Abstract
Introduction
Objectives
Methods
Stream data archive
Water balance and nutrient transport modeling
Web interface
Results
GIS database
Stream data archive
Water balance and nutrient transport modeling
Web interface
Links to other related projects
Discussion
Technology transfer
Technology commercialization
Scientific and academic achievement
Literature cited
(c)terrestrial mobilization and riverine transport / processing models to route water and water borne materials into and through river corridors; and,
(d)a
World Wide Web server that acts as an online reference source for digital
map and station-based monitoring data (www.gm-wics.unh.edu).
This poises us for additional detailed analysis of nutrient fluxes to the Gulf of Maine coastal zone. In the follow-up project (The Gulf of Maine Watershed Information and Characterization System (GM-WICS): Outreach Activities Using Innovative Information Technology), also funded by CICEET, we intend to continue to carry-out nutrient transport modeling for basins within the Gulf of Maine Watershed, improve model validation datasets and increase regional understanding of the seasonality of nutrient transport and snowpack nutrient storage through snow and river sampling in New Hampshire and Maine watersheds. Through a companion project with the Plum Island Ecosystem Long Term Ecological Research Group, we will focus on these issues for the Ipswich and Parker Rivers in northern Massachusetts.
Keywords:Gulf
of Maine Watershed, Nutrient Flux, Pollution Sources
The
management of inland landscapes and freshwater resources is a key determinant
of coastal zone water quality, but one that is still poorly understood
(Hobbie 2000, NAS 1994). Within this context, river loadings of biotically-active
elements are known to have increased several-fold since the beginning of
the Industrial Era (Nixon 1995, Smil 1990). And, from numerous individual
river basin and coastal zone studies (e.g. the Baltic region, Rosenberg
et al. 1990; Mississippi River / Gulf of Mexico, Ortner and Dagg 1995,
Turner and Rabelais 1994; Northern Adriatic, Justic et al. 1995a,b; the
Black Sea, Mee 1992) we know that excessive river borne nutrients, shifts
in nutrient limitation, coastal eutrophication, toxic phytoplankton blooms,
and bottom-water hypoxia go hand-in-hand. Loadings associated with population
growth and the byproducts of economic development (i.e. non point source
inputs from agriculture and atmospheric deposition, sewage loadings, etc)
are now an important contributor to riverine fluxes of nutrients to the
coastal zone. The quantitative significance of such fluxes is not restricted
to the local scale. Indeed, they have been shown to be significant over
continental domains and at the regional scale as exemplified by the Northeastern
US (Howarth et al. 1996).Because
of its high population density and industrial activity, New England provides
an excellent opportunity for better understanding how human-dominated landscapes
ultimately interact with coastal ecosystems over a multi-state domain.
By their very nature, rivers display spatial variability in terms of both discharge and biogeochemical flux. From a scientific perspective our work seeks to improve understanding of how drainage basin attributes contribute to the observed patterns of fluvial transport. We seek to identify which key determinants – mappable using our GIS systems – can account for these patterns. In addition, rivers display an enormous degree of temporal variability that is of direct consequence to coastal receiving waters. The variations are river –specific and are a manifestation of drainage basin characteristics which themselves may change substantially over time. For example, for suspended sediment flux, the key controlling factors are both static (i.e. drainage basin area, geology and soils, presence of wetlands and lakes) and time-varying (i.e. precipitation and snowfall patterns, land management, creation of impoundments) (Milliman and Syvitski 1992, Hadley and Walling 1985).
Despite the importance of land-coastal zone interactions, we did not have, for the Gulf of Maine, a consistent system for monitoring and assessing the dimensions of such change at anything but the local scale, for example at NSF Land-Margin Ecosystem Study (LMER) or National Estuarine Research Reserve (NERR) sites. Our primary aim in this proposal was to develop a diagnostic GIS-based methodology to monitor and interpret the status of the Gulf of Maine watershed with respect to watershed characteristics and to the fluvial transport of water and nutrients.
