Nutrient Control Actions for Improving Water Quality in the Mississippi River Basin and Northern Gulf of Mexico (2009)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Allocating Nutrient Load Reduction Targets Decisions and policies for reducing nutrient loadings in watersheds and tributaries across the Mississippi River basin are complicated by numerous geographic, economic, legal, historical, and political factors. This section addresses question 2 in this committee’s statement of task to discuss “alternate methods to allocate load reductions.” It identifies several factors to be considered in setting allocations and discuses two fundamental considerations in these decisions: equity and cost effectiveness. As explained below, there is a good rationale for considering both factors in load reduction plans. ESTIMATING LOADS, REDUCTION TARGETS, AND SPATIAL DISTRIBUTION OF SOURCES Two key decisions need to be made before nutrient load reductions can be allocated. The first is to determine a target for the reduction of aggregate loads reaching the Gulf of Mexico. The second is to determine the spatial units to which load reductions are to be allocated. In other words, it is necessary to establish a target as to how much reduction is to be achieved, then to decide how the aggregate reduction will be divided among spatial units within the basin. Targets and Spatial Units Given the uncertainty regarding the amount of nutrient load reduction that may be necessary to reduce the areal extent of hypoxia, it may be prudent to set a series of interim targets over time in an adaptive, incremental approach to load reduction allocation. The need to proceed adaptively in addressing the nutrient loadings-Gulf of Mexico hypoxia challenge was explained in the 2007 EPA SAB report on hypoxia: Accordingly, it is even more important to proceed in a directionally correct fashion to manage the factors affecting hypoxia than to wait for greater precision in setting the goal for the size of the zone. Much can be learned by implementing 41 

42 NUTRIENT CONTROL ACTIONS FOR IMPROVING WATER QUALITY management plans, documenting practices, and measuring their effects with appropriate monitoring programs (USEPA, 2007, p. 2). Once load reduction targets are set, those reductions must be allocated among spatial units. The two main types of spatial units for allocating reductions are states, or watersheds within states. Because the Clean Water Act encourages states to assume the primary responsibility to address water quality, state boundaries are logical political boundaries for dividing responsibilities for load reductions. However, the amounts of loads delivered to the gulf differ greatly among the Mississippi River basin states, and also differ greatly in different watersheds in the same state (Figure 3). Therefore, allocation of load reductions to highest priority watersheds will result in a very different spatial pattern of allocations from one based solely on states. If federal funds are intended to be targeted to watersheds likely to have the most cost-effective impact on reducing nutrient loads delivered to the northern Gulf of Mexico, load reduction allocations based on watersheds are essential. Allocations to states then would be determined by summing allocations to watersheds within states. Any allocation to an interstate watershed would have to be apportioned among states based on the watershed area within each state. Point and Nonpoint Sources Nutrient loads in the Mississippi River basin are dominated by nonpoint sources. This dominance of nonpoint source loadings is very different from some other river basins where interstate initiatives have been taken to reduce nutrient loads. For example, in-basin nonpoint source loads in the Connecticut River Basin delivered to Long Island Sound were only 33 percent of the total pollutant loadings (NY State Dept of Env. Protection and CT Dept of Env Protection, 2000). The high percentage of nutrient loadings contributed by nonpoint agricultural sources across the Mississippi River basin presents a special challenge for administering water quality improvements actions pursuant to the Clean Water Act, because those nonpoint sources cannot be regulated directly at the federal level by a permitting process. FACTORS IN LOAD REDUCTION ALLOCATION DECISIONS Once load reduction targets for nitrogen and phosphorus for the northern Gulf of Mexico have been established, those reductions must be allocated among priority watersheds across the Mississippi River basin. There is previous experience in making similar decisions in two large U.S. watersheds—the Chesapeake Bay and North Carolina’s Neuse River. One useful guide is a formal analysis of alternative approaches to reducing nonpoint nitrogen loads 

ALLOCATING NUTRIENT LOAD REDUCTION TARGETS 43 delivered to the hypoxia-plagued Neuse River estuary by 30 percent (Schwabe, 2001). A structural model was used to compare costs of a uniform rollback strategy with a cost-minimization strategy, taking into account heterogeneity of biochemical and physical factors across subareas within the basin and fate and transport of nitrogen in streams that deliver loads to the estuary. The uniform rollback policy places greater weight or considerations of equity, while the cost- minimization policy places greater weight on considerations of cost- effectiveness. Of course, issues of equity versus efficient use of limited financial resources abound in all types of public decisions (further discussion of this topic as it relates to water resources decisions is in Druzrik and Theriaque, 1996). Balancing Equity and Cost Effectiveness Equity Decisions upon tradeoffs among the several aspects of equity and cost- effectiveness are central to the process of allocating nutrient load reductions. A simple model may help in illustrating the concept of equity. Let the basinwide target reduction be represented by T tons per year, and let the number of watersheds to which T is allocated be N. Then, T must be divided among the N units included in the management program, taking into account the percentages of loads from those sources that are delivered to the Gulf of Mexico. Let Li be the load in tons per year generated in watershed i; di is the fraction of Li that is delivered to the Gulf ; and pi is the fraction of Li that is to be reduced. Then, the load reduction equation can be written: p1d1L1 + p2d2L2 + p3d3L3 + . . . . + pNdNLN = T The allocation problem is one of selecting the set of load reduction factors, pi, i = 1,2,3…N. For nonpoint sources, a simple formula for allocating load reductions that has been used in the Chesapeake Bay Program and was actually adopted for the Neuse River Basin, is the uniform rollback strategy (p1 = p2 = … = pN). That concept of equity is accepted by many stakeholders, state land and water managers, elected officials, and other parties. It may be equitable only in a limited sense, however, because it does not account for large variations in percentages of loads within watersheds that are delivered to the Gulf of Mexico (see Figure 3). Uniform reductions of loads delivered to the Gulf of Mexico would require that reduction percentages assigned to watersheds be weighted by 

