An evaluation of the Port Hacking Estuary Project from the viewpoint of applied science

Abstract

The Port Hacking Estuary Project, a model-guided study of the flow of carbon through a small Australian estuary, is reviewed from the viewpoint of applied science. The Project did not reach its goal of constructing a predictive dynamic model of carbon flow in the South West Arm of Port Hacking and key ambiguities in project design and execution that inhibited progress are identified. It is suggested that the model structure chosen to be compatible with time and manpower constraints did not allow sufficient mechanistic contribution to attract the support of the experimental participants. It is a general characteristic of ecosystem modelling to date that a complete (self-consistent) set of data is assumed before the model is constructed. Since no group, including our Project, has been able to collect such a complete set of data from the specific ecosystem, recourse to the literature is regularly made to complete the model. This means most of the data are of unknown reliability, and this fact implicitly demeans the (often hard won) data from the specific ecosystem; this, in turn, leads to decreasing interest in the model from the experimental participants and the group eventually splits apart. To avoid this scenario, I suggest more emphasis on the data that has been collected from the specific ecosystem: in particular, periodic syntheses of the available data set not supplemented by data of unknown reliability.