A golf course superintendent who wishes to remain anonymous wrote with an interesting question. Here's our exchange.

I have been using the growth potential model for 2 seasons - but very religiously this season. I have some brief thoughts that I wanted to pass along, and was curious what you thought regarding the following:

Would the growth potential model more accurately match the growth of the grass if soil temperatures were used rather than air temperatures?  Or canopy temperatures?

I am sending this email today because of some observations this weekend. We had our first cold snap of the season - average temperatures of 51, 49, 47, 53 [that would be from about 8 to 12 ºC] this past Thursday - Sunday.  However, we had full sun in the afternoons and our soil temperatures at 2" [5 cm] depths and canopy temperatures (in the full sun) both exceeded the air temperatures. It does not appear that our grass slowed its growth much. This is small sampling in a short period of time, and my guess is over a long time frame (the interval between fertilizer applications) the peaks and valleys will smooth out.

Going into and coming out of winter are when I notice the greatest variances.  Honestly, I don't know that the variations will have much impact on N amounts. I have not monitored soil temps religiously enough to cross reference with the growth potential model, but plan to next year.

Ultimately, just curious what you thought about the impact of soil or canopy temps on the growth potential model.

Here's my response.

Real quick answer: the GP optimum temperature and equations are set based on the general AIR temperatures for growth written about in Dr. Beard's Turfgrass: Science & Culture book. So as far as general use, the GP equation is meant to generate a value between 0 and 1 that is calculated from air temperature and related to the optimum air temperature given by Dr. Beard.

The application of GP is not an exact science and is not reality -- it is just a number between 0 and 1 (or between 0 and 100%) that can be put to use in various ways. As you will have seen, and as I and others around the world have found, despite the fact that GP is just a number, and is not in and of itself reality, it can be practically useful in a lot of maintenance work. 

Here's my guess at what happens with weather like this past weekend. I think the variances you see in spring and fall may also be due to the optimum temperatures changing -- grasses adapt a bit when it is hot or cold, so there is just no way that one can model growth exactly with this type of equation. The enzymes in the leaf change, or the activation temperatures for certain reactions, with changes in temperature. So I think as long as things generally seem to even out over say a week -- a week's average of GP should correspond to the way one sees the grass actually growing -- then I would be comfortable that the number is useful. 

Canopy temperatures, I don't know enough to consider.

Soil temperatures -- one could do this type of model using soil temperature, and one would adjust the optimum temperatures to do so, probably. But a lot of research about soil temperatures is related to root growth, not so much shoot growth. So I still like to work with air temperatures and try to estimate shoot growth. My guess is that soil temperature would not add a whole lot to the accuracy of the model. The reason being that soil temperature is a wave function following air temperature. The soil temperature just follows the air temperature and doesn't go to extremes of highs and lows and has a time lag depending on depth. My measurements of soil temperature and air temperature have been in Asia and in California and Oregon. When I've done that, the soil temperature is usually quite similar to the air temperature (see Figure 10 here), and it has been somewhat predictably lagging behind air temperature. So I just don't think there is a huge amount of information in soil temperature that would improve accuracy. 

I think it is awesome you are applying this, and I think it is an excellent idea to try different measurements or adjustments that better make it fit the observed growth or grass conditions. 

One more thing comes to mind regarding GP. I look at it as a passive predictor, not as a driving or controlling force. So the GP for me is very much a "potential" but it is not something that is "causing" the grass to do anything. In that sense, if the grass was healthy last week, and had some ample amounts of nutrients, and the soil temperatures and air temperatures were reasonable and then dropped off, you can see that the driving force would be the inertia of the grass just growing along. The GP as a "potential" would not immediately slam the brakes on growth. But over the long term, the potential makes sense.