Conceptual Models > Program coupling

PEST + PHREEQC issue in optimization process

(1/1)

Ardenis:
My PhD research is focused on fluoride adsorption in volcanic soils. I use PHREEQC to model this reaction, and the predictions are then compared with experimental measurements. I've successfully integrated PEST with my PHREEQC script, using the method described in Bachmaf and Merkel's 2011 paper (Complexation modeling 123). However, when I start the optimization with PEST, PHREEQC opens, and I have to run the script and close it at each optimization step. After a while, PEST crashes and I don?t know when I should end the optimization. Thie whole process seems terribly inefficient. Would someone be able to advice on how to handle the optimization process with PEST correctly? Thank you very much for your time.

dlparkhurst:
I don't understand about opening and closing; PEST should run PHREEQC at each iteration and then pull out results through the instruction file.

With surface complexation, the number of sites and log K are highly correlated, so it may be difficult to arrive at a unique optimization. My only advice is to start simply with as few sorption species as possible, and also, at least initially, to fix the number of sites.

Another possibility is to use PhreePlot. It has an optimization capability and is focused specifically on PHREEQC.

Ardenis:
Thank you very much for your answer ! I must have made a mistake when installing PEST or in one of my PEST files. I will check that ! And I will also take a look to PhreePlot to check if my issue can be solved with this tool.

Navigation

[0] Message Index