Page 1
1
Reducing Phosphorus Impairments with Nutrient Trading
Arun Khatri-Chhetri
Former PhD graduate student
Natural Resource Economics
West Virginia University
Morgantown, WV 26506
Email: [email protected]
Phone: 540-239-0702
Alan R Collins, Professor
Agricultural and Resource Economics
West Virginia University
Morgantown, WV 26506
Email: [email protected]
Phone: 304-293-5486
Selected Paper prepared for presentation at the Agricultural & Applied Economics
Association’s 2013 AAEA & CAES Joint Annual Meeting, Washington DC, August 4-
6, 2013
Copyright 2013 by Arun Khatri-Chhetri and Alan R Collins. All rights reserved. Readers may make
verbatim copies of this document for non-commercial purposes by any means, provided that this
copyright notice appears on all such copies.
Page 2
2
Abstract
The Greenbrier River watershed in West Virginia suffers from severe algal bloom
problems. A combination of hydrological and economic models is used to assess the physical
and economic feasibility of generating total phosphorus (TP) credits to offset a proposed TP
standard of 0.5mg/l for wastewater treatment plants (WWTP). The results reveal that most
cropland and only a proportion pasture and grassland would need base management practices to
TP runoff from agricultural land in order to fulfill WWTP’s TP load reductions. All four TP
credit market scenarios resulted in a cost savings compared to no market, with potential cost
savings of up to $1.2 million/year over WWTP upgrades to meet the TP standard.
Page 3
3
Reducing Phosphorus Impairments with Nutrient Trading
Introduction
During recent years there has been growing attention to control nutrients related water
pollution in the United States. A large amount of nutrients is annually discharged into the water
bodies from both point and non-point sources. This large influx of nutrients is causing serious
water quality problems such as algal blooms, eutrophication, and hypoxia in many streams,
rivers and coastal waters (Lopez et al. 2008; Diaz and Rosenberg 2008; Selman and Greenhalgh
2009). Reducing nutrient loads in the streams and rivers is essential for water quality
maintenance and restoration.
Excess growth of algae from nutrient enrichment is a major problem in many streams and
rivers of West Virginia (WV). Greenbrier River has the most severe algae problems of all West
Virginia Rivers (WVDEP 2010). The primary source of the problem has been identified as high
phosphorus discharges from the seven municipal wastewater treatment plants (WWTPs) in the
watershed combined with discharges from agricultural sources and septic systems (WVDEP
2011). Reducing phosphorus discharges in the watershed is a key challenge for West Virginia
Department of Environmental Protection (WVDEP) to improve water quality in the Greenbrier
River. Recently, WVDEP proposed a total phosphorus (TP) discharge standard of 0.5 mg/l for all
WWTPs in the Greenbrier River watershed. Under this new requirement, many existing
permitted WWTPs need to make significant upgrades in their treatment plants to comply with the
discharge limitation which requires large amount of capital investments.
Recent literature indicates that nutrients trading between point and nonpoint sources can
lower the cost of meeting water quality goals in a watershed (Kardos and Obropta 2011;
Newburn and Woodward 2012; Houtven et al. 2012). As the WVDEP moves toward assigning
Page 4
4
TP discharge limitation, the trading of TP credits between WWTPs and agricultural sources
presents a cost saving alternative to treatment facility upgrades. Farms located upstream can
implement agricultural best management practices (BMPs) on their lands and reduce TP
discharge at a lower cost than WWTPs with treatment plant upgrades (Ribaudo and Nickerson
2009; CTIC 2011). Nutrient trading would enable WWTPs to pay for upstream improvements to
lands that drain into an impaired water body.
The goal of this paper is to present an assessment of both the physical and economic
feasibility of nutrient credit trading in the Greenbrier River watershed to lower the costs of
complying with possible TP discharge limitation standards for WWTPs in the Greenbrier River.
We examine four different nutrient trading scenarios by considering: two trading ratios between
WWTPs and farmers (1:1 and 2:1) and two baseline requirements for agricultural sources
(existing level of BMPs and 100% nutrient management plan). A nutrient management plan
(NMP) has been proposed as baseline requirements for agricultural sources in water quality
trading guidelines proposed in West Virginia (WVWRI 2008).
This study uses hydrological and economic models to assess the physical and economic
feasibility of TP trading in the Greenbrier River watershed. The physical feasibility analysis
includes the estimation of nutrient reduction requirements for the WWTPs (potential demand for
nutrient credits) and nutrient reduction potential from the agricultural sources (potential supply
of nutrient credits) in the watershed. The economic feasibility analysis includes estimation of
costs of TP credit generation from the agricultural sources, cost of TP reduction for the WWTPs,
demand for and supply of TP credits, and cost savings for individual WWTP as well as aggregate
across all WWTP in the Greenbrier River watershed.
Page 5
5
The Greenbrier River watershed consists of 1640 square miles in the southeastern corner of
WV (Figure 1). It is part of Mississippi River Basin that drains nutrients to the Gulf of Mexico.
Estimates show that WV contributes relatively less phosphorus to the Mississippi River and the
development of hypoxia in the Gulf of Mexico (Alexander et al. 2008). Still, phosphorus
discharged from WV contributes to interstate problems and development of hypoxia in the Gulf
of Mexico. Any phosphorus reduction activities in small sub-watershed of Mississippi River
basin can contribute to the reduction of algae problem in the local water body as well as hypoxia
in the Gulf of Mexico. Thus, it is anticipated that all jurisdictions within the Mississippi River
basin required reducing phosphorus loadings to help reduce water quality problems in the rivers
and coastal waters.
