Final Report for FNE11-718
Project Information
This project measured the time to harvest for many plantings of 2 varieties of lettuce, one variety of broccoli and one variety of corn, and provides a planning tool for scheduling succession planting of crops. The main variables are seen to be temperature above the plants “base growing temperature”, and the amount of light as the calendar year progresses. The temperature is the largest variable, year to year. The amount of light varies through the season but is repeatable year to year.
As more growers practice winter growing, these issues will become even more pronounced. Winter growing was not be addressed by this project, but the results should be useful. For instance, knowing the effects of time and temperature numerically will allow growers to make more educated (calculated) investments in supplemental heating and lighting or crops in the fall, winter and spring.
Results from 2011 season and data show that the time temperature and light requirements are:
Broccoli, 34,000 GDDf*HD (40 F base), from Seeding.
Corn, 27,000 GDDf*HD (50F base)
Greenstar Lettuce, 25,000 to 30,000 GDDf*HD (40 F base), from Seeding.
Vulcan Lettuce, about 28,000 GDDf*HD (40 F base), from Seeding.
So, this data suggests that if one had a greenhouse with lighting and heat so that Light and Temperature were controlled, And if the Light was on 16 hours per day, and the temperature was a constant 75F, that Vulcan Lettuce should grow from seed to a full head in 50 days (assuming a 40 F base). Similarly, if to save fuel, the greenhouse was at 65F, the same lettuce should take 70 days to grow to full size.
Of course, conditions in the field are much more variable and complex.
Introduction:
Diversified farms grow many crops. Grocery store, farm stand, and restaurant customers want a steady supply of “every” crop. Scheduling plantings for constant supply of crops through the year is a major planning question for growers. Planning a constant supply of many crops is a challenge because time to maturity varies throughout the season, and there are no planning tools to help growers predict time to harvest.
Blue Ox Farm is a certified Organic Vegetable Farm in Enfield NH with about 10 acres tilled in 2011. We grow for the local Coop, local Restaurants, Farmers Market, and CSA. We grow about 1 to 1.25 acre each of Lettuce and Broccoli, and about ½ acre of corn. The remaining is a wide variety of crops. Most of the soils are on the wet side, new fields are less wet.
Becky Sideman was the technical advisor on the project. Cynthia Walthour did the seeding, most of the tracking of the plants through the seeding and transplanting, and much of the data collection during harvest (particularly Lettuce), with Dave Winters also helping with the lettuce harvest. The corn and Broccoli were harvested by Dave Winters, Jake Torrey, Dave Gagne, and Steve Fulton through the season (the same people also did the data recording on the cards). Steve Fulton did some of the tracking and most of the analysis of the data, Steve was also responsible for setting up and placing the thermal probes.
Planning a years production of crops that are planted in succession requires experience, plus a bit of guessing what the weather might be during the season. This report presents a methodology to eliminate some of the guessing that is based on historical weather information from a growers area.
Plants need time, light, and heat to grow. Growing Degree Days (GDDf) are a relatively well known method to understand biological processes such as when an insect pest will appear, or when a crop will be mature. Data for crop maturity shows that the GDD for a crop to mature is not constant, but varies through the season. Some, or maybe all, of the variation is due to day length variation.
A proposed means to combine the effects of time, light, and heat are the Growing Degree Day * Hours Daylight (GDDf*HD) .
Of course, conditions in the field are much more variable.
Cooperators
Research
All crops in this study were started as transplants in 2 inch soil blocks, grown to a transplantable size, and transplanted using a waterwheel. A set of waterproofed card stock tags (5.5 x 8.5) was used to track each planting, see photos.
At seeding the seeding date was recorded on the tags with a permanent marker and they were placed in the transplant trays. Each tray in the planting was also marked with a wooden marker that had the crop name and seeding date.
At Transplant the transplant date was recorded on the tags and the tags were placed on wire hoops, one at each end of the planting. The wooden tags were (generally) also placed in the row in case the plastic coated tags blew before the wire hoops were set.
During harvest the tags were used to record the dates harvested, and as much as possible, the amount harvested.
• Broccoli. We recorded the date harvested, and the approximate weight of crowns harvested. (we harvest in 5 gallon buckets, one 5 gallon bucket being about 10 pounds. For fractions of a bucket we estimated the weight based on the volume).
• Corn. We recorded the date and the number of top ears harvested
• Lettuce. We recorded the date and the number of heads of lettuce harvested. We also tried to record heads that were harvestable (we have a ¾ pound minimum head size). We also tried to leave at least one head per planting and see when that head bolted to establish a maximum harvest window.