44 NUTRIENT CONTROL ACTIONS FOR IMPROVING WATER QUALITY delivery factors, namely pidi = pjdj or pi = pj(dj/di) for all pairs of spatial units. Equity considerations also extend to methods for financing control strategies. If the costs to one group (call them Group A) are borne by owners in that group, and costs to a second group (call them Group B) are covered from general tax revenues, Group A will not consider the result to be equitable. That condition is a real one—many point source dischargers currently are paying for construction, operation and maintenance largely from their own source revenues, while costs of management practices for nonpoint sources (e.g., buffer strips, water controls, tillage practices, nutrient control actions, etc.) are subsidized to a significant extent from state and federal tax revenues. Yet another equity consideration is ability-to-pay. There are significant differences in per capita income among the Mississippi River basin states, the lower basin states generally having lower incomes than in the upper basin. Such differences among landowners and producers who would be affected by load reductions requirements also enter into these decisions and policies. Another consideration is past actions taken to reduce pollutant discharges. Setting load allocation targets conceivably will affect the discharges of many different sources and parties. Invariably, some of these parties will have taken few past measures to reduce nutrient yields, while other parties will have made stronger efforts to reduce nutrient yields. It is important that these past efforts be recognized in setting future allocation targets. In setting future allocation regimes, parties who have implemented past nutrient reduction measures should receive some credit for these actions. Cost Effectiveness Cost effectiveness of load reductions also is an important consideration. Analysis clearly has shown that heterogeneity of soils, slopes management practices, characteristics of tributary streams, and unit costs can have significant effects on costs of reducing loads to downstream water bodies subject to severe hypoxia (Schwabe, 2001). These important geographic differences—which can be substantial in adjoining sections of the same watershed—point to the importance and value of “precision agriculture” practices (Cox, 2008). If a cost-effectiveness approach is to be pursued, actions or policies that distribute financial assistance uniformly across all watersheds or across all municipalities will be counter-productive. Targeting requires that funds be disproportionately—and more efficiently—distributed to watersheds and municipalities with higher nutrient loads and high delivery coefficients. Development of a credible, formal least cost model for the entire Mississippi River Basin (or portions of the basin subject to significant nutrient loads) is not likely to be completed in the near future. At least one study is underway to construct a least cost model for the upper portion of the basin (Rabotyagov et al., 2007), and where other credible results are available they 

ALLOCATING NUTRIENT LOAD REDUCTION TARGETS 45 may be used to guide allocation decisions. Even in the absence of formal optimization models of least cost approaches and policies, guiding principles can be derived from existing studies of watersheds and regions within the basin. Among the important guiding principles are: 1) some management practices are generally more cost effective than others in particular settings and, more specifically, 2) management practices on watersheds with higher loading densities and higher delivery coefficients are likely to be more cost-effective than on watersheds with lower densities and lower delivery coefficients. There is also interest in the prospects of market-based approaches, such as tradable permits or allowances, to manage water quality across a watershed. Interest in market-based approaches to water quality management stems, in part, from the extensive use of tradable permits to manage air pollution. There has been less experience in water quality trading than in air quality, especially in watersheds where nutrient loads are dominated by nonpoint sources. Most examples of trading in water quality have been among point sources or where point sources have been allowed to purchase offsets from nonpoint sources. Nevertheless, there is potential for market-based approaches to manage water quality more cost effectively and these should be encouraged. As was observed in the previous 2008 NRC report, “. . .water quality trading regimes could become more useful and widespread over time as monitoring improves and as stricter water quality criteria are adopted” (NRC, 2008, p. 181-182). SETTING LOAD ALLOCATIONS FOR THE MISSISSIPPI RIVER BASIN As the preceding section has explained, there are alternative methods and multiple factors to consider in allocating load reductions. Furthermore, final decisions about load reduction targets are not based fully on scientific and engineering factors and also must consider social, economic, and political issues. In developing a load reduction allocation scheme for the Mississippi River basin, the experience in allocation of load reductions for the Chesapeake Bay merits careful consideration (Box 4-1). There is no standard formula or practice for setting these targets, and practices in one watershed may not transfer perfectly to another. Important differences between the Chesapeake Bay system and the Mississippi River/Gulf of Mexico system must be kept in mind. In particular, the Mississippi River basin is much larger and extends over 31 U.S. states and six different EPA regions. Nevertheless, there are important parallels between these two systems: both are affected by downstream water quality problems of nutrient overenrichment, large percentages of these nutrients derive from agriculture in both systems, and both systems extend over many different states and thus necessitate interstate approaches and cooperation for effective water quality administration. The experience of addressing Chesapeake Bay watershed nutrient yields and 