Our initial focus was on developing this overall framework for three test sites: the Great Bay NH and Wells ME Estuarine Reserves, and the Parker River/Plum Island Sound Land-Margin Ecosystem Study Area in northern Massachusetts. This initial development effort proceeded in close conjunction with the proposed work of F. Rubin (CICEET project 22, 1997-2000) to develop land cover templates and that of T. Loder (CICEET project 11, 1997-1999) to monitor river discharge and nutrient chemistry in these study sites.
Our overall goal was to produce a synoptic picture of the status of fluvial inputs to the Gulf of Maine coastal zone and support this activity with a GIS-based analysis of how riverine biogeochemical fluxes are influenced by seasonal and interannual variability as well as anthropogenic activities that take place within drainage basins. The capacity for monitoring and interpretation are necessary precursors for correctly understanding changes in the long-term state of the region’s coastal zone and formulating strategies for protecting these resources, which increasingly are subject to human disturbance.
In order to accomplish this goal we focused on four objectives.
1.Development of a data bank of relevant GIS-based data sets from which individual drainage basins of the Gulf of Maine were classified and characterized. Synoptic statistics related to population density, land use, soils, level of industrial activity, for example, were also be assembled.
2.Bringing together Canadian and United States data to compile an archive of available discharge and water chemistry data sets that serve as model calibration / validation targets.
3.Creation of terrestrial mobilization and riverine transport / processing models to route water and water borne materials into and through river corridors.
4.Development of a World Wide Web server and data examination tools that act as online references for digital map and station-based monitoring data. Providing an interface by which the state of Gulf of Maine drainage systems can be assessed by scientists, coastal managers, and citizen groups.
During the early phases of the project we focused on the approach of White et al. (1992) due to its ability to treat point and non-point sources from urban and non-urbanized watersheds. We also considered specific elements of other distributed and semi-distributed drainage basin models, such as the EPA BASINS modeling framework, the EPA NPSM (Non-Point Source Model) N), HSPF (Johnson et al. 1980), ANPS (Young et al. 1987), and ANSWERS (Beasley et al. 1980). Our initial focus was on the SPARROW methodology, applied by Smith et al. (1994), Smith et al. (1993), and White et al. (1992). This methodology uses a straightforward set of constitutive equations applied to each element of the simulated river networks. For dissolved biotically active chemicals in rivers,White et al. (1992) assume a simple first-order decay.
After careful consideration of existing models, we decided to develop our own nutrient transport model. Our detailed review of several prominent models (Brown & Barnwell, 1987; Johnes, 1996; Smith et al., 1997; Johanson & Imhoff, 1984) determined these models to be inadequate for our purposes. The weaknesses include issues such as the need for inordinate quantities of detailed nutrient inventory data (Johnes, 1996; Export Coefficient Model), mismatch between spatially-detailed but temporally static statistical flux models which give only mean annual flux estimates (Smith et al. 1997; SPAtially Referenced Regression On Watershed attributes (SPARROW)), complex process-based kinetics for instream processes to the virtual exclusion of land-based nutrient loading (Brown & Barnwell, 1987; Enhanced Stream Water Quality Model (QUAL2E)), and highly parameterized, vector representations of water quality processes (Johanson & Imhoff, 1984; Hydrological Simulation Program-Fortran (HSPF)).
The NTM hydrology sub-model is based on our existing Water Balance Model (WBM). Detailed water budget calculations trace precipitation (as rain or snow), through snowpack growth and decay, soil recharge and drainage, plant canopy interception and transpiration, surface runoff, groundwater recharge and discharge, and stream flow. NTM simulations using site-specific parameterizations have produced encouraging results for several Gulf of Maine watersheds including the Oyster and Lamprey Rivers in NH, the Parker and Ipswich in MA, and the Androscoggin/Kennebec in ME. An NTM simulation of three years (1985-1987) resulted in an underestimate of 35 mm in cumulative runoff, representing a less than 3% difference from the observed value.