Nutrient Trading
Environmental economists have advocated market-based strategies for water pollution
control (Freeman 2003; Tietenberg 2007). Nutrient credit trading, one such strategies, is
receiving substantial attention from many researchers, policy makers, and local water resource
organizations as a means of meeting nutrient load limits in a cost-effective way. As the marginal
costs of additional nutrient reductions from point sources are high and because non-point sources
can reduce nutrients at a lower cost than point sources, watershed-based nutrient trading between
point and non-point sources offers an alternative to the technology based pollution control
approach (Fang 2005; Shabman and Stephenson 2007; Ribaudo and Nickerson 2009). The
United States Environmental Protection Agency (USEPA) also recognizes watershed-based
nutrient trading programs for minimizing overall costs of water quality protection and restoration
(USEPA 2008).
Page 6
6
Nutrient dischargers (point and non-point sources) in a watershed face different pollution
abatement costs. When a credit market is available, sources with low marginal abatement costs
can abate more than their required amount of pollution and then sell their excess abatement
credits to high-cost abatement sources. When a nutrient credit market is available, sources with
high marginal abatement costs can comply with a regulatory discharge limit at lower cost by
purchasing credits rather than installation of expensive nutrient removal technology (Fetch 2000;
Horan and Shortle 2011; Newburn and Woodward 2012). However, the economic and
environmental benefits from a nutrient trading market depend upon market design parameters
such as the trading ratio between point and nonpoint sources, baseline requirements for nonpoint
sources, and the effluent limitations for point sources considered to establish a water quality
trading (WQT) program in a watershed (Khatri-Chhetri 2012).
A high trading ratio increases the price of nutrient credits in a WQT market, hence
reducing total volume of credits traded (Malik, et al. 1993; Horan and Shortle 2005). Most
trading programs require participating agricultural nonpoint sources to meet a certain level of
pollution reduction before best management practices (BMPs) can generate nutrient credits
(USEPA 2007). This is called a baseline requirement and it has a major impact on the cost of
pollution reduction from the agricultural sources and overall gains achievable from the trading in
a watershed (Gosh et al. 2011). In general, regulatory authority sets effluent limitations for point
sources based on the desired water quality level in the water body. The effluent limitation
enforced to the point sources is the key factor in determining demand for nutrient credits in a
watershed (King and Kuch 2003).
Since the late 1980s, nutrient trading between point and nonpoint sources has been the
subject of research and experiments in different locations in the U.S. However, very limited
Page 7
7
programs are currently active and successful in reducing nutrients discharge at the watershed
level (Breetz et al. 2004; Selman et al. 2009). There are 21 water quality trading (WQT)
programs that are active as of 2012 in the U.S. TP is most common pollutant traded in the
currently active WQT programs. Of the 21 programs, 15 target phosphorus trading between point
and nonpoint sources. There are very limited ex-post evaluations of nutrient trading programs.
An ex-post evaluation of Ohio’s Great Miami Water Quality Trading Program indicates that the
program has been quite successful in reducing nutrient discharges in a cost-effective way
(Newburn and Woodward 2012).
Theoretical Models
Modeling Point Sources (Waste Water Treatment Plants)
A point source (i.e. WWTP) is assumed to discharge pollution directly into the water
body and it controls discharges by selecting wastewater treatment technologies. Let wik1 denote
the total quantity of water outflow from the ith
firm following treatment with technology k1 and
eik1 denote the nutrient concentration of the outflow water. The total amount of nutrient outflow
(TP) from the ith
firm after treatment is eik1w
ik1. The firm faces the abatement cost, (
),
which depends on the level of TP reductions by choosing a abatement technology (k1). This is a
continuous, twice differentiable function and C' > 0 and C" > 0. A TP discharge cap is set below
the current discharge level for this firm so that the total discharge cannot exceed firm’s mandated
pollution discharge level, . The i
th firm aims to minimize its abatement cost subject to the set
discharge constraint.
Min (
)……………………………………….. (1)
Subject to
≤ ……………………………………………. (2)
Page 8
8
and ………………………………………………….. (3)
The constraint (2) shows that the amount of total emissions of TP must not exceed the set
discharge limit. This minimization problem can be solved by using the Kuhn-Tucker conditions.
(
)
)…………………. (4)
By differentiating this with respect to we get the Kuhn-Tucker conditions for the optimum:
(
)
[ (
) ]
≤ , [
] …………… (6)
; ………………………………………………….. (7)
λ represents the marginal abatement costs (MAC) of the firm, expressed as (
)
. This λ
is positive only when the TP discharge constraint (2) is binding.