We had no issues with the seeding. We did not seed on the exact dates on the seeding schedule all the time, but we use the actual date seeded, not the planned dates for data analysis.
Transplanting was done when the plants were ‘ready’ to transplant, and when the tractor was available with a crew. This was a once or twice a week occurrence. The first three plantings of corn were transplanted on the same day as the field was not ready, and the ground was cool, so the first planting of corn was pretty large (about 1 foot tall), generally we transplanted corn at 4-6 inch. Similarly, the first 3 plantings of lettuce, and 2 plantings of broccoli were planted on bare ground, without raised beds or plastic due to practical issues. One planting of lettuce (of each type mid season) bolted in the trays as the plants were too small one week, and bolted by the next.
Harvest. We had some problems with the permanent markers writing on the plastic tags when the tags were wet with rain, or morning dew. So the yield data was not as comprehensive as we wanted (but yield was not a primary focus of this study).
Fields used are flat and open with full sun. Soils are primarily Silt loams with Moderate drainage, Pootatuck, with some Rippowam (poor drainage). The first three Lettuce, and 2 broccoli were on Agawam fine sandy loam.
Temperature was measured using ONSET TMC50-HD thermal probes, and recorded on ONSET HOBO model U12-008 data loggers. The thermal probes with the U12 data logger are listed as being accurate to +0.45F (for 0 to 50C range), the U12 Data loggers record the data to 0.05 F resolution. The data-loggers were programmed to take thermal measurements every 30 minutes (48 per day). Thermal probe response time is listed as 2 minutes in air.
Daylight hours was taken from Website for standard day length in Lebanon, NH. This was not measured.
Data analysis was done with HOBOWARE version 3.1.2 and Microsoft EXCEL.
Soil temperature probes were all placed 4 inches below the soil surface, with at least 12 inches of cable buried as well. A 4 inch deep trench was dug, and the thermal probe pushed into the side wall of the trench so that it would be under un-disturbed soil. The trench was then filled. There were two air temperature probes for each crop.
Crop Lower Air Probe height Upper Air Probe Height
Broccolli ~6 to 12 inch above ground ~1.5 ft above ground
Corn 1.5 ft above ground 4 to 5 ft above ground
Lettuce ~6 to 12 inch above ground ~1.5 ft above ground
There are many definitions of Growing Degree Day (GDD). The general concept is that biological activity (growth for plants) is related to the amount of Time at or above a temperature.
Growing degree day for a single day is:
GDD (Growing Degree Day)= ((Max Temp+Min Temp))/2-(Base temp)
(this is the average temperature for the day, less the base temperature. It is never negative.)
Min temp = minimum air temperature that day
Max temp = Maximum air temperature that day
Base Temp = base temperature, below which no plant growth is presumed to occur.
I used Hoboware version 3.1.2 to calculate all GDD presented herein. The two methods used were the Average data method, this is based on the minimum and Maximum temperatures for the day, as well as the base temperature (this only requires the daily HIGH and LOW to calculate). I also used the “Actual” data method which is a more continuous integration of the data. Air Temp#1 was used for analysis. If not specifically identified, the data presented is done with the average method so that it is applicable to those without data collection systems, but with a means to record and track the daily high and low temperatures. Limited comparison of the two methods shows that they are similar.
For instance, based on the Lettuce data set, there were 5,136 GDDf*HD (average)in 2011 based on 40F base temperature and using the AVERAGE method, using the same data and the ACTUAL method for GDD, we had 4,591 GDDf*HD (actual) in 2011. The Broccoli temperature data set shows 4,808 (ave) vs 4,535 (Actual). I suspect that the difference between Lettuce and Broccoli probes are that the Lettuce probes was in full sun, the Broccoli probes was shaded by the plant when the plant grew up.
Base temperatures were determined from references.
Broccoli, 40 F
Corn, 50 F (proposal says 60F, further research says 50F is the best estimate in the literature)
Lettuce, 40 F (Wikipedia says 42F (5.5 C), I use 40F in calculation)
This is defined (by me) as GDD for a day, times the day length (in hours). Day length is defined by the length of time between actual dawn, to actual twilight. (NOT civil time, which is 1 hr longer per day).
I calculate (using Excel and average annual temperatures) that the Enfield Lebanon area will have 4,468 GDDf with a 40 F base. Data set for 2011 says that 2011 season had 4808 GDDf (40 deg base) from May 10 to 12-1-2011 (using the Average method, which is based on the min and max temp in a day, not an integration). So, 2011 was warmer than average. This was my guess without looking at the data.
