Dear Dibakar,

Thank you for the paper. I'm still wondering how accurate these models really are? In my mind if they are empirical they are only as accurate as the input values?

Kind regards,

Clint

Field hydrologists, soil scientists, water quality specialists and similar technical specialists would spend substantially more time inventorying conditions and applying appropriate BMPs to protect water quality, identifying any special mitigation needs for activity than running models.  Models do have their uses and are helpful in trying to understand water quality dynamics at various hydrological scale, physiographic conditions and land use scenarios.  But as time passes with more experience and training, professionals learn to read the land and have a working knowledge of acceptable and questionable practices aand actions.  Often the question is not what model was used, but were BMPs identified and applied to fit the activity and circumstance.

Models may have many limitations unless they have been validated to the conditions under analysis that may include collecting more data in the verification process.  Too many want to generate numbers or answers with a model without reviewing the model assumptions, limits and need for validating or adjusting when applied to conditions outside of the boundaries that developed the model. 

There are many physical, chemical and biological aspects to water quality - what model can address them all?  My guess is that research is going to continue to spend significant time and resources in modeling.  The models that are developed when transferred to professionals and technical specialists, improve understanding perhaps and may help provide justification.  But many times, the practical application of activities for water quality maintenance and improvement is not as dependent on models, as it depends on who is planning and implementing work, and making decisions.  

"if they are empirical they are only as accurate as the input values".

The advantage of such empirical models is that they are very close to the conditions, in which they were created. So it would be always useful for you to invest in some your own model unless it is too expensive. If it is too expencive, then you are right and it is better to use other methods of research and analisys.

Dear Dr Clinton, thanks for your teaching question which seems to be around a basic concept: the chosen boundary conditions. That is, if the boundary conditions are really those of the so called "catchment"?. Aquatic systems are often considered to be limited by the bottom of the lake or river, and by the nearby hills (surface watershed); limits that surely require to be clearly supported by solid data. There is a need to validate these conditions, showing the lack of input from the presence of groundwater flows that account for a considerable larger percentage of the existing water. In fact, the main starting model of any study area should be to establishing a satisfactory low-cost conceptual model. This model might be used in producing the correct questions to the observed field answers... 

Do get from Mexico saludos,jjoel carrillor  

Well, All answered  before are appreciated, and from 31 years job in groundwater quality and Hydro-geochemistry interest, models in regional studies give an idea to right decision taken by  water politic management. Local model interprets hydrological and hydrochemical phenomena with best  recommendations proceeds to final solution for related problems which finally solved by in situ model. so that each model aimed to the needed targets(No problems with cheep or expensive investment),

Dear All,

Thank you for your answers. Very valuable to get a wide perspective. 

Regards,

Clint