Use of historic data, degree days and
agrometeorological variables, to build
prediction models of head lettuce
harvest date in northern Greece
This dissertation was written as part of the MSc in Sustainable Agriculture and
Business at the International Hellenic University. Lettuce, is an annual plant, of the
Asteraceae family (Table 1.1). It is mostly grown as a leafy vegetable. In
Mediterranean regions it is best grown during spring and autumn. It is difficult to
cultivate it during summer, so off season cultivation presents a commercial interest
for many growers. It is very difficult to predict plant growth based on calendar days
due to the fact that temperatures can vary greatly between years. Growing degree
days (GDD), are based on actual temperatures, so they are a simple and accurate
way to predict cultivation durations. The aim of this study was to build a model to
help growers predict the duration of cultivation per season, using GDDs.
This study was done at Vezyroglou Farm, Shinas, Alexandria, Imathia, Greece. Historic
weather data for the last five years were used to calculate the expected GDD per day
of the year. Also, the harvest dates for the 3 main lettuce types cultivated at
Vezyroglou Farm (Romana, Iceberg, Butterhead), were recorded.
The required GDDs for the beginning and the end of harvest, were calculated for these
3 lettuce types for the last 3 years. Because of the fact that every lettuce variety has a
different rate of development, the required GDDs for every period were different. The
relationship between the transplant date (from 1st to 365th day) and cultivation
duration (Days to harvest) from transplant to harvest for all three types of lettuce was
calculated. 2nd degree polynomic equations were fitted for the prediction of number
of days to harvest for a specific transplant date to calculate for the beginning and the
end of harvest. For the model creation, the required GDD accumulations were used
for every day of transplant in order to predict harvest days, as well as the window of
harvest. Using the real GDD (from the real climatic data every year) from the day of
transplanting until any day the model updates the predictions. This way, the remaining
GDD until harvest can be calculated and by these, the new predicted harvest days, as
well as the new predicted cultivation durations. The model was tested in the first 5
transplantings of 2021 and was quite accurate.
4
With this model, a grower is very easy to predict when the harvest will occur, and it
will get more accurate, the closer we get to the harvest. This can prove very useful for
lettuce growers planning. In the future, it would be very interesting to investigate
which climatic parameters affect the yield of lettuce. This way it is possible to upgrade
the present model in order to predict apart from the harvest dates, the expected yield
for every transplanting of lettuce.
Collections
Show Collections