The NTM nutrient sub-model (Figure 1) predicts nitrogen flux at the mouth of a watershed by linking measured hydrologic and nitrogen (N) inputs, the hydrology sub-model, and results of a process-based terrestrial ecosystem model. Nitrogen deposition occurs as both wet and dry deposition to the soil surface unless snowpack is present, in which case N is sequestered in the pack until snowmelt begins.Nitrogen uptake and mineralization are determined via statistical relationships with temperature and soil moisture. These relationships were derived from the well- validated terrestrial ecosystem model, PnET-CN (Aber et al., 1997; Aber and Driscoll, 1997).Nitrate is dissolved in soil water, so leaching to groundwater or runoff to wetlands, lakes, and streams is controlled by hydrologic turnover times.The system of differential equations in both the hydrology and nutrient sub-models are solved via an adaptive Runge-Kutta differential equation solver.
Figure
2 shows the current hydrologic and nutrient flux results for the Lamprey
River watershed, Rockingham County, NH (474 km2). Graph A shows
mirror images of the observed runoff and NTM predicted runoff. Graph B
shows NTM predicted nutrient flux. The current lumped-watershed version
of NTM, which considers upland connected to downslope wetlands, is being
further developed, through our CICEET 2000 project and our work on the
Plum Island Ecosystem (PIE) Long Term Ecological Research (LTER) project,
in a spatially distributed GIS framework so that it may be used to evaluate
the influence of land use and land-use topology on water and nutrient runoff.
Archival time series data sets stored in GM-WICS for runoff, precipitation,
and air temperature and gridded fields of precipitation, air temperature,
elevation, soils, and land cover can be used as inputs to the NTM.
Work
under our current CICEET project (Grant #NA97OR0338) has combined (i) data
development, (ii) GIS analysis, and (iii) modeling efforts.Data
development and GIS analysis efforts have focused on creating a GIS database
for the entire Gulf of Maine Watershed. Currently, the Gulf of Maine Information
and Characterization System (GM-WICS) contains data sets describing the
soils, vegetation, river network typology, meteorological data, and stream
discharge for the Gulf of Maine Watershed. Nutrient input and export data
is also being collected and added to this database. This database provides
the necessary input layers for spatially-distributed watershed modeling
using the Water Balance Model (Vörösmarty et al., 1998), Water
Transport Model (Fekete et al., 2000; Vörösmarty et al., 1989)
and Nutrient Transport Model (under development). Modeling efforts have
focused on construction and testing of a Nutrient Transport Model (NTM)
which combines a physically-based extension of the WBM/WTM hydrology model
with a nutrient biogeochemistry module. Currently, NTM is being developed
to simulate nitrogen transport, and will be expanded to model other constituents.
As
an aid to model development, we have been collecting time series data from
local coastal rivers to determine nutrient loads into the Great Bay NERR.
This monitoring has been done through the current CICEET-funded project
by the Co-I (T. Loder: Innovative Technologies for the Measurement of Fluvial
Inputs of Nutrients to Estuarine Systems, NOAA Grant #NA97OR0338). Figure
3 shows fifteen-day running averages of the dissolved inorganic nitrogen
concentration for the Lamprey, Salmon Falls and Androscoggin/Kennebec Rivers.
The Lamprey River and Androscoggin /Kennebec River graphs (Figs 3a and
3c) show seasonal variations in concentration, as would be expected from
many previous studies, e.g. Davis and Keller (1983) and Webb and Walling
(1985). The Salmon Falls River (Fig 3b) concentrations are much higher
and do not show the seasonal variations.Such
behavior is characteristic of landscapes with high nutrient loading (Stoddard,
1994) and indeed this river is known to have high nitrogen inputs.
