Endothermic “warm blooded” organisms like birds and mammals maintain relatively constant body temperatures, so their metabolic reactions and development rates are fairly consistent over time. However, body temperatures of ectothermic “cold blooded” organisms (i.e. plants, insects, fungi and bacteria) are close to ambient temperature, so their rates of metabolism and development rates are strongly influenced by the temperature of their environment. Several other factors like moisture, competition (i.e. crop spacing and weed density) and pest damage can influence crop development rates, but time and temperature (degree-days) can often predict maturity more accurately than just time (calendar days). We’ve all seen crops grow quickly when temperatures are optimal, and slowly or not at all when it’s too cold or too hot.
Oregon State University Extension and the OSU Integrated Plant Protection Center are working with seed companies and local farmers to develop a degree-day scheduling website for vegetable growers. Croptime will predict harvest dates for vegetable varieties chosen by collaborating growers and seed companies (see Table 1). Ed Peachey and Aaron Heinrich (OSU Extension) are also developing some weed models that will predict when viable seeds are set. Growers will be able to use this information to reduce weed seed rain in vegetable rotations. Dan Sullivan (OSU Extension) is explaining how thermal time can improve our understanding of the nitrogen cycle. The OSU team hopes to put at least 50 variety specific models and three weed models online by late 2016, but some crops will need more research before the models are ready.
Around 1730 René A. F. de Réamur first used mean daily air temperatures to predict plant development. Since then biologists have been improving crop models. Figure 1 illustrates how sine curves can estimate degree-day accumulation between a lower threshold (the temperature below which the organism does not develop) and upper threshold (the temperature above which the organism does not develop). The volume of the shaded area represents the degree-days accumulated on those two days. For example, Jubilee sweet corn has lower and upper thresholds of 50°F and 86°F respectively, and requires 1539 degree-days to reach fresh market harvest. On a cool spring day with a low of 44°F and a high of 62°F, Jubilee will accumulate only 6 degree-days, on a warm summer day with a low of 60°F and a high of 85°F, about 22 degree-days are gained. Visit this UC Davis site for a more in-depth discussion of degree-day concepts.
Orchardists regularly use degree-days to predict insect pest phenology (i.e. codling moth and filbert worm) and disease risk (i.e. apple scab and fire blight). Some degree-day models have been developed for vegetable crops and pests, but fresh market vegetable growers normally rely on calendar days to maturity provided in most seed catalogs. Frank Morton, owner of Wild Garden Seed in Philomath, Oregon breeds vegetables and sells organic seed. Frank explains, “The normal ‘days to maturity’ varietal information available in most seed catalogs is not useful to farmers, except in a vague relative sense. If seed breeders and catalogs could provide a degree-day index for their vegetable varieties, farmers would be able to more accurately model their crop delivery schedules in years of unusual weather patterns or extremes.”
David Brown from Mustard Seed Farms in St. Paul, Oregon is perhaps the only fresh market vegetable grower in Oregon who already uses degree-days to schedule crops. He has developed his own degree-day models for broccoli and some other crops. “I have used degree days for over 20 years to schedule successive plantings of vegetables... more information based on some research would be helpful in refining my schedules and maybe even using the information for more crops.” Our goal is to make reasonably accurate vegetable degree-day models accessible to more vegetable growers.
In the spring of 2016, growers will be able to use the first Croptime models to schedule plantings and predict harvest dates to plan a consistent supply. We have developed a new Google maps interface to make it easier to select the best nearby weather station (figure 2). Up to four planting dates can be entered at a time.
To use the Croptime calculator: follow these steps in figure 2:
- Choose a reliable local weather station in the map
- Select the "CROPTIME models" category
- Select the model fro the crop and variety you are growing
- Enter up to four successive planting dates
- Choose full or condensed output. Full output shows DD accumulation every day from first planting to last harvest, condensed output only shows dates when a phenological event occurs (i.e. flowering or maturity)
- Hit "Click here" to generate model output
- Scroll down the output sheet to see predicted harvest dates based on your planting dates.
