Pontius Jr, R.G., L. Claessens, C. Hopkinson Jr, A. Marzouk, E.B. Rastetter, L.C. Schneider, J. Vallino. 2000. Scenarios of land-use change and nitrogen release in the Ipswich watershed, Massachusetts, USA. in Parks, B.O., K.M. Clarke, M.P. Crane, editors. 2000. Proceedings of the 4th international conference on integrating geographic information systems and environmental modeling: problems, prospects, and needs for research; 2000 Sep 2-8; Boulder, CO. Boulder: University of Colorado, Cooperative Institute for Research in Environmental Science. (www and CD).

Scenarios of land-use change and nitrogen release in the Ipswich watershed, Massachusetts, USA

Robert G Pontius Jr, Luc Claessens, Charles Hopkinson, Abdelkrim Marzouk, and Laura C Schneider, Edward B Rastetter. 


We construct scenarios of land-use change from 1991 to 2101 and analyze their implications for nutrient release in the Ipswich River Watershed, Massachusetts, USA. In the process, we create a GIS-based method that can be applied elsewhere to simulate human impacts across a landscape. The research makes important links in integrative modeling because the Ipswich River flows into Plum Island Sound, a salt marsh estuary. The estuary and its watersheds are a Long Term Ecological Research site funded by the National Science Foundation.

The land-use change component of the analysis concerns four land types: forest, residential & commercial, agriculture & open, and wetlands. We have land use maps from 1971, 1985 and 1991. We model land-use change based on spatial physical factors, legal constraints and extrapolations of quantities of change. We calibrate maps of suitability for deforestation with maps of real change between 1971 and 1985 by using multi-criteria analysis. The maps of 1971 and 1985 serve also as the basis to extrapolate the quantity of deforestation predicted for the future. The extrapolated quantities and calibrated suitability maps predict the location of deforestation between 1985 and 1991. The predicted deforestation maps are validated with the map of real forest area of 1991. Relative Operating Characteristic (ROC) and variations of kappa index of agreement measure the validation at various resolutions. In the most successful simulation run at the finest resolution Kno=84%, Klocation=15%, Kquantity=99% and Kstandard=14%, ROC=72%. To predict land-use change into the future, we use our validated simulation method to sketch a pair of scenarios. The scenarios compare a future with restrictive laws concerning land use versus a future without such laws.

The nitrogen component of the analysis is based on our study of the relation between land cover and nutrient concentrations in streams that drain small catchments within the Ipswich River watershed. This analysis is based on an empirical relationship between upstream land-use distribution and nitrate levels in small streams, which we obtained from intensive water nutrient sampling campaigns. For every decade in our two land-use change scenarios we apply this empirical nitrate loading relationship to several hundreds of sub-catchments. The results are time series of spatially distributed surfaces of nitrate loading, with corresponding transects of potential nitrate concentrations for the Ipswich River stream network. Simulation results for both development scenarios indicate a doubling in nitrate delivery to the Plum Island Sound estuary from 1991-2101. These results are preliminary, especially since they don't include any in-stream processing component. However, they demonstrate that future residential development could lead to a substantial increase in nutrient delivery, which could have serious implications for eutrophication of downstream riverine, wetland and estuarine ecosystems.


Deforestation, kappa, land use, Massachusetts, nitrogen, relative operating characteristic, validation.


4th International Conference on Integrating GIS and Environmental Modeling


See Figures Page for this article.


Copyright 2000 Graduate School of Geography, Clark University