To
support project outreach efforts, we also have been working with colleagues
from the Marine Biological Laboratory in Woods Hole (C. Hopkinson, E. Rastetter,
J. Vallino; NSF-LTER Grant #OCE-9726921) to monitor and understand nutrient
outflows from the Parker and Ipswich River systems into Plum Island Sound,
MA. In addition, we have completed four snow and river sampling trips to
characterize the nutrient storage of snowpack in the Androscoggin/Kennebec
basins and are working with volunteers from the Friends of Merrymeeting
Bay, and seven schools in the Androscoggin/Kennebec basins to develop additional
river and snowpack time series data sets.
In
the case of the Parker and Ipswich Rivers, similar per unit area rates
of nutrient export for nitrogen have been recorded (B.J. Peterson, MBL,
unpublished data) despite the differences between these basins with respect
to point and non-point source loadings.The
trapping and loss of nutrients inside wetlands is a possible mechanism
(Jacks et al., 1994; Jansson et al., 1994) and we are exploring how the
spatial organization of the upstream landscape ultimately defines the mouth-of-river
fluxes.These site-specific results
form the focus for our NTM simulation studies, and are important in providing
guidance in our effort to expand our understanding to larger landscape
units including the Androscoggin/Kennebec and ultimately the entire GOM
watershed. For example, it was found in the Lamprey and Oyster Rivers that
the majority of the annual nitrogen loading occurs during the winter and
snowmelt time frame, as it does in the larger and more spatially complex
Androscoggin/Kennebec.
We
believe the inclusion of regional scale studies is important since such
work will help to identify critical and emerging problem areas. This has
been a central focal point of our work. This is especially important as
the region continues to develop economically. We believe that important
insights can be gained by comparative monitoring/modeling studies of river
basins and the manner in which they load constituents into the coastal
zone (Hobbie 2000, National Academy of Sciences 1994, Boynton et al. 1982).
The products of this research have not been patented, copyrighted, or licensed in anyway. We are currently requesting permission to disseminate Canadian hydrologic and climatic data as a part of our Gulf of Maine Watershed Web site.
Undergraduate Student (senior honors project):
Matt Hill, Civil Engineering Department, University of New Hampshire
Graduate Student:
Kim Bredensteiner, Earth Science Department, University of New Hampshire
The data sets and finding presented above have been summarized and presented as a set of documents. These are listed below:
Reports
and Papers
Vörösmarty,
C.J. and B.J. Peterson. 2000. Macro-scale models of water and nutrient
flux to the coastal zone. In: J. Hobbie (ed). Estuarine Science: A Synthetic
Approach to Research and Practice. Island Press.
Bredensteiner,
K., C. Vörösmarty, T. Loder, E. Penfold, and W. Wollheim. 2000.
Regional Estimation of Snowpack Nutrient Storage. (In preparation for submission
to Estuaries).
Bredensteiner,
K. 2000. Gulf of Maine Watershed Information and Characterization System
Database Development (Update of 1999 M. Hill report). Water Systems Analysis
Group Technical Paper. University of New Hampshire, Durham NH.
Hill, M. 1999. Data Development for the Gulf of Maine Basin Characterization System. Senior Honors Project Final Report. Civil Engineering Department, University of New Hampshire, Durham NH.
Presentations
Bredensteiner,
K., C. Vörösmarty, T. Loder, E. Penfold, and W. Wollheim. 2000.
Regional Estimation of Snowpack Nutrient Storage. Poster. Long Term Ecological
Research Network All Scientists Meeting. Snowbird, UT.
Bredensteiner, K. 1999. Overview of Gulf of Maine Project. Plum Island Ecosystem. Long Term Ecological Research Site. All Scientist Meeting. Marine Biological Lab, Woods Hole MA.
Beasley,
D.B., L.F. Huggins, and E.J. Monke. 1980. ANSWERS: A model for watershed
planning. Transactions of the ASAE 23: 938-944.
