This is a subject on which I have done a lot of research, but I have little hands on experience so far.

Generally, it is pretty easy to measure ph and electrical conductivity with low cost electronics. DO is more expensive. The electrical conductivity will tell you in aggregate how many ions you have in solution but it won't tell you how much of any particular ion... ie NH4+, NO3-, NO2-, or for that matter, K+, or P3-.

So far, the best technology that I have been able to find that does that directly are semi-permeable membrane sensors. Basically they are a membrane that sorts out just the ion you are looking for, and then does a electrical conductivity test on the solution you get. Last time I checked, they were expensive (>$500per), short lived (<3 months in continuous use), finicky (require constant recalibration), but available from Honeywell. For some reason I'm not able to track them down right now on Google though?

I concluded that there was basically nothing out there that does the job very well, and even less that is remotely affordable.

But, I have an idea that might help solve that?

The idea is basically to run an automated system that draws off a fixed volume of water fro you system, and then adds the correct number of drop of reagent to the sample. Then it shines light through it, and reads the color with a simple spectrometer like one of these...

http://publiclab.org/wiki/spectrometer

Depending on how many reagents you had you could automatedly test for basically anything you wanted. and it would be pretty cheap and durable.

But it is not a product you can purchase and would require some work to get going.