46 NUTRIENT CONTROL ACTIONS FOR IMPROVING WATER QUALITY BOX 4-1 The Chesapeake Bay Program: An Example of Interstate Water Quality Monitoring and Nutrient Control Actions In considering approaches to reducing northern Gulf of Mexico hypoxia, water quality experts and decision makers often look to modeling, monitoring, load reduction allocation, and related efforts that have been undertaken for the Chesapeake Bay. Efforts to reduce nutrient loadings to the bay and to develop a basinwide, nutrient management program date back to the 1980s. The Chesapeake Bay Program was founded in 1983 as a regional partnership to direct bay restoration. Program members include Maryland, Pennsylvania, and Virginia; the District of Columbia; the Chesapeake Bay Commission (a tri-state legislative body); the US EPA, and citizen advisory groups (for more information visit: http://www.chesapeakebay.net/overview.aspx; accessed September 11, 2008; also see NRC, 2008 for more discussion of the program). The program today encompasses a range of scientific and nutrient management components and includes: a coordinated water quality monitoring program; interstate information management arrangements; consistent water quality standards; and tributary watershed cap load allocations. Many aspects of the Chesapeake Bay experience are relevant to creating a similar science-based nutrient control program for the Mississippi River basin and northern Gulf of Mexico. Scientifically, that program includes an interstate information management system, basinwide water quality monitoring, and integrated water quality modeling and data analysis. Regarding nutrient control efforts, participants in the program agreed to annual nitrogen load and sediment load reductions and to a basinwide permitting strategy. The process by which nutrient load caps were allocated is particularly relevant to this report. As explained in the previous 2008 NRC report: Final basinwide nutrient cap loads were allocated to the nine major tributary basins. Basin allocations were further divided and assigned to each of the six watershed states and the District of Columbia based on principles of fairness and equity. . . Individual states have the option to further sub-divide their major tributary basin cap load allocations into 44 state-defined tributary strategy sub-basins (NRC, 2008). Finally, the time requirements to establish, develop, and extend the various components of the Chesapeake Bay Program should be kept in mind. As mentioned, the monitoring and nutrient control efforts in the Chesapeake Bay date back to the early 1980s, and it has taken decades for the program to develop into its current state. The development of a similar program for the Mississippi River basin and northern Gulf of Mexico clearly will require a similar amount of time—if not longer, given the greater size of the Mississippi River basin. If the Mississippi River basin states and the federal government are to establish a similar program of water quality monitoring and modeling (some of which are reflected in this report’s recommendation for the “NCII”) and nutrient control actions some time in the foreseeable future, it will be important to initiate soon similar monitoring, evaluative, and nutrient control actions. 

ALLOCATING NUTRIENT LOAD REDUCTION TARGETS 47 downstream water quality impacts represents a significant effort that the federal government and Mississippi River basin states should look to in considering a future allocation scheme and process. In particular, Chesapeake Bay program components that should be considered in establishing a similar process for the Mississippi River basin are: • The extensive water quality monitoring system; • The use of water quality models to inform a host of administrative decisions; • The process of agreeing to cap nutrient loads; • Dividing allocations by major river sub-basin; and, • Further dividing allocations on successively smaller watersheds. Regardless of the method chosen for allocation, however, the allocators should also consider the potential need for future adjustments to the overall goal. Given the recommended use of interim goals, allocators need to be sensitive to the possibility that early investments in “hard” technologies could limit future choices in adaptive management. In other words, allocators should consider the possibility that early commitments to certain technologies may commit the overall adaptive management strategy to limited paths. Finding/recommendation 7: In working toward a load reduction allocation scheme, the EPA, USDA, and the Mississippi River basin states should draw upon the experience in the Chesapeake Bay in allocating nutrient loading caps. In doing so, the following principles for allocating cap load reductions should be considered: • Select an interim goal for nutrient load reductions as the first stage of an adaptive, incremental process toward subsequent reduction goals; • Target watersheds to which load reductions are to be allocated; • Adopt an allocation formula for distributing interim load reductions to targeted watersheds within the basin that balances equity and cost-effectiveness considerations; • Allow credit for past progress; and • Encourage the use of market-based approaches to allow jurisdictional flexibility in achieving nutrient load reductions. It bears keeping in mind, however, that such markets do not automatically lead to satisfactory outcomes. Such markets require some regulatory caps on nutrient losses in order to operate. 

48 NUTRIENT CONTROL ACTIONS FOR IMPROVING WATER QUALITY