Average temperature 2011
40 F base, average method 4,468 GDDf 4,808 GDDf
50 F base, average method 2,603 GDDf 3,114 GDDf
60 F base, average method 1,309 GDDf 1,810 GDDf
Method for forecasting when a crop will be ready to harvest
Planning a years production of crops that are planted in succession requires experience, plus a bit of guessing what the weather might be during the season. This report presents a methodology to eliminate some of the guessing that is based on historical weather information from a growers area.
Plants need time, light, and heat to grow. Growing Degree Days (GDDf) are a relatively well known method to understand biological processes such as when an insect pest will appear, or when a crop will be mature. Data for crop maturity shows that the GDD for a crop to mature is not constant, but varies through the season. Some, or maybe all, of the variation is due to day length variation.
A proposed means to combine the effects of time, light, and heat are the Growing Degree Day * Hours Daylight (GDDf*HD) .
The data suggests that the GDD*HD (Growing Degree Days * Day Light Hours) is as follows
Everest Broccoli, 34,000 GDDf*HD (40 F base), from Seeding.
Mystique Corn, 27,000 GDDf*HD (50F base) ), from Seeding
Greenstar Lettuce, 25,000 to 30,000 GDDf*HD (40 F base), from Seeding.
Vulcan Lettuce, about 28,000 GDDf*HD (40 F base), from Seeding.
So, this data suggests that if one had a greenhouse with lighting and heat so that Light and Temperature were controlled, And if the Light was on 16 hours per day, and the temperature was a constant 75F, that Vulcan Lettuce should grow from seed to a full head in 50 days (assuming a 40 F base). Similarly, if to save fuel, the greenhouse was at 65F, the same lettuce should take 70 days to grow to full size.
Of course, conditions in the field are much more variable. Figure 1 shows the average daily temperatures (high and low) for Lebanon NH. It also shows the number of daylight hours in Lebanon NH (http://www.weather.com/outlook/travel/businesstraveler/wxclimatology/daily/USNH0123 reference).
Figure 2 shows the GDDf (Growing Degree Days in Fahrenheit days) through the year, also with a 40 F base temperature. This was developed using the Average Annual temperature from a weather station near the farm that reports summary temperatures to the Internet, not Data taken at Blue Ox Farm.
Figure 3 looks at the same information, only on a cumulative basis, also for a 40F base. Note that I have added arbitrary 3F warmer, and 3F cooler lines in the plot to start to think about how temperature variations in the real world should effect the growing of plants (relative to a schedule).
Similarly Figure 4 shows at the Cumulative GDDf*HD through the year.
So if a Lettuce seed is planted on May 27 (day 147), it should be harvestable 28,000 GDDf from that date, or July 31 in an average year, July 26 in a 3F warmer year, and August 8 in a 3F cooler year (all in Enfield). Vulcan seed planted in 2011 on May 27 was harvested July 23rd! Greenstar was July 24. 2011 was warm. (or the Blue Ox fields are warmer than the not-fully-specified sites in Lebanon or Enfield from the Internet from which the 2010 and the average historical temperature Information was taken). Figure 5 compares the 2011 data to average temperature data from the Internet for “the area”.
Figure 6 shows that 2011 was generally warmer than 2010 most all summer and fall. (or that the two locations are this different).
Planning tools for growing
The main output from this project was to be a planning tool, or tools, that could be used. This section is that tool. These plots are valid for Enfield NH. Farms at the same latitude will have the same day length, temperatures vary as well. These plots are made in Excel and require the day length, and the high and low temperatures for a location to make the predictions.
Graphically we can show when we seeded the plantings, and when we harvested to see what happened in 2011. Figure 7 and 8 show the harvest date vs seeding date for lettuce. These give a guide for when the crop is harvestable vs the date seeded.
We then calculate when the lettuce would be ready during a year with average temperatures. This is seen in Figure 9, as well as with the 2011 data.
Similarly for Broccoli, Figure 10 shows the Harvest date vs seeding date data, and Figure 11 shows calculated values for an average year.
Note that the data from 2011 shows that Everest Broccoli came in earlier than forecast. Of course there is season to season variation, but also, all crops were started in a heated greenhouse, this was all plantings prior to June 1. Further, the first two plantings of broccoli were planted in different soil, and covered by row cover for about 2 weeks, a further boost to growth.
What is happening in the fall is not as clear. In large part this is due to 2011 fall being warm, but also, the historical data cuts out the peak temperatures, which may be significant in marginal growing conditions.
Lettuce shows the same effects.