Let us assume that nutrient trading market exist in the watershed and firm i can purchase
TP credits, , from other sources. Now, the i
th firm’s total discharge is:
≤ +
……………………………………………………………. (8)
Where =
The total cost of firm is:
(
)
………………………………………… (9)
Where, Cek is the cost of operating the kth
technology and p is the per unit price of TP credit that
prevails in the nutrient trading market. It is assumed that the total quantity of water inflow to the
firm i’s treatment plant and the total quantity of water outflow from the ith
firm following
treatment with technology k1 remains the same. Now the firm faces the problem of minimizing
total costs which consists of abatement costs and cost of TP credits as follows:
Page 9
9
Min (
)
……………..………………….. (10)
Subject to (3) and (8)
The constraint (8) shows that the amount of TP discharge from the firm i must not exceed the
mandated amount of TP discharge plus amount of TP credits purchased in the nutrient trading
market. The solution for the minimizing problem can be reached by solving following Kuhn-
Tucker conditions:
(
)
[
(
) ] …………………………………. ……… (12)
……………………………………………………………….. (13)
A comparison of equations (5) through (7) with equations (11) through (13) shows that p = λ =
MAC to be a sufficient condition for this cost minimization problem. Each firm’s MAC curve
represents a TP credit demand function which can be represented by qid (p,
). The market-
level demand for TP credits can be obtained by aggregating individual point source’s demands:
D(p) = ∑
p, ).
Modeling Non-Point (Agricultural) Sources
Building on the model of Peterson et al. (2005), we assume that agricultural production
exhibits constant returns to scale which can be expressed in per acre term. Let, y = y1, y2…….yJ
denote a vector of yields of J crops, x = x1, x2, ……xk denotes a vector of K inputs, p and s, are
vector of output and input prices. Total cost of crop production is denoted by c. Profit for a farm
in the absence of TP abatement is:
Page 10
10
TP discharges are generated based on the level of inputs (x): In the absence of TP
controls, firm chooses zero abatement and produces y* level of output discharging r* level of TP.
In a nutrient market, a nonpoint source would have the incentive to reduce their TP
discharge level through abatement activities (z). z = (z1, z2, z3……zj) denotes a vector of z
abatement activities in a farm. The profit for a ith
farm in the adoption of TP abatement activities
is:
( )
Both (inputs) and (abatement activities) have positive effects on cost function. Thus,
( ) is a continuous, twice differentiable function and ci' > 0 and ci" > 0. The TP discharge
after implementation of abatement activities is: . A firm cannot implement any
abatement activity more than the land area that the firm own. Thus, the first constraint is:
………………………………………………………… (16)
Where, Ai is the total agricultural land of the ith
farm. TP reduction from the ith
farm r* =
. In order to participate in the nutrient market, farmers need to satisfy certain baseline
level of TP reduction qb (Ai) specified by the regulatory authority. This baseline depends on total
agricultural land of a firm. TP reduction available for credits in a nutrient market is:
……………………………………….. (17)
The ith
farm seeks to maximize its profit from the agricultural production.
Max ( ) ……………………………………………………… (18)
Subject to: (16) and (17)
This maximization problem can be solved by using the Kuhn-Tucker conditions.
( ) { }…………….. (19)
By differentiating this with respect to we get the Kuhn-Tucker conditions for the optimum:
Page 11
11
( )
[ ( ) ]
, [ ] …….. (21)
; ………………………………………………………... (22)
The λ represents the marginal abatement cost (MAC) of the firm, expressed as ( )
,
This λ is positive only when the TP reduction constraint (17) is binding. z is most likely zero
when the farm doesn’t experience any benefit from the abatement activities and then λ is zero.
When a nutrient trading market exists, farm profit is assumed to be maximized by gains
from both crop production and TP credit sales. The decision problem facing the ith
farmer is:
Max ( ) {( ( )) } ……………………………. (23)
Subject to:
………………………………………… (24)
and (16)
Where, qs denotes TP credit generation after meeting the baseline requirement qb(Ai). The
solution for the maximizing problem can be reached by solving following Kuhn-Tucker
conditions:
[ ] ……………………………………………………. (26)
…………………………………………………………………… (27)
A comparison of equations (20) through (22) with equations (25) through (27) shows that p = λ =
MAC to be a sufficient condition for profit maximization problem. For the ith
farmer, the supply
Page 12
12
function of TP credit is . The market level supply curve is obtained by aggregating the
supply across all abatement activities and farmers: ∑ ∑
( ).
Data and Methods
Estimation of Potential TP Credit Demand
Estimating the potential demand for TP credits from WWTPs required information about
current treatment processes and TP discharge levels, TP load reduction requirements to meet
discharge limitations, and costs of technology upgrades for TP reduction. The cost of meeting a
0.5 mg/l TP discharge standard depends upon what standard the current treatment meets and the
costs associated with upgrading the current treatment system. Information on the type of
wastewater treatment used by WWTPs within the Greenbrier River watershed was obtained from
the Clean Watersheds Needs Survey (USEPA 2008). The existing nutrient discharge data for all
WWTPs in the watershed were obtained from the EPA’s online Discharge Monitoring Report.
Data included the size of facilities (Million Gallon per Day), average daily loadings (lb.) and
concentrations (mg/l) of TP discharges from the facilities (USEPA 2010).
For the purpose of the nutrient trading analysis, TP load reduction requirements were
estimated for all National Pollutant Discharge Elimination System (NPDES) permitted WWTPs
under the proposed TP limitation (0.5mg/l) in the Greenbrier River watershed by the WVDEP.
The TP load reduction requirements for individual WWTPs were estimated based on their
current average daily load (lbs.), design flow, and TP limit (mg/l) for the WWTP in the
watershed. Cost estimation methods developed by the Nutrient Reduction Technology Cost Task
Force, Chesapeake Bay Program (CBP 2002) were used to estimate the costs of TP reductions
for each WWTP in the Greenbrier River watershed.