Growers can use models to plan successive crops or predict harvest dates for full season crops in the spring. During the season producers can run models again to access more up-to-date and accurate harvest predictions. Bob Egger from the Pumpkin Patch on Sauvie Island, OR, explained how a steady flow of crops like cabbage could benefit his farm. “When we have a couple weeks of wet weather in spring we could use Croptime to choose varieties we might not be familiar with but would help keep our production up. The big buyers don’t waste time with you if you don’t have the right product available at the right time.”
The Croptime site uses actual weather data up to the day before a model is run, then 5-day forecasts followed by 30-year average temperatures. Tanya Murray previously with Sauvie Island Organics near Portland, Oregon planned each week’s CSA share carefully. “The dramatically different weather we have had this spring and last makes it hard to know what to expect. Croptime will help our farm use degree-days to predict maturity.” Len Coop (OSU Integrated Plant Protection Center) is improving the accuracy of long term forecasts by converting the output of NOAA weather models to degree-days.
Arcadia broccoli reportedly takes 63-94 days to mature depending on the seed catalog referenced. The preliminary Croptime model predicts 66-103 days between transplant and maturity from 2011-2015 at the North Willamette Research & Extension Center in Aurora, OR. Days to maturity vary with planting time and year (figure 3). Early spring planted Arcadia broccoli takes 20-30 days longer to mature than mid-summer plantings. Development also progresses more slowly in cooler years (2011-2012) than warmer years (2013-2015). Crops planted one month apart matured 14-26 days apart (not shown). Croptime models are being developed in irrigated Willamette Valley fields. In some regions and cropping systems, environmental factors not well tested here may be more important (i.e. moisture, day-length or upper thresholds).
Developing new models
We hope to continue developing degree-days models over time, and would like to include new models for winter vegetables and possibly cover crops. Since vegetable varieties change regularly we hope to eventually work with others to collect field data and develop models. We are also developing a Vegetable Growth Stage Guide and standard protocols to improve consistency of field observations (figures 4-6).
Croptime includes cool season crops like cabbage and spinach, and warm season crops like peppers and winter squash. Cool season crops have cooler lower and upper thresholds (i.e. 32°F and 70°F for broccoli). Warm season crops have warmer thresholds (i.e. 52°F lower threshold for sweet pepper); it normally doesn’t get hot enough in the Willamette Valley to identify upper thresholds for warm season crops. One data set consists of crop development observations at one location and planting date. Models require at least eight to ten data sets for each crop to verify threshold temperatures in the literature. Then four to five data sets are often enough to estimate the number of degree-days to maturity for each subsequent variety of the same crop, as long as threshold temperatures are the same for different varieties of the crop. So far preliminary thresholds have been identified for broccoli, sweet pepper, cucumber, winter squash and sweet corn. 2015 data is now being incorporated into these models.
Crop development observations were made under a variety of production methods such as organic, conventional, black plastic, bare ground, direct seeded and transplanted crops. Separate models may be needed for some of these practices. One day we may be able to adjust models to account for some factors such as the warming effect of black plastic mulch.
Vegetable degree-day models can be a more accurate crop scheduling method than calendar days. Producers and buyers using Croptime may be able improve the consistency of supply, and plan harvest crews and marketing activities more accurately. The weed models may help reduce weed seed rain in crop rotations, and the nitrogen information will add to our understanding of nitrogen cycling in organically managed soils. We hope the website will help growers and produce distributors improve efficiency, profitability and sustainability.
You can learn how to use the new system at the first Croptime workshop at the North Willamette Research & Extension Center on Feb 11th from 10:00-2:30 (contact Heidi Noordijk to register). We are also offering a double session at the OSU Small Farms Conference on Feb 20th in Corvallis.
Acknowledgments Croptime is funded by WSARE Research & Education award number SW12-037, additional funding from Clackamas Extension Innovation Fund. Photos by Heidi Noordijk.