Figure 12 shows the Harvest date vs seeding date data for Corn, and Figure 13 shows calculated values for an average year.
Corn, Mystique (variety from Johnny’s Selected Seeds, listed by JSS as a 74 day maturity)
Total Planned 9,500
Total Seeded 11,165
Total transplanted 5,080 (?)
Total Harvested 4,628 (?)
Total sold 7,031 (?)
Assume Base temperature = 50F
Figure 14 shows the days to harvest for the corn used in the experiment, Mystique.
Figure 15 shows the same data, but plotted harvest date vs Seeding date so one can see when to seed for a given Harvest date.
Broccoli (Everest from Seedway) 40F Base Growing Temperature
Total Planned 11,280
Total Seeded 11,385
Total transplanted 8,900 (on record)
Total Harvested and sold as Crowns 2,060# (on record)
Total sold as crowns 2,256# (includes 2nd variety not in study)
Data for Broccoli
Figure 17 and 18 show the days to harvest, and the Harvest date data plotted for Everest broccoli.
Figure 19 shows Growing degree days to harvest for Broccoli, 2011 season. Note DDf is Degree Day Fahrenheit. Note that the first 6 plantings were started in the greenhouse and in the greenhouse until transplant. So, will have a different # GDD than outside, and thermal probes started May 9, 2011. The first 6 plantings are not shown in the “from seed to harvest” plot above.
Figure 20, GDD*DH vs date seeded shows that this is almost constant. Something that would be expected if we were accounting for everything of importance. A positive sign that this is an accurate representation. Note that the first 6 plantings did not have full set of temperature data from seeding and are not plotted for that reason.
Lettuce, Greenstar
Total Planned 6,270
Total Seeded 6,380
Total transplanted 4,410
Total Harvested 1,993
Total sold as Greenleaf 3,894
Total sold as Lettuce 2,718
Total sold as Lettuce, leaf 148
Totals include Tropicana in Greenleaf,
Totals include Tropicana and red leaf for leaf lettuce, plus romaine as Lettuce
Figure 21 shows the Days to harvest vs date seeded for Greenstar Lettuce. Another Physiological limit is bolting. This sets the longest that the lettuce can be in the field. The bolting limit is not in the harvest window, so potential harvest window is larger than presented herein, as the maximum days to harvest is a few days shy of bolting.
Figure 22 shows the Harvest date vs date seeded for Greenstar Lettuce.
Figure 23 shows the GDD*HD vs date seeded for Greenstar Lettuce. This is not quite a constant. Not sure why it is not constant, but it is between 25,000 to 30,000, about the same as Vulcan as you will see soon.
Total Planned 4,510
Total Seeded 4,510
Total transplanted 3,045
Total Harvested 1,428
Total sold as Redleaf 2,772
Total sold as Lettuce 2,718
Figure 24 shows the Days to harvest vs date seeded for Vulcan Lettuce. Another Physiological limit is bolting. This sets the longest that the lettuce can be in the field. The bolting limit is not in the harvest window, so potential harvest window is larger than presented herein, as the maximum days to harvest is a few days shy of bolting.
Figure 25 shows the Harvest date vs date seeded for Vulcan Lettuce.
Figure 26 shows the GDD*HD vs date seeded for Vulcan Lettuce.
All temperature probes seem to be working correctly, and agree reasonably well with each other when at the same conditions.
Figure 27 shows the Minimum and Maximum Air temperature from the Broc1 probe for the season.
Air temperatures over the crops are more easily understood (to date) by looking at data over a period of a few days, rather than the season. Figure 28 shows the temperatures in the crops from July 1 through July 7. Soil temperatures (at 4 inch depth) vary much more than I expected during the day. The broccoli soil temperature is consistently lower temperature than the corn soil temp and the Lettuce soil temp. It is not understood what happened to Lettuce air temp #1 on 7-7. Likely there is a shielding from the sun issue.
Figures 29 to 35 show individual days through the growing season, roughly every 30 days. The broccoli soil temperature is consistently lower temperature than the corn soil temp and the Lettuce soil temp. Note that the agreement between all thermal probes is generally pretty good.
Figure 36 shows the soil temperatures for the season. Hard to tell what is happening here as there is so much information.
Figure 37 shows the temperature difference between the Broccoli and lettuce soil temperatures. Broccoli soil temperature Averages 1.6F below Lettuce soil temp. Both are on white plastic, so presumably the Broccoli plants shade the plastic more than the lettuce.
Figure 38 shows the temperature difference between the Lettuce and Corn soil temperatures. Corn