Billen,
G., C. Lancelot, and M. Meybeck. 1991. N, P, and Si retention along the
aquatic continuum from land to ocean. In: R.F.C. Mantoura, J.M. Martin,
and R. Wollast (eds.) Ocean margin processes in global change. New York:
John Wiley & Sons Ltd.
Boynton,
W.R., W.M. Kemp, and C.W. Keefe. 1982. A comparative analysis of nutrients
and other factors influencing estuarine phytoplankton production. In: Kennedy,
V.S. (ed). Estuarine Comparisons. Academic Press. New York.
Bredensteiner,
K., C. Vörösmarty, T. Loder, E. Penfold, and W. Wollheim. 2000.
Regional Estimation of Snowpack Nutrient Storage. (In preparation for submission
to Estuaries).
Fekete, B.M., C.J.
Vörösmarty, andR. Lammers.
2000.Multiple-resolution stream
networks for macro-scale hydrology: Algorithm development and analysis.Hydrological
Processes.(In Preparation).
Hadley,
R.F. and D.E. Walling (Eds.). 1985. Erosion and Sediment Yield: Some Methods
of Measurement and Modelling. Geo. Books, Norwich, England.
Hem,
J.D., 1989, Study and interpretation of the chemical characteristics of
natural water: USGS Water-Supply Paper 2254, 264 p.
Hobbie,
J.E. (ed). 2000. Estuarine Science: A Synthetic Approach to Research and
Practice. Island Press. Washington, DC.
Howarth,
R.W., G. Billen, D. Swaney, A. Townsend, N. Jaworski, K. Lajtha, J.A. Downing,
R. Elmgren, N. Caraco, T. Jordan, F. Berendse, J. Freney, V. Kudeyarov,
P. Murdoch, and Z. Zhao-Liang. 1996. Regional nitrogen budgets and riverine
N & P fluxes for the drainages to the North Atlantic Ocean : Natural
and human influences. Biogeochemistry 35: 75-139.
Jacks,
G., A. Joelsson, and S. Fleischer. 1994. Nitrogen retention in forest wetlands.
Ambio 23: 358-362.
Jansson,
M., R. Andersson, H. Berggren, and L. Leonardson. 1994. Wetlands and lakes
as nitrogen traps. Ambio 23: 320-325.
Johanson,
R.C., Imhoff, J.C., and Davis, H.H., Jr., 1980, Users manual for hydrological
simulation program-FORTRAN (HSPF): U.S. Environmental Protection Agency,
Office of Research and Development, Environmental Research Laboratory Report
EPA-600/9-80-015, 678 p.
Justic, D., N.N. Rabelais, and R. E. Turner. 1995a. Stoichiometric nutrient balance and origin of coastal eutrophication. Marine Pollution Bulletin. 30: 41-46.
Justic,
D., N.N. Rabelais, R. E. Turner, and Q. Dortch. 1995b. Changes in nutrient
structure of river-dominated coastal waters: Stoichiometric nutrient balance
and its consequences.Estuarine
Coastal Shelf Science.40:339.
Mee,
L.D. 1992. The Black Sea crisis: A need for concerted international action.
Ambio 21: 278-286.
Meybeck,
M. 1993. C, N, P, and S in rivers: From sources to global inputs. In: R.
Wollast, F.T. Mackenzie, and L. Chou (eds.) Interactions of C, N, P, and
S Biogeochemical Cycles and Global Change. Berlin: Springer-Verlag. pp.
163-93.
Milliman,
J.D. and J.P. Syvitski. 1992. Geomorphic/tectonic control of sediment discharge
to the ocean: The importance of small mountainous rivers. Journal of Geology
100: 525-544.
National
Academy of Sciences.1994.Priorities
for Coastal Ecosystem Science. Committee to Identify High-Priority Science
to Meet National Coastal Needs.National
Academy Press, Washington, DC.88
pp.