Page 13
13
Total costs of upgrading WWTP facilities to meet new TP standard include both
increased capital costs and operations and maintenance costs (O&M) upgrading from the existing
treatment facility. All capital and O&M cost estimates for the facilities in the Greenbrier River
watershed were adjusted to 2011 US$ using the US producer price index. The real treasury
interest rate for different maturities was used for annualized cost of capital investments. The real
treasury interest rate for the year 2011 with 20-year maturity life was 2.1 percent. Per unit costs
($/lb.) of TP reductions were calculated based on total annual cost of facility upgrades and TP
reduction requirements for each facility. Average costs represent the incremental costs required
to achieve a 0.5 mg/l TP concentration for WWTP discharges.
Estimation of Potential TP Credit Supply
A geographical information system (GIS) based water quality model (MapShed) was
used to estimate agricultural nonpoint source TP loadings at a sub-watershed level. Mapshed
was developed by Evans and Corradini (2012) and integrates two models (GWLF and PRedICT)
to estimate nutrients and sediment runoff from various land use categories. Table 1 presents a list
of GIS data layers prepared for MapShed on the Greenbrier River watershed. The MapShed
model was run to calculate stream flow and TP loads from 1990 to 2011 for 14 sub-watersheds.
The model was validated and calibrated based on statistical tests (Nash-Sutcliffe coefficient (r2)
and mean difference (t-test) between observed and estimated data. The model provided daily,
monthly, and annual TP loadings from various non-point sources including stream flows and TP
concentration (mg/l).
Two BMPs for crop land (cover crop and nutrient management plan) and two BMPs for
pasture and grassland (rotational grazing and nutrient management plan) were selected to
estimate TP credit supply. These BMPs were selected as the most likely for implementation
Page 14
14
based on a survey of farmers in Greenbrier County. A survey of farmers in the watershed
showed that the percentages of farms with existing BMPs were: 15% for cover crop, 17% for
nutrient management plan on pasture and grassland, 24% for nutrient management plans on
cropland, and 10% for grazing land management (Khatri-Chhetri 2012). To estimate a TP credit
supply function, BMPs were selected in order of average per pound TP reduction cost with the
lowest cost BMP being considered first into the MapShed simulation assuming that farmers will
prefer to implement low cost BMPs initially to generate TP credits.
BMP implementation costs were collected from the USDA NRCS West Virginia
payment schedules for the 2012 Environmental Quality Incentive Program (EQIP) (NRCS,
2012). The EQIP payment schedule was provided based on a BMP unit cost basis ($/acre).
Linear cost functions were assumed for all BMPs. The average cost of TP reduction ($/lb.) was
assumed to represent the minimum price that a farmer would be willing to accept to sell his/her
TP credit in the nutrient market.
Nutrient Trading Scenarios
A total of four different nutrient credit market scenarios (Table 2) were examined by
considering: two trading ratios between point and nonpoint sources (1:1 and 2:1) and two
baseline requirements for agricultural sources (existing BMPs level and 100% nutrient
management plan). The aggregate demand and supply functions were estimated for TP under
these four nutrient trading scenarios. A graphical method was applied to find out the equilibrium
market price under each scenario. Total costs with and without TP trading market and costs
saving in TP trading market for each WWTP were estimated. We also estimated aggregate cost
saving across society with a TP trading market under four nutrient trading scenarios.
Page 15
15
Results and Discussion
Estimated TP Loading in the Watershed
A validated and calibrated MapShed model was used to estimate existing TP loadings
considering the current level of BMPs on crop and pasture/grass lands. The water quality model
was validated by comparing estimated versus observed monthly data for stream flow. The Nash-
Sutcliff coefficient of model efficiency for the calibrated model was 0.63 and R2 value in
regression analysis between observed and estimated was 0.71. A null hypothesis of no difference
between observed and estimated monthly stream flows in the simulation period cannot be
rejected. Our water quality model was not calibrated based on observed and simulated TP
concentration data. Sufficient monthly monitored data for TP concentration were not available
for the Greenbrier River to compare observed and simulated results. TP loading estimates
represented average annual loading over a 21 year period (1990 to 2011).
The MapShed model gave an estimated TP loading from all sources of 375,421 pounds
per year. Agricultural sources were projected to be the largest single contributor at 56% of this
total and WWTPs contributed an estimated 10% of TP loadings. The remaining discharges come
from the forest lands, wetlands, and groundwater sources. Existing BMPs were estimated to
reduce total TP loadings by about 15% in the watershed.
Potential TP Demand and Supply
The total annual amount of TP load reduction requirement for each WWTP in the
watershed was estimated based on a 0.5 mg/l effluent standard for WWTPs in the watershed,
their current level of nutrient concentration in mg/l, daily nutrient discharges in pounds (lb.), and
each facility’s discharge flow in million gallons per day (MGD). Table 3 presents existing TP
loads from the seven NPDES permitted WWTPs, load reduction requirements at 0.5mg/l TP
Page 16
16
limit and average costs ($/lb.) of additional TP reduction for each WWTP. Total TP reduction
requirements for the WWTPs in the watershed was estimated to be 22,428 lb./year. This TP
reduction requirement for WWTPs was considered as the potential aggregate TP credits that
could be utilized in the watershed. The average per unit cost ($/lb.) of additional TP load
reductions from the WWTPs ranged between $49 and $877. The weighted average (based on
discharge flows) of TP reduction costs across all seven WWTPs was $114 per lb.
