (e.g., wastewater treatment plants, significant tributary or stormwater inflows, inlets, etc.). The accepted number of segments per reach is supported by the results of multivariate analyses The variables selected for the analyses were salinity, DO, turbidity, chlorophyll a, TP, and TKN, with a 1988 - 1991 period of data record. For each lagoon reach, principal component analysis of these data identified the principal variable(s) responsible for inter-segment variability. In ML, the principal variable was turbidity; in the BRL and IRL-B, it was salinity; and in the IRL-IR, it was both turbidity and salinity (SAS 1988).

The grouping of "like" stations or a segmenting of each of the lagoon reaches, based on principal component analysis, was confirmed by cluster and kriging analyses and univariate analyses. Sampling stations (Figure 1) were selected within each segment based on their proximity to existing seagrass transects. Station selection was also based on major or representative land-based activities and resource features within the IRL Basin (e.g., wastewater treatment plants, inlets, and tributary inflows from significant urban watersheds and relatively undeveloped watersheds).

As an aid to the foregoing discussion, an schematic (Figure 3) depicts the sequence of statistical tests used to determine segmentation (the number and placement of lagoon segments). First, the IRL Basin was treated as four separate reaches: Mosquito Lagoon (ML), Banana River Lagoon (BRL), Indian River Lagoon - Brevard County (IRLB), and Indian River Lagoon - Indian River County (IRLIR). Water quality variables that can affect seagrasses, and for which there are sufficient data, were selected to delineate reach segments. The water quality data for each reach were assumed to be normally distributed, and have independence of observation and homogeneity of variance over the period of record. Second, each reach was initially divided into segments based on a visual discrimination of water quality differences between station groupings aided by spatial and temporal data distribution plots. Then, multivariate analyses, including proc manova and principal component analysis, served to accept or reject the segmentation hypotheses, as stated below. Once the segments of each reach were delineated, one or more stations were chosen to represent each segment. Most stations were chosen based on their proximity to existing seagrass transects.

1) Mosquito Lagoon: (Ho - null hypothesis; Ha - alternative hypothesis)

Ho: Total segments = 0

Ha: Total segments 0 (at least 3 segments based on visual discrimination)

Result: Reject Ho; Mosquito Lagoon has 4 segments (based on turbidity)

2) Banana River Lagoon:

Ho: Total segments = 0

Ha: Total segments 0 (at least 3 segments based on visual discrimination)

Result: Reject Ho; Banana River Lagoon has 3 segments (based on salinity)

3) IRL-B (Brevard County)

Ho: Total segments = 0

Ha: Total segments 0 (up to 8 segments based on visual discrimination)

Result: Reject Ho; Indian River Lagoon (IRL-B) has 7 segments (based on salinity)

4) IRL-IR (Indian River County)

Ho: Total segments = 0

Ha: Total segments 0 (at least 3 segments based on visual discrimination)

Result: Reject Ho; Indian River Lagoon (IRL-IR) has 3 segments (based on turbidity and salinity)

General Sampling. Three near-surface samples per station will be collected during a tidal phase each month to capture any tidal or seasonal trends in the lagoon. Near-bottom nutrient samples will be collected at each station at least once during a tidal cycle (~12 hr) each month. Photosynthetically active radiation (PAR) measurements for each station will be done after 10:00 a.m. at least once a month. Sampling will be done on the Tuesday following the second Monday of each month, or on the day(s) agreed to by the WQMN group, or as soon as possible after that day if inclement weather prevents sampling. Water samples will be shipped within the appropriate holding time to a centralized laboratory for chemical analyses. Each county will continue to sample and measure for the different parameters (meteorological and physical) listed in Table 2. Monitoring schemes and measurements for these physical characteristics should involve in situ methods. The list of water column chemical properties shown in Tables 3 were the major consideration for the centralization of laboratory analyses in the IRL.

Near-Surface and Near-Bottom Sampling. It is recommended that at least one of the three samplings (Circuits 1, 2, or 3 as shown in Figure 2) during the tidal cycle run include a near-bottom water sample. The near-surface sample can be taken using a water grab sampler (Van Dorn). Prior to and at the time of sampling from the near-surface or near-bottom depth, it is very important that there is minimal sediment disturbance. The station should be approached slowly, ideally with boat engine cut-off as the boat is coasting into position, and then the anchor gently lowered. The sample should be taken as far away from any points of possible disturbance (anchor, engine, other sampling activities, etc.). The recommended depth at which the near-bottom sample would be taken using a "bilge pump" or peristaltic pump is 0.3 m above the bottom. The IRL database (1988-1991) reveals that the depth of the water column at most of the retained stations is usually more than 1.25 m. If the mean water column depth is less than 1 m, no near-bottom sample will be taken; only the near-surface sample.

Evaluation of WQMN Performance

Evaluation of the redesigned IRL-WQMN will be performed after three years of implementation. The questions that will be addressed in the evaluation are:

1. Can the modified sampling design meet the requirements of a long-term, ambient water quality monitoring program focused on the seagrass environment?

