Those of us who are considering gypsum or alum treatment can easily do a simulation.

Take 3, 1 gallon samples, in clear or semiclear plastic jugs. Use one as a control, dont treat it, just leave it alone and watch it for a week or so. If the control sample settles, you have a physical disturbance issue. Physical disturbance is often the culprit if you have high hardness (100 mg/liter+) and also excessive turbidity. Treat one sample with gypsum. 1.4 grams per gallon sample is equivalent to 1000lbs per acre ft. You can break that down into smaller increments if you want to be precise in determining threshold dosage. Some samples dont clear, and they are easily ruled out as candidates. Also do the alum simulation, .067grams per gallon is equal to 50lbs per acre ft of pond water. Small gram scales are easy and cheap. No need to do the slurry method suggested in some of the publications.

The simulation can save you money and give you insight. Make sure to do your gypsum simulation with product from your would-be supplier. Same with alum. Sone samples are not good candidates. Different types of clay respond differently. Often the chocolate milk samples are easiest to clear, and the slightly turbid samples are stubborn or ruled out as candidates. Often folks confuse stained water with clay turbidity.

Hope this helps