Draft Objectives for Non-Tidal Water-Quality Monitoring Defining the objectives of a monitoring program is an essential first step for data-collection network design. The objectives define the purpose and scope of the monitoring program and in that way provide the basis for determining if a given network design is adequate. For example, a monitoring program with the objective of quantifying loading from first-order streams could have a network design that is much different than a monitoring program with the objective of determining compliance with water-quality standards. In either case, the objectives must be defined in order to determine if a given network design will be adequate. Furthermore, the objectives should be defined in sufficient detail to determine the criteria for an "adequate" network. In many cases, as they evolve over time, monitoring programs begin to be used for multiple objectives. Objectives are often added to the existing network as new needs arise. Often, the data-collection network is less than optimal for more than one objective, because it was not designed for objectives that began after the network was in place. In such cases, the network design should be evaluated periodically in order to determine if it is adequate for all of the purposes for which it is being used. When there are multiple objectives, adequacy should be considered separately for each in order to identify weaknesses and necessary improvements for individual objectives. The criteria for data collection for all the objectives can then be combined to develop the overall data-collection network. The non-tidal water-quality monitoring network in the Chesapeake Bay watershed is an example of a data-collection network that has multiple objectives, which have evolved over time. The original network consisted of federal, state and local agency water-quality monitoring networks that were designed separately with different (although often similar) objectives. Those individual networks are now viewed as one data-collection network for the purposes of Chesapeake Bay restoration. The Chesapeake Bay Program (CBP) Non-tidal Water-Quality Monitoring Workgroup is now conducting an effort to evaluate the adequacy of the overall network for addressing the objectives for which it is already being used. Where weaknesses in the overall network are identified, recommendations will be made to the agencies that conduct the monitoring in the hope that they will be able address the weaknesses. Scope of Monitoring Objectives for Non-tidal Redesign This document is intended to identify and describe the primary objectives of the non-tidal water- quality monitoring network for use in Chesapeake Bay restoration. Once the objectives are defined, criteria can be established that describe the characteristics of a network design that is adequate for addressing the objectives being considered. Comparison of the necessary criteria ------- with those of the actual monitoring network will provide a basis for evaluating the existing network and for making recommendations for improvement. The evaluation being performed by the CBP Non-Tidal Water-Quality Workgroup is primarily focused on nutrients and sediment, which are considered to be the greatest problems for water quality and ecological integrity in the Chesapeake Bay. Other water-quality constituents such as toxic contaminants are also important and the need for monitoring of those constituents is described in the CBP monitoring strategy. However, monitoring design for toxic contaminants is beyond the scope of this work and objectives for toxic contaminant monitoring will not be considered here. The Non-Tidal Water-Quality Workgroup recommends that a separate network design effort be established for monitoring toxic contaminants. This evaluation is focused primarily on water quality and related issues in the non-tidal part of the watershed. Non-tidal living resource monitoring is also needed, but is not considered here. Separate efforts by the Living Resources Sub-committee Monitoring and Modeling Workgroup will address the need for living resource monitoring in the non-tidal part of the watershed. Finally, this evaluation is focused primarily on basinwide water quality and loading to the Chesapeake Bay. The objectives considered here will be those that consider a whole watershed perspective as compared to more local issues such as TMDL's and local compliance with water- quality standards. While such local issues are related to the water quality and ecological integrity of the Bay, they will be quite variable across the watershed and outside the influence of the CBP. Thus only the broad spatial scale objectives will be considered here. Definition of Monitoring Objectives The objectives for the non-tidal water-quality monitoring network are based on commitments listed in the Chesapeake Bay 2000 agreement. Non-tidal monitoring data are necessary to address a number of commitments. Among those are the following: Nutrients and Sediments 3.1.1 • Continue efforts to achieve and maintain the 40 percent nutrient reduction goal agreed to in 1987, as well as the goals being adopted for the tributaries south of the Potomac River. 3.1.2 • By 2010, correct the nutrient- and sediment-related problems in the Chesapeake Bay and its tidal tributaries sufficiently to remove the Bay and the tidal portions of its tributaries from the list of impaired waters under the Clean Water Act. 3.1.2.5 • By 2003, work with the Susquehanna River Basin Commission and others to adopt and begin implementing strategies that prevent the loss of the sediment retention capabilities of the lower Susquehanna River dams. Non-tidal monitoring data are essential for tracking the effectiveness of the 40 percent nutrient reduction goal (3.1.1) for reducing stream loads. Combined with the watershed model, the data can also be used for identifying reasons for observed responses of streams. Non-tidal monitoring is also essential for developing management strategies that will lead to removal of the Bay from ------- the Clean Water Act list of impaired water bodies (3.1.2) and to understanding and managing the sediment retention capabilities of the lower Susquehanna River dams (3.1.2.5). Specific uses of the non-tidal water-quality monitoring data for achieving the commitments listed above include: 1. estimating the amounts of nutrients and sediment that reach the Bay from non- tidal parts of the watershed (load estimation); 2. identifying the watershed sources and basin characteristics that affect nutrient and sediment transport (load estimation); 3. tracking temporal changes in concentration and loading, and relating them to natural variation or management actions (trend analysis); 4. relating temporal changes in loading to temporal changes in Bay water quality (trend analysis); 5. calibrating the watershed model in order to evaluate the potential benefits of management action scenarios (calibrate watershed model); 6. performing studies to understand the processes that affect nutrient / sediment generation, loss and transport (research). Generally, these examples and most others can be grouped into three categories: 1) load estimation; 2) trend analysis; 3) watershed modeling and 4) research. The non-tidal monitoring network should provide data that are adequate for these applications and ideally the network would be designed for these applications. The current network, however, consists of the combination of multiple smaller networks that were designed for other purposes. Thus the non- tidal network evaluation is intended to identify gaps and weaknesses for the objectives described above. More detailed discussions of these objectives are provided below. Load