We considered upstream-only TP credit trading between WWTPs and farmers in the
watershed. In the upstream-only trading system, buyers (WWTPs) can purchase TP reduction
credits only from the upstream sellers (farmers). This assumption avoids the potential for
development of nutrient “hotspots”. The PRedICT tool in the MapShed was used to simulate
different BMP scenarios at 14 sub-watershed levels in order to estimate TP credit supplies.
BMPs were selected by per pound cost with lowest being first into the MapShed simulation
assuming that farmers will prefer to implement low cost BMPs initially to generate nutrient
credits. Each WWTP’s reduction requirement was simulated with a different upstream BMP
scenario to meet its targeted TP load reduction.
Table 4 presents TP credit supplied from the agricultural sources under two baseline
requirements (existing BMPs and 100% NMP) for agricultural sources and two trading ratios
(1:1 and 2:1) between agricultural sources and WWTPs. The pasture and grassland was a
dominant source of TP credits in the Greenbrier River watershed. Nutrient management plans
provided the largest amount of TP credits (70 to 88%) under existing BMPs baseline, while
rotational grazing provided most the TP credits (77 to 88%) under a NMP baseline.
Implementation of four agricultural BMPs; cover crop and nutrient management plan
on crop land and nutrient management plan and rotational grazing on pasture/grass land could
Page 17
17
generate sufficient TP credits to fulfill WWTP’s TP reduction requirements at 0.5 mg/l TP limit
in the watershed (Table 5). As low cost BMPs were selected first in the simulation process,
farmers were assumed to implement nutrient management plans and cover crops 100% of crop
lands (including existing BMPs) under the existing BMP baseline in order to generate low cost
TP credits to supply in a credit trading market. Nutrient management plans on pasture and grass
lands ranged from 15 to 36% coverage depending upon the trading ratio. Similarly, from 34%
and 65% of pasture and grassland would need to implement rotational grazing on the watershed
with cover crop on 100% of crop lands to fulfill WWTP’s TP reduction requirements under the
100% NMP baseline requirement (Table 5).
The average costs of nutrient reduction from BMP implementation on crop and
pasture/grass lands are presented in Table 6. The average per unit cost ($/lb.) of TP reduction
from BMPs on crop and pasture/grass lands ranged between $33 and $57 with an existing BMPs
baseline and increased to $51 and $339 under a 100% NMP BMPs baseline. As expected, when
all farmers must implement a NMP on their agricultural land before generating TP credits, BMP
costs per lb. of TP reduction rise. Total cost of TP credits generation included both costs of
NMP baseline requirement and cost of implementing additional BMPs (e.g. cover crop or
rotational grazing). Out of four BMPs considered in this study, nutrient management plans on
crop land generated TP credits at lowest cost followed by cover crop on the crop land, nutrient
management plan on the pasture land, and rotational grazing on the pasture land, which was the
most expensive BMP to generate TP load reductions. The costs of TP reductions from
agricultural sources were considerably lower that the costs of TP reductions for most WWTPs
(Tables 3 and 6).
Page 18
18
Cost Savings from TP Trading
We estimated aggregate costs of compliance of TP standard (0.5mg/l) for each WWTP
under without and with TP credit trading scenarios. Table 7 presents total compliance costs
when: (1) WWTPs must upgrade their treatment plant with no credit market, and (2) when
WWTPs can choose between an upgrade or purchase of TP credits in a market. The four market
scenarios are presented in this table and compared to plant upgrades to compute cost savings.
The total compliance cost for all seven WWTP upgrades to meet 0.5 mg/l TP standard in the
absence of TP credits market was $2.4 million per year. This WWTP upgrading costs to meet 0.5
mg/l TP standard ranges between $0.19 million and $0.53 million.
Equilibrium credit prices were computed for TP credits in each trading market scenario.
The equilibrium price of TP credits in ranged from $52 to $239 per lb. of TP. In a market, each
WWTP was assumed to minimize compliance costs by choosing either a treatment plant upgrade or
purchase of TP credits to meet the new TP standard. The percentage of cost savings with a TP credit
market ranged from 23% to 50%. Two WWTP (Union Public Service District and the City of
Marlinton) received the greatest cost savings from participation in credit markets. The largest
cost savings ($1.21 million/year) occurred under current BMPs baseline requirement and a 1:1
trading ratio. Six out of seven WWTPs were projected to participate in this market.
Cost savings declined with higher baseline requirements (100% NMP versus existing
BMPs) and higher trading ratios (2:1 versus 1:1). Even under the most stringent baseline (100%
NMP) and least favorable trading ratio (2:1), there was a 23% cost savings from implementation
of a credit market. These results were consistent with literature results that showed high trading
ratios and baseline requirement for agricultural sources reduced the cost saving from a nutrient
trading market (Gosh et al. 2011; Smith et al. 2010; Fang, et al. 2005).
Page 19
19
Conclusions
Both physical and economic feasibility are essential for a successful nutrient trading
program in a watershed. A combination of hydrological and economic analyses has been
presented here to identify the feasibility of total phosphorus (TP) credit trading program in the
Greenbrier River basin. Estimates from the hydrological model indicated that the agricultural
non-point sources contribute the majority of TP in the watershed. Implementation of agricultural
BMPs on crop and pasture/grasslands could generate enough credits to compensate WWTPs’ TP
reduction requirements to meet a proposed TP discharge standard of 0.5mg/l. However, 100%
cropland participation in BMP implementation would be required in the watershed along with
substantial pasture and grassland participation, up to 65%. We conclude that the TP trading is
physically feasible in the watershed.
