GDD Inaccuracies

How the growing degree calculation is made and one of the limitations (that growing degree days are linear and crop development is not linear) have been pointed out. There are other limitations. One that should be obvious from the rules previously stated is the limitation of what to do with the low or the base temperature. If the threshold temperature, called THR, is 50° F, the actual low temperature is 50° F and the high temperature is 70° F, there were 10 growing degree days during the 24-hour period. During the period, the temperature warmed up to 70° F, and later assume it cooled back down to 50° F. If lines connect the temperature values, the area under the "curve" represents crop growth/development potential (Fig. 3.14). However, temperature seldom varies between the daily low and high value in a manner linear with time. Hence, the "area under the curve" is not totally realistic.


gddassumption Fig. 3.14 GDD assumes temperature changes linearly throughout the day. Click on figure for actual temperature change.

Now, observe a third inaccuracy of the growing degree day, the condition wherein the low temperature falls below 50° F. At 50° or above, and with averaging over some days, errors would be small. Consider that the high was 70° but the low temperature fell to 40°. Remember the rule. If the low temperature is below 50°, call it 50°. So, move it up and call it 50°. The growing degree day accumulation for that day remains at 10. Even though the night temperature had been 40°, there would still be a 10 just like during the previous day. There is some error involved in this (Fig. 3.15). The plant did not start growing until the temperature reached 50. So there were some hours between the 40° and 50° when it would be assumed that the plant was growing, but it really was not. This error gives the plant more growing degree days than the plant actually recognized or received. So there is an area of error and, of course, the area of error may be compounded by the evening cooling as well.

gddassumption2
Fig. 3.15 Error in GDD calculation. The GDDs for this day would be 10. But the growth area would actually be less because the crop would not grow until temperatures moved above 50 °F.

With this error, introduced by assuming that the temperature was 50° even though it was only 40°, plants do not grow or develop as fast as anticipated.

There are three inherent errors: First, the inherent error that growing degree days assume linear growth (Fig. 3.11); second, the "shape" of the temperature trend with time (Fig. 3.14); third, the error of adjusting the minimum temperature up to the base when we calculate, so there is a little area of mistake if things under the area of the curve are considered (Fig. 3.15).

Now, someone might say, "Okay, fix it. An account needs to be made for this area that has an error." This has been done. A lot of work was done at Iowa State University, primarily by entomologists, to do just that. During a 24-hour period, there will be a minimum temperature that existed at some time,usually just before sunrise. Soon after the sun comes up, the temperature starts to increase, usually quite rapidly. Then the increase rate diminishes, and a peak is reached for the day. Temperature will fall to whatever the low for the next night might be. It may be different from the previous night. Nevertheless, it will taper off. This peak will probably come around 1 p.m. standard time or at least some time in the afternoon. The low point will probably come around 6 a.m., actually just before sunrise or just the minute or two after sunrise.

A temperature curve of what might happen during a day is shown in Figure 3.16 (warming up quite quickly when the sun comes up, usually if it's a clear day, hitting a peak around 1 p.m. standard time, then tapering off slowly until 6 a.m. the next morning to the minimum for the 24-hour period). Someone might say, "We would do better to draw a curve something like a sine curve rather than a growing degree day straight line that indicates changes from 6 a.m. to 1 p.m. and then another straight line for the period 1 p.m. to 6 a.m. the next day." A method was developed for calculation of growing degree days called the "sine wave method" that assumes that the day heats up slowly and reaches a peak around 1 o'clock. It then cools off in a uniform way according to a smooth sine curve.

a
Fig. 3.16 Sine wave approximation to daily temperature.

A growing degree day calculation according to a sine curve makes it possible to correct one of the errors described earlier. If the minimum temperature were somewhat below threshold (Fig. 3.16b), the temperature of the day rose, and attained a peak, the sine wave would define the hour it will cross the threshold temperature. The plant will grow according to the area beneath that curve and be considerably more accurate than the straight line method.

GDD Advantages

This more accurate way is not generally used. An interesting study was done a few years ago. An entomologist in a neighboring state expressed some criticism. He said, "Iowa developed the sine method of calculating growing degree days, while Iowa is not using them. Iowa is using growing degree days by the old Weather Service method which has been used since 1900." The reply was, "There are two reasons we do that! The first reason is, Iowa has used the Weather Service method since 1900. Everyone has that record; hence, ongoing measurements and data are consistent with the historical record. The second reason is, data indicate that the growing degree day method (after the inaccuracies of the growing degree days were adjusted to base 50) gives better results with the Black cutworms than the method that uses the sine wave calculations."

The results are better using the original method, for black cutworms under field conditions at least, than for using the sine wave method. The sine wave method is better for use with insects reared inside the laboratory. For those that are growing outside, the straight line, old-fashioned method turns out to be the best. Why it turned out to be the best is not really known. Some ideas have been suggested. The worms move around to places having different temperatures, Maybe their activity is sort of compensated for in the error in the method. Certainly the worms do move around. If the temperatures get too hot, they dig a hole in the ground and go down to where it is cool enough for them to be comfortable. When the temperatures are appropriate on the surface, they come up to the surface and feed and do the things that cutworms do, as observed by the entomologists and this author. The results of cutworm activities are the reasons why our growing degree day calculations seem to be acceptable for Black cutworms.

There is one more reason to use the old-fashioned method for calculation, which is really the overriding reason, the accuracy of the system. When making a calculation, one does not need to have any calculation better than the accuracy of the system.

What is the biggest error in the system? Is it the calculation of the growing degree day, or is it the measurement of the temperature of the air itself? A thermometer is not perfect. The thermometer reads to the nearest degree or records something on the order of ±1° F. There is a 1° error just in the thermometer. What about the error across a field? From one side of the field to the other? On a south-facing slope, and a north-facing slope, the plants grow differently. Even if the soils are the same, there is different response to temperatures, different emergence times, and different development rates. The temperature variation across the field itself may be on the order of 5° F. If there is a variability of 5° across the field itself, the error that comes from not calculating growing degree days, the very best way pales next to the error of the temperature variability of the environment.

When growing degree days are observed, the hope is that this process is something a little better than the crop calendar that was considered previously. The crop calendar gives the average dates of planting, the average date of emergence, the dates of tasseling, and the dates of maturity as calendar dates. The growing degree day adjusts for temperature and indicates whether or not the temperature effect has been significant enough to move away from the average dates, and about how far. That can be of some benefit, and that is all it is--just SOME benefit. There is no reason to think that a growing degree day will be a perfect model for what a crop will do.