Estimation Much of the water-quality data that is collected is used to quantify the amount of nutrients or sediment that is transported from the drainage above the sampling location. This information is used to perform mass-balance studies to identify the predominant sources of nutrients or sediment upstream or to quantify the amount going downstream for the purpose of evaluating impacts there. Load estimation is a critical task for the Bay restoration because it forms the basis for managing nutrient transport to the Bay. For the load estimation objective, water-quality samples are usually collected in association with a stream gage where continuous discharge data are available. Intermittent water-quality samples are collected, transported to a laboratory and analyzed to provide concentration data. Generally, concentration data are available in relatively low frequency, which may range from 10 to 50 values per year. Because concentration data are usually available in low frequency, statistical tools are used to estimates stream loads. Most of these tools are designed to optimize the ------- accuracy and precision of load estimates based on covariance between concentration and discharge. Some specific examples of load estimation include the following: 1. The spatial distribution of load estimates are often related to the spatial distribution of sources (e.g. - STP's) or basin characteristics (e.g. - hydrogeomorphic region) to assess the importance of those features for determining the amount of nutrient or sediment transport. 2. Load estimates are often used to compare the export rates of different drainages so that they can be ranked to set priorities for management actions. 3. Load estimates are used to generally characterize watersheds in order to communicate with the public such as in the web-based Watershed Profiles. Water Quality Trends Trend analysis is performed to evaluate temporal changes in measures of water quality. Trends are usually analyzed to define statistically significant changes over time, the direction of change and possibly the amount of change. Results of those analyses provide information to evaluate the potential benefits or ineffectiveness of management actions or natural variations of which mangers should be aware. Trend analysis is generally performed by visually evaluating the time series of monitoring data and by applying a statistical test to determine if any change is significant. Thus monitoring over several years is required to identify trends. Some monitoring sites would be more advantageous for this objective solely because a long historical record already exists. Stations for water quality trends may require less intensive sampling regimes than those needed to estimate loads because the higher frequency variations are less important to long term changes. Watershed Modeling Monitoring is required to estimate parameters for the watershed model being used by the CBP to evaluate management scenarios. Scenarios such as the more widespread application of Best Management Practices (BMP's) will need to be evaluated. Some scenarios will include activities intended to ameliorate a specific type of environmental stress, such as erosion from fields or stream banks. Other scenarios may have a more general scope, such as locating housing developments away from streams and other critical habitats. The objective of this type of forecasting is to permit the CBP to compare the contribution of management activities to reducing nutrient and sediment loads. Still other scenarios will incorporate the continued growth of the population of the watershed and the accompanying changes in land use. Modeling will be used to forecast water quality as a consequence of future land use. The number of calibration sites needed for this purpose is unclear, but in general, the accuracy of the model will be improved by a larger number of monitoring stations that are broadly representative. Such an array of sites could improve the capability of the model to discriminate ------- between watersheds. Most of the current sites represent large watersheds with heterogeneous land uses, whereas model calibration could be more precise when based on smaller watersheds with homogenous land cover. Sampling regimes and station locations are dictated by the requirements of the model, but generally the model requires calibration data for a variety of substrates and land use types. Emphasis is given to sampling to estimate those parameters to which the model is most sensitive. Specific land-use classes should be included to determine how well the model functions over a wide range of land-use categories. Additional data may be required to validate model performance. Research/Education Uses The results of monitoring are useful to researchers and educators. As the cohesiveness and quality of the monitoring data improve, the data will become increasingly useful for determining long-term changes in the Chesapeake Bay ecosystem. Several long-term changes in the watershed are occurring now and will attract increasing attention as the shorter-term objectives of the CBP are met. One of the most notable is the increase in urbanization accompanied by the conversion of forest and agricultural land to housing. Another trend is the expected increase in atmospheric C02 concentration, which may result in increases in primary productivity and the sequestering of more nutrients in the non-tidal portions of the watershed. If the increase in CO2 is accompanied by an increase in global temperature, it is likely to produce a change in farming practices as well as shifts in the composition of forested ecosystems. All of the trends listed above will interact in ways that are very difficult, if not impossible, to predict. The monitoring data will become an important tool to evaluate the effects of changing land-use as it interacts with potential changes in global climate. Long-term non-tidal monitoring data is also valuable in the study of small single-land-use watersheds for the purpose of understanding processes that affect nutrient and sediment loading. For example, long term monitoring of predominantly agricultural watersheds, combined with other types of information, can be useful for understanding the processes that affect the transport of nutrients and sediment from such areas. Similarly, long-term monitoring of predominantly forested watersheds can be used to understand how nutrients and sediment might be transported under relatively pristine conditions. Process oriented research such as that described for single land uses can be valuable for the goals of the Chesapeake Bay program because it provides information that supports the development of reduction goals and the improvement of modeling tools. For that reason, long-term monitoring should be used to support such efforts whenever possible. For the reasons stated above, research is important for the goals of the Chesapeake Bay Program, but is generally considered to be a secondary objective of the non-tidal water-quality monitoring network. Various research efforts have used water-quality data from the network and it is expected that this will continue in the future. However, research objectives can be quite variable across the watershed and can change with time. Thus it is difficult to include research as a broad objective for the design of a non-tidal water-quality monitoring network. The Non-Tidal Water- Quality Workgroup would like to express support for the use of water-quality monitoring data in research efforts and will provide support where possible. However, research cannot be considered a primary objective for the design of the network and will not be used as a basis for that purpose. ------- |