TP trading in the Greenbrier River watershed has the potential to provide cost savings
over treatment plant upgrades to meet a 0.5 mg/l TP standard for WWTPs. About 86% of
WWTPs in scenario 1, 57% of WWTPs in the scenarios two and three, and 29% of WWTPs in
the scenario four could experience cost savings compared to treatment plant upgrade costs by
purchasing TP credits in a nutrient market.
The cost saving analyses for the individual WWTP in four different nutrient trading
scenarios indicated that all four scenarios showed lower costs with TP credit markets versus
without a market, although the degree of WWTP participation differed. With a 1:1 trading ratio
and existing BMPs baseline, a TP credit market resulted in a projected cost saving of $1.21
million annually. The cost savings were lowered (only 23%) and most WWTPs would not
participate in a TP credit trading market under less favorable market conditions (a 2:1 trading
ratio and 100% NMP baseline). Thus, in the case of TP in the Greenbrier River, policy makers
Page 20
20
could select any of the four market scenarios and there would still be potential cost savings over
having no market. However, cost savings have the potential to be twice as great with lesser
baseline requirements (current BMPs vs. 100% NMP) and a higher trading ratio (1:1 vs. 2:1).
The more stringent baseline requirement of 100% NMP raises market compliance costs slightly
more than a lower trading ratio (2:1).
This paper presented how a combination of hydrological and economic methods can be
applied to assess the feasibility of nutrient trading in a watershed. This research assumed that all
farmers and WWTPs will participate in the potential nutrient trading program in the watershed.
In reality, they might be reluctant to participate in the WQT program despite the potential
benefits from the participation. There are a number of attitudes or values that cause farmers to be
reluctant adopting conservation practices. A strong pride in private property, a history of tensions
with industrial actors, or a desire to be recognized for land stewardship are few of the attitudes
and values that can establish powerful norms of behavior discouraging trades (Breetz et al. 2005;
Mariola 2009). This study did not quantify the effects of those variables in a nutrient trading
program. Research is needed to explore the potential impacts of such variables on the nutrient
trading market.
Acknowledgements
This research was conducted under a project entitled WRI 111 – Kanawha River Basin
Nutrient Trading Feasibility Assessment funded by the U.S. Environmental Protection Agency.
The authors wish to thank Dr. Michael Strager for his help in water quality modeling.
Page 21
21
References:
Alexander, R. B., R. A. Smith, G. E. Schwarz, E. W. Boyer, J. V. Nolan, and J. W. Brakebill.
2008. “Differences in Phosphorus and Nitrogen Delivery to the Gulf of Mexico from the
Mississippi River Basin”. Environmental Science & Technology 42 (3): 822–830.
Breetz, H. L., K. Fisher-Vanden, H. Jacobs, and C. Schary. 2005. “Trust and Communication:
Mechanisms for Increasing Farmer's Participation in Water Quality Trading”. Land
Economics, 81(2): 170-190.
Breetz, H. L., K. Fisher-Vanden, L. Garzon, H. Jacobs, K. Kroetz, and R. Terry. 2004. Water
Quality Trading and Offset Initiatives in the U.S.: A Comprehensive Survey. Dartmouth
College, New Hampshire.
Chesapeake Bay Program (CBP). 2002. Nutrient Reduction Technology Cost Estimations for
Point Sources in the Chesapeake Bay Watershed. Chesapeake Bay Program, Annapolis,
MD.
Conservation Technology Information Center (CTIC). 2011. Wabash River Watershed Water
Quality Trading Feasibility Study Final Report. Conservation Technology Information
Center. West Lafayette, IN.
Diaz, R., and R. Rosenberg. 2008. “Spreading Dead Zones and Consequences for Marine
Ecosystems.” Science 321(5891): 926–929.
Evans, B. M., and K. J. Corradini. 2012. MapShed Version 1.0 Users Guide. Penn State Institutes
of Energy and the Environment, Pennsylvania State University, PA.
Faeth. P. 2000. Fertile Ground: Nutrient Trading Potential to Cost Effectively Improve Water
Quality. World Resources Institute, Washington DC.
Page 22
22
Fang, F. K., W. Easter, and P. L. Brezonik. 2005. “Point-Nonpoint Source Water Quality
Trading: A Case Study in the Minnesota River Basin”. Journal of the American Water
Resources Association 41: 645-58.
Freeman, A. M. 2003. The Measurement of Environmental and Resource Values: Theory and
Method (2nd
edition). Resources for the Future, Washington, DC.
Ghosh, G. M. Ribaudo, and J. Shortle. 2011. “Baseline Requirements Can Hinder Trades in
Water Quality Trading Programs: Evidence from the Conestoga Watershed”. Journal of
Environmental Management 92: 2076-84.
Horan, R. D., and J. S. Shortle. 2011. “Economic and Ecological Rules for Water Quality
Trading”. Journal of the American Water Resources Association, 47: 59-69.
Horan, R. D. and J. S. Shortle. 2005. “When Two Wrongs Make a Right: Second-Best
Point/Nonpoint Trading Ratios”. American Journal of Agricultural Economics, 87 (2):
340-352.
Houtven, G. V., R. Loomis, J. Baker, R. Beach, and S. Casey. 2012. “Nutrient Credit Trading for
the Chesapeake Bay: An Economic Study”. Chesapeake Bay Commission, Annapolis,
MD.
