PARENT SESSION
Oral Session # 1: Ecological Theory I: Theory; Scaling.
Presiding: M Pascual
Monday, August 4. 8:00 AM to 11:30 AM, SITCC Meeting Room 100.

Discrete-time dynamic environ analysis of indirect effects in ecological networks: Basic considerations.

Patten, Bernard^*,1, Borrett, Stuart ¹, Whipple, Stuart², Christian, Robert³, Thomas, Cassondra⁴, ¹ Institute of Ecology, Athens, GA² Skidaway Institute of Oceanography, Savannah, GA³ Biology Department, Greenville, NC⁴ Department of Environmental Sciences, Charlottesville, VA

ABSTRACT- Environ analysis is a set of input-output methods to analyze the environments within systems (termed environs) of the system's component parts. Analyses of internal pathways, throughflows (summed node flows), and node storages provide core methodologies, from which additional analyses for utility and control are derived. Existing methods are restricted to time-invariant (stationary) systems at steady state. Needed are methods for more realistic cases, like food webs and biogeochemical cycles, which are neither stationary nor at steady state due to time variation of internal structure and function, or inputs. Methods for time-continuous dynamic analyses are cumbersome, and discrete-time approaches have yet to be attempted. This paper explores a discrete approach for the forward-time (output-environ) version of throughflow analysis; methods for throughflow input environs, and storage input and output environs, follow parallel development. Three cases are considered: (1) intensive analysis of nonstationary systems with constant input, and extensive analyses of (2) stationary and (3) nonstationary systems with time-varying input. For Case 1 we address how to couple discrete steps in the time series, and investigate how dominant indirect effects typically observed for stationary systems are altered by discrete-time successions of flow-generating matrices. For Cases 2 and 3, we also consider how to couple discrete steps in the time series, we determine what fractions of extended pathway networks are engaged during each time period, and we explore the effects of time-series truncation on indirect-effects dominance. We also consider alternative scenarios with respect to the correspondence between time and pathway length, and we show that as time increases with increasing length, the relative role of indirect effects declines. In all cases, however, indirect effects still dominate. We illustrate our results with a discrete-time model (16 seasons, over 4 years) of nitrogen loading in the Neuse River Estuary, North Carolina.

Key words: biogeochemical cycling, environ analysis, indirect effects, Neuse River Estuary