Faculty Seed Grant Projects

Integrating an economic input-output model with watershed and energy models using a system-of-systems approach

  • Dr. John Little, Civil and Environmental Engineering
  • Dr. Cayelan Carey, Biological Sciences
  • Dr. Faye Duchin, RPI School of Humanities & Social Science/Economics
  • Dr. Adil Godrej, Civil and Environmental Engineering
  • Dr. Erich Hester, Civil and Environmental Engineering
  • Dr. Lamine Mili, Electrical Engineering
  • Dr. Adrian Sandu, Computer Science

There is growing recognition of the need for a system-of-systems approach when attempting to understand complex societal challenges such as sustainability. Consider, for example, the food, energy, and water systems, which are intimately interconnected, meaning that we need to understand their interdependencies before we can formulate effective solutions. To address this challenge, we will begin the development of an interdisciplinary, system-of-systems framework using the Chesapeake Bay Watershed as our study region. We will focus on agriculture and food, water at the watershed scale, and energy generation and distribution.

We choose for our economic model a family of input-output models. The input-output database covers all producing sectors of the economy, allowing for detailed disaggregation of the sectors of particular interest including agriculture. The EPA’s Chesapeake Bay Program maintains a publicly-available release of a Chesapeake Bay Watershed model which can be used to simulate management scenarios. Energy generation and distribution models for the Chesapeake Bay Watershed are managed by energy utility companies operating within the study region. Our intention is to integrate an economic input-output model more intimately than has previously been achieved with both watershed and power grid models. Such a temporal coupling, where economic, watershed, and energy models run forward in time, seamlessly exchanging information at each time step, has never been accomplished before and is entirely novel. Our specific objectives are therefore to:

  1. Develop a preliminary, spatially-resolved, economic input-output model for the Chesapeake Bay Watershed that is dynamic in distinguishing the timing of a sector’s water withdrawals and discharges and energy requirements;
  2. Acquire and develop an input-output database for the Chesapeake Bay Watershed that includes new variables and parameters that are not standard components for economic analysis and that is organized within a Geographic Information System;
  3. Integrate the economic input-output model with the existing, spatially-resolved, dynamic, watershed and energy models; and
  4. Evaluate the potential to generalize the system-of-systems approach to other regions and other systems, including demographic and social systems.

The knowledge gained during this project will enable us to acquire preliminary data and develop conceptual analyses that will form the basis for strong proposals for follow-on funding on the integration of economic input-output models with watershed and energy models.

Economic model: The economic input-output model takes the form of a linear program, with the primal results in physical units of output and resource use, the dual results in corresponding costs and prices, and well-defined complementarity relations between changes in quantities and costs under alternative scenarios about the future. The minimal requirements of inputs from the other models are the spatially-distinguished stocks of relevant resources, namely water, energy, and land suitable for agriculture and other economic uses. The economic model provides the other models with the volumes and spatial distribution of demand for land, water, energy, and other resources and goods.

Watershed model: The Chesapeake Bay Watershed Model is comprised of a suite of linked models, principal amongst these being watershed, airshed, land use change, and estuary models. The watershed model itself uses over 2,000 segments with an average segment (land area) size of 83 km2 and typically executes on a one-hour time-step. Each of these spatial and temporal resolutions can be lumped or averaged over a larger space or time period if needed.

Energy model: The energy model consists of several electric power plants connected to a threephase high-voltage power transmission network that provides electric energy to medium and lowvoltage power distribution networks, which in turn serve a host of geographically dispersed loads. These are essentially standard power grid models that are maintained and operated by energy utilities within the study region.

Model integration: The economic, watershed, and energy models will be integrated in a spatially-resolved and dynamic fashion. For example, if the watershed model projects decreasing water availability in an agricultural region, this will automatically be reflected in a reduction in water availability for economic activities. This will have other economic consequences, namely some combination of increased price for water, technological changes that are water-saving, substitution of water-intensive production by imports, or regulation to influence the costs or allocation of water. These changes will affect all economic activities directly or indirectly and in particular the amount of water withdrawn and discharged, constituting an input to the watershed model. Given that the watershed and energy models are highly resolved, a critical aspect of the research is the spatial and temporal resolution at which the economic model can be implemented.

The project will take place during the 6-month period from 1 December 2016 to 1 June 2017.