Nixon,
S.W., J.W. Ammerman, L.P. Atkinson, V.M. Berounsky, G. Billen, W.C. Boicourt,
W.R. Boynton, T.M. Church, D.M. Ditoro, R. Elmgren, J.H. Garber, A.E. Giblin,
R.A. Jahnke, N.J.P. Owens, M.E.Q. Pilson, and S.P. Seitzinger. 1996. The
fate of nitrogen and phosphorus at the land-sea margin of the North Atlantic
Ocean. Biogeochemistry 35: 141-180.
Ortner,
P.B. and M.J. Dagg. 1995. Nutrient-enhanced coastal ocean productivity
explored in the Gulf of Mexico. EOS 76, 97, 109.
Rosenberg,
R., S. Elmgren, S. Fleischer, P. Jonsson, G. Persson, and H. Dahlin. 1990.
Marine eutrophication case studies in Sweden. Ambio 19:102-108.
Smil,
V. 1992. China’s environment in the 1980s: Some critical changes. Ambio
21:431-436.
Smith,
R.A., R.B. Alexander, G.D. Tasker, C.V. Price, K.W. Robinson, and D.A.
White. 1993. Statistical modeling of water quality in regional watersheds.
Watershed 93: A National Conference on Watershed Management.
Smith,
R.A., R.B. Alexander, and G.E. Schwarz. 1994. Nutrient transport modeling
of U.S. watersheds. Preliminary Proceedings of the IGBP Inter-Core Project
Workshop on Modeling the Delivery of Terrestrial Materials to Freshwater
and Coastal Ecosystems, Institute for the Study of Earth, Oceans, and Space.
University of New Hampshire, Durham, NH.
Turner,
R.E. and N. Rabalais. 1994. Evidence for coastal eutrophication near the
Mississippi River delta. Nature 368: 619-621.
Vörösmarty,
C.J., B. Moore III, A.L. Grace, M.P. Gildea, J.M. Melillo, B.J. Peterson,
E.B. Rastetter, and P.A. Steudler, 1989. Continental-scale models of water
balance and fluvial transport; an application to South America, Global
Biogeochemical Cycles, 3:241-265
Vörösmarty, C.J., C.J. Willmott, B.J. Choudhury, A.L. Schloss, T.K. Stearns, S.M. Robeson, and T.J. Dorman. 1996. Analyzing the discharge regime of a large tropical river through remote sensing, ground-based climatic data, and modeling. Water Resources Research 32:3137-50.
Vörösmarty, C.J., C.A. Federerand A. Schloss. 1998.Potential evaporation functions compared on U.S. watersheds: Implications for global-scale water balance and terrestrial ecosystem modeling.Journal of Hydrology 207:147-69.
Vörösmarty,
C.J., M. Meybeck, B. Fekete, and K. Sharma. 1997a. The potential impact
of neo-Castorization on sediment transport by the global network of rivers.
Human Impact on Erosion and Sedimentation: Proceedings of the Rabat Symposium.
IAHS Publication #245.
Vörösmarty,
C.J., K.P. Sharma, B.M. Fekete, A.H. Copeland, J. Holden, J. Marble, and
J.A. Lough. 1997b. The storage and aging of continental runoff in large
reservoir systems of the world. Ambio 26: 210-219.
Vörösmarty,
C.J. and B.J. Peterson. 2000. Macro-scale models of water and nutrient
flux to the coastal zone. In: J. Hobbie (ed). Estuarine Science: A Synthetic
Approach to Research and Practice. Island Press.
White,
D. A., R.A. Smith, C.V. Price, R.B. Alexander, and K.W. Robinson. 1992.
A spatial model to aggregate point-source and nonpoint-source water-quality
data for large areas. Computers & Geosciences 18(8): 1055-1073.
Young, R. A., C.A. Onstad, D.D. Bosch, and W.P. Anderson. 1987. AGNPS: Agricultural Nonpoint Source Pollution Model: A watershed analysis tool. U.S. Department of Agriculture - Agricultural Research Service