An example of ample DLI

Figure 1. The snow-covered landscape of central Hokkaido in mid-March 2024 when the growth potential was 0 but the DLI was fine.
Figure 1. The snow-covered landscape of central Hokkaido in mid-March 2024 when the growth potential was 0 but the DLI was fine.

In a recent ATC Doublecut discussing light and temperature, I mentioned that I’d calculate potential DLI for the recording location, check the actual DLI for the recording location, and share that information.

Figure 2. The actual and potential DLI for each hour of March 16, 2024, at Asahikawa, Japan.
Figure 2. The actual and potential DLI for each hour of March 16, 2024, at Asahikawa, Japan.

I’ve looked that up, made a chart, and wanted to explain this again real quick with one more example.

My assessment, after checking data for a lot of places, is that temperature generally plays a controlling role in how much the grass can grow. Light from the sun1 rarely provides a continuous limit on growth, especially on cool-season grass growth, in the same way.

Figure 1 shows a snow-covered landscape. Grass is not growing. But there’s plenty of photosynthetically active radiation. I looked up the DLI data for a nearby JMA station with a sensor, and those data are shown in Figure 2. On March 16, the day I recorded that episode, the DLI was 32.4 mol/m2. If it wasn’t for the partly cloudy skies, the DLI would have been close to 39.6 mol/m2. Note that even this early in the season, that’s more light than cool-season grass can use.2

Meanwhile, in all this glorious sunshine, the high temperature on March 16 at this location was 3.8 °C, and the low temperature was -7 °C, with an average temperature of -1.3 °C. That gives a C3 growth potential (GP) of 0.

Does the GP prediction improve by adding an adjustment for light on this date, and in this location? I would argue that an adjustment doesn’t improve anything in this case.


  1. This is not a discussion of tree or structural shade. This is about sun, clouds, location on the planet, and day of the year, and the amount of light that reaches the earth’s surface. ↩︎

  2. The midday PPFD was often above 1,000 µmol/m2/s, which is the conventional light saturation point for C3 grasses. ↩︎

Related Posts

Next
Previous