2. In the short term, can the modified WQMN meet the calibration needs of the PLR Model? Can the WQMN, in conjunction with other seagrass diagnostic projects and the PLR Model, establish better water quality targets for the IRL, and accurately measure IRL conditions relative to those targets?

3. How can the redesigned WQMN be further modified to ensure that the objectives or questions of the monitoring program are sufficiently addressed?

The cost of the IRL-WQMN within the SJRWMD is substantial. In order to derive the most benefit from the network, it is essential to periodically (at least every three years) evaluate its expected performance. This performance information will provide the basis for determining the feasibility of proposed sampling strategies and optimizing the overall monitoring effort. Initially, this evaluation will take place following a first year of sampling based on the modified network.

References

Brown, D. W., W. E. Kenner, J. W. Crooks, and J. B. Foster. 1962. Water resources of Brevard County, Florida. 104 p.

Higman, J. 1994. Mosquito Lagoon water quality status: spatial and temporal patterns. SJRWMD. 123 p.

SAS Institute, Inc. 1988. SAS/STAT User’s Guide. Release 6.03. SAS Institute, Cary, NC. USA. 494 p.

Sigua, G. C., W. A. Tweedale, J. D. Miller, and J. S. Steward. 1996. Inter-agency implementation of modified water quality monitoring program for the Indian River Lagoon: Methods and QA/QC Issues. Technical Memorandum # 19. St. Johns River Water Management District, Palatka, Fl. 36 p.

Steward, J. S., R. Virnstein, D. Haunert, and F. Lund. 1994. Surface water improvement plan for Indian River Lagoon. 119 p.

Steward, J. S. and J. A. VanArman. 1987. Indian River Lagoon Joint Reconnaissance Report. Contract Report Nos. (CM-137 and CM-138). Florida Office of Coastal Management. Department of Environmental Regulation. 367 p.

 

 

Figure 1. Indian River Lagoon showing the old water quality monitoring stations.

 

 

Figure 2. Sampling strategy for Volusia County—example.

 

 

 

Figure 3. Flowchart of statistical tests for the segmentation of the Indian River Lagoon (ML-Mosquito Lagoon; BRL-Banana River Lagoon; IRLB-Indian River Lagoon-Brevard County; IRLIR-Indian River Lagoon-Indian River County).

Table 1. Sampling Stations Being Monitored per Modified IRL-WQMN

in the Indian River Lagoon

Sampling Station

IRL Segment

Sampling Agency

V02

Mosquito Lagoon

Volusia County

V11

Mosquito Lagoon

Volusia County

V17

Mosquito Lagoon

Volusia County

ML02

Mosquito Lagoon

NASA-Dynamac

I02

Indian River Lagoon

SJRWMD

I07

Indian River Lagoon

SJRWMD

I10

Indian River Lagoon

SJRWMD

I13

Indian River Lagoon

SJRWMD

I16

Indian River Lagoon

SJRWMD

I18

Indian River Lagoon

Brevard County

I21

Indian River Lagoon

Brevard County

I23

Indian River Lagoon

Brevard County

I27

Indian River Lagoon

Brevard County

B02

Banana River Lagoon

NASA-Dynamac

B04

Banana River Lagoon

Brevard County

B06

Banana River Lagoon

Brevard County

B09

Banana River Lagoon

Brevard County

IRJ01

Indian River Lagoon

Indian River County

IRJ04

Indian River Lagoon

Indian River County

IRJ05

Indian River Lagoon

Indian River County

IRJ07

Indian River Lagoon

Indian River County

IRJ10

Indian River Lagoon

Indian River County

IRJ12

Indian River Lagoon

Indian River County

 

 

 

Table 2. Water Column Physical Parameters for the Indian River Lagoon Water

Quality Monitoring Network

Physical Parameters

Unit

Water Temperature

degrees Celsius

pH

pH units

Dissolved Oxygen

mg/L

Conductivity

mhos/cm

Salinity

parts per thousand

Secchi

meters

Depth of Collection

meters

Depth of Sample Site

meters

Air Temperature

degrees Celsius

Wind Direction

degrees

Wind Velocity

miles per hour

Cloud Cover

percent

 

Table 3. Water Column Chemical Parameters for the Indian River Lagoon Water

Quality Monitoring Network

Chemical Parameters

Unit

Color

PCU

Turbidity

NTU

Total Organic Suspended Solids

mg/L

Total Inorganic Suspended Solids

mg/L

Chlorophyll a

g/L

Chlorophyll b

g/L

Chlorophyll c

g/L

Pheopigments

g/L

Chlorophyll a corrected

g/L

Chlorophyll a / Pheopigment Ratio

 

Total Organic Carbon as C

mg/L

Total Kjeldahl Nitrogen as N

mg/L

Nitrate + Nitrite as N

mg/L

Dissolved Kjeldahl Nitrogen as N

mg/L

Total Phosphorus as P

mg/L

Total Orthophosphorus as P

mg/L

Dissolved Phosphorus as as P

mg/L

Silica as SiO2-D

mg/L