There are many models to assess estuarine and coastal ecosystems in general. In fact according to the conditions of each ecosystem can be created for each particular model. But there is one that I think is very good, is simple, practical and easy to understand and apply, is the model Land-Ocean Interactions in the Coastal Zone (LOICZ). It is a simple model that describes completely if your system like you can be the basis for your own model, in this page you will find more information http://www.loicz.org/, I hope you serve.

Regards

Your question potentially opens a very broad discussion because "the ecological effects of chemicals in marine and estuarine systems" covers a lot of fields going from the effect of heavy metals on marine organisms to the effect of nutrients loads on aquatic production or trophic ecosystem structure. So it is a bit difficult to point you towards one particular model or the other.

Yet, I guess there are a few things that can already be said, in particular with respect to the data requirements you are refering to.

As a very first approach, you can do some mass balance budget on your system if you can contrain its physical domain and assume it homogeneity (which is a strong assumption, especially in a estuary). That's basically what the LOICZ 'model' is. Based on very simple data (salinity, volume, DIN, DIP) and assuming redfield ratio, you get an approximation of the NEM of a system. This is still very crude and I wouldn't qualify that as a model in the sense that is not dynamic and you can't do any simulation.

A step further, based on the same approach but with the implementation of a simple reaction network, you can derive a box model and make simple simulations over time with varying environmental parameters. This will give you qualitative insight regarding the evolution in time of your system but there won't be any representation of the spatial heterogeneity (except if you design several boxes). I would suggest to have a look at the work of Fred T. Mackenzie (http://www.soest.hawaii.edu/oceanography/faculty/mackenzie.html) who made several box models for the coastal zone at the global scale but there exist local and regional box models too.

If you want a representation of the hydrodynamics of your system, which is often crutial to understand the dynamics of an estuary for example, you'll need to use a Reactive Transport Model. It will solve for the hydrodynamics of your system and provide a representation of the transport (by diffusion and advection) of the chemical species in your system. The reaction part will be your biogeochemical module and can be anything depending on what you want to study.

Depending on what you are interested in as far as chemistry and ecology go, you have a whole suite of biogeochemical model of various complexities. Starting with a N, P, Z, D model (for Nutrient, Phytoplankton, Zooplankton and Detritic matter) and going to very complex mechanistic models providing insight on the physiology of several phytoplanktionc groups.

Of course, the more complex the model, the more data you'll need to contrain all the parameters (and later validate the results). Besides, the representation of the hydrodynamics will require a lot of additional physical data and, the more dimensions you want to represent (i.e. you can decide to ignore depth variation and use a 2D model), the higher the data requirement. On the plus side, this spatial representation also allows you to use parametrization varying in space (like along the salinity gradient of an estuary as you suggest in your question).

I realize it is a lengthy response but I hope (some of) it helps. As far as naming models goes, there are so many that I wouldn’t be sure where to start so I would suggest you to have a look at this paper which compares several 3D ecological models and their performances in the same region: Moll A, Radach G (2003) Review of three-dimensional ecological modelling related to the North Sea shelf system - Part 1: models and their results. Progress in Oceanography 57: 175-217