Kardos, J. S., and C.C. Obropta. 2011. Water Quality Trading Model Uncertainty Analysis of a
Point-Nonpoint Source Phosphorus Trading Program. Journal of the American Water
Resource Association, 1-12.
Khatri-Chhetri, A. 2012. A Feasibility Assessment of Water Quality Trading in the Greenbrier
River Watershed, West Virginia. Ph.D. Dissertation, West Virginia University.
Page 23
23
King, D. M., and P. J. Kuch. 2003. “Will Nutrient Credit Trading Ever Work? An Assessment of
Supply and Demand Problems and Intuitional Obstacles”. Environmental Law and
Review, 33: 10352-10368.
Lopez, C.B., Jewett, E.B., Dortch, Q., Walton, B.T., Hudnell, H.K. 2008. Scientific Assessment
of Freshwater Harmful Algal Blooms. Interagency Working Group on Harmful Algal
Blooms, Hypoxia, and Human Health of the Joint Subcommittee on Ocean Science and
Technology. Washington, DC.
Malik, A., D. Letson, and S. Crutchfield. 1993. “Point/Nonpoint Source Trading of Pollution
Abatement: Choosing the Right Trading Ratio”. American Journal of Agricultural
Economics, 75: 959-967.
Mariola, J. M. 2009. Are Markets the Solution to Water Pollution? A Sociological Investigation
of Water Quality Trading. PhD Dissertation Submitted to the Graduate Program in Rural
Sociology, Ohio State University.
Morgan, C., and Wolveton, A. 2005. “Water Quality Trading in the United States”. Working
Paper No 05-07. U.S. Environmental Protection Agency, National Center for
Environmental Economics, Washington DC.
Natural Resources Conservation Service (NRCS). 2012. WV 2012 EQIP Payment Schedule.
West Virginia Natural Resources Conservation Service, Morgantown, WV.
Newburn, D. A., and R. T. Woodward.2012. “An Ex Post Evaluation of Ohio’s Great Miami
Water Quality Trading Program”. Journal of the American Water Resources Association
48: 156-69.
Ribaudo, M. O., and C. J. Nickerson. 2009. "Agriculture and Water Quality Trading: Exploring
the Possibilities." Journal of Soil and Water Conservation, 64(1):1-7.
Page 24
24
Sado, Y., and R. N. Boisvert. 2007. “Potential Cost Savings from Discharge Permit Trading to
meet TMDLs for Phosphorus in the Passaic River Watershed”. Selected Paper prepared
for Presentation at the American Agricultural Economics Association Annual Meeting,
Portland, OR, July 29-August 1, 2007.
Selman M., and S. Greenhalgh. 2009. Eutrophication: Policies, Actions, and Strategies to
Address Nutrition Pollution. WRI Policy Note No. 3, World Resource Institute,
Washington D.C.
Selman, M., S. Greenhalgh, E. Branosky, C. Jones, and J. Gulling. 2009. Water Quality Trading
Programs: An International Overview. WRI Issue Brief No. 1. World Resource Institute,
Washington, D.C.
Shabman, L., and K. Stephenson. 2007. “Achieving Nutrient Water Quality Goal: Market- Like
Principles to Water Quality Management”. Journal of the American Water Resources
Association 43(4): 1076‐1089.
Smith, C. M., J. M. Peterson, and J. C. Leatherman. 2010. “Simulation of Factors Impeding
Water Quality Trading Market Performance”. Paper presented at the Agricultural &
Applied Economics Association 2010 AAEA, CAES, & WAEA Joint Annual Meeting,
Denver, Colorado, July 25‐27, 2010.
Summers, J. 2008. Assessment of Filamentous Algae in the Greenbrier River and other West
Virginia Streams. West Virginia Department of Environmental Protection, Charleston,
WV.
Tietenberg, T. T. 2007. “Tradable Permits in Principle and Practice”. In J. Freeman and C. D.
Kolstad (eds), Moving to Markets in Environmental Regulation: Lessons from Twenty
Years of Experience. New York: Oxford University Press.
Page 25
25
United State Environmental Protection Agency (USEPA). 2008. Handbook for Developing
Watershed Plans to Restore and Protect Our Waters. EPA 841-B-08-002, Office of
Water, Washington, D.C.
United States Environmental Protection Agency (USEPA). 2007. “Water Quality Trading
Toolkit for Permit Writers”. EPA-833-R-07-004, Office of Water Management,
Washington, D.C.
United States Environmental Protection Agency (USEPA). 2008. Clean Watersheds Needs
Survey 2004 Report to Congress. Office of Water, Washington, D.C. Online available at:
http://www.epa.gov/cwns/2004rtc/cwns2004rtc.pdf
United States Environmental Protection Agency (USEPA). 2010. “Discharge Monitoring Report
(DMR) Pollutant Loading Tool”. Office of the Water, Washington D.C. Online available:
http://cfpub.epa.gov/dmr/
West Virginia Department of Environmental Protection (WVDEP). 2010. West Virginia
Integrated Water Quality Monitoring and Assessment Report 2010. Division of Water
and Waste Management, Charleston, WV.
West Virginia Department of Environmental Protection (WVDEP). 2011. Greenbrier River State
of the Watershed 2011. Division of Water and Waste Management, Charleston, WV.
West Virginia Water Research Institute (WVWRI). 2008. DRAFT West Virginia Water Quality
Nutrient Credit Trading Program Guidelines. West Virginia University, Morgantown,
WV.
West Virginia Water Research Institute (WVWRI). 2008. DRAFT West Virginia Potomac River
Basin Water Quality Nutrient Trading Program. West Virginia University, Morgantown,
WV.
Page 26
26
Figure 1. Location of Greenbrier River watershed
Page 27
27
Table 1: GIS data layers prepared to develop water quality model in the MapShed.
File Name Description Data Source
Shape Files
Watershed Basin Basin boundary used for modeling
(polygons)
Natural Resource Analysis
Center, WVU
Streams Map of stream network (lines) US Geological Survey
Soils Soil characteristics data (polygon) Geospatial Data Gateway of
USDA-NRCS
Point sources
Point source discharge locations
(points)
USEPA
Weather stations
(Rainfall and
Temperature data)
Weather station locations (points) US National Climatic Data
Center
Grid Files
Land use/cover Map of land use/cover classes Natural Resource Analysis
Center, WVU
Elevation
Digital Elevation Model (DEM) file
WVGISTC 2012
Table 2: Descriptions of total phosphorus (TP) credit market scenarios.
Trading
Scenario
Trading Ratioa TP Standard for WWTPs
(mg/l)
Baseline Requirements
for Agricultural Sources
1 1:1 0.5 Existing BMPs level
2 1:1 0.5 Existing BMPs level
3 2:1 0.5 100% NMP
4 2:1 0.5 100% NMP
Note: NMP = Nutrient Management Plan
a Two pounds of phosphorus reduction from the agricultural sources equal to one credit in 2:1
trading ratio.
Page 28
28
Table 3: Estimated annual total phosphorus (TP) load reduction requirements at 0.5mg/l TP
standard for NPDES permitted WWTPs in the Greenbrier River watershed.
Wastewater
Treatment Plant
Annual TP
Load
(lb.)
Annual TP Load Reduction
Requirement
(lb.)
Average Cost of TP Load
Reductions
($/lb)
Town of Alderson 2,783 2,357 108
City of Ronceverte 13,332 7,502 63
Union Public Service
District 11,780 608 877
Pence Springs 5,872 5,141 49
City of White Sulfur
Springs 6,815 4,819 95
Town of Hillsboro 1,247 1,149 168
City of Marlinton 1,172 852 319
Weighted Average
114
Note: Weighted based on WWTP’s discharge flow (MGD)
Table 4: Total phosphorus credit supply from the agricultural sources under two baseline
requirements for agricultural sources and trading ratios.
Best Management
Practice
Baseline: Existing BMPs Baseline: 100% NMP
TR: 1:1 TR: 2:1 TR: 1:1 TR: 2:1
lb. / year
Cover crop 4,387 2,194 5,151 2,576
NMP (crop land) 2,409 1,205
NMP (pasture/grass
land)
15,697 19,040
Rotational grazing 17,293 19,855
Total
22,493
22,438
22,444
22,430
Note: TR = Trading Ratio, NMP = Nutrient Management Plan
Page 29
29
Table 5: Estimated BMPs coverage (%) on all agricultural lands in the Greenbrier River
watershed under four total phosphorus credit trading scenarios.
Best Management Practice (BMP) Baseline: Existing BMPs Baseline: 100% NMP
TR: 1:1 TR: 2:1 TR: 1:1 TR: 2:1
NMP (crop land) 76% 76% - -
Cover crop 85% 85% 100% 100%
NMP (pasture/grass land) 15% 36% - -
Rotational grazing - - 34% 65%
Note: TR = Trading Ratio, NMP = Nutrient Management Plan
Table 6: Average per unit cost of total phosphorus (TP) reduction from BMPs on crop and
pasture/grass lands
Best Management Practice
(BMP)
TP ($/lb.)
Baseline: Existing BMPs
TP ($/lb.)
Baseline: 100% NMP
Nutrient Management Plan
(NMP) (crop land)
32.99
(12.07 - 46.54)
Cover crop 36.75
(13.28 – 49.50)
51.44
(18.56 – 67.89)
NMP (pasture/grass land) 56.72
(37.02 – 97.82)
Rotational grazing 339
(219-371)
Note: Values in parenthesis indicate cost range.
Page 30
30
Table 7: Compliance costs for WWTP upgrades and cost in a total phosphorus (TP) credit
market.
Wastewater
Treatment Plant
(WWTP)
No TP Credit Market TP Credit Market Available
Compliance costs for
WWTP upgrades to
meet a TP standard
($)
Costs in a nutrient market to purchase TP credits
or WWTP upgrades to meet a TP standard ($)
Baseline: Current BMPs Baseline: NMP
1:1 2:1 1:1 2:1
Town of Alderson 253,507 131,992 226,272 253,378 253,507
City of Ronceverte 470,246 420,112 470,246 470,246 470,246
Union Public
Service District 533,265 34,048 58,368 65,360 103,968
Pence Springs 254,120 254,120 254,120 254,120 254,120
City of White Sulfur
Springs 456,441 269,864 456,441 456,441 456,441
Town of Hillsboro 192,616 64,344 110,304 123,518 192,616
City of Marlinton 272,059 47,712 81,792 91,590 145,692
Totals 2,432,254 1,222,192 1,657,543 1,714,653 1,876,590
% of cost saving
compared to no
market
49.75 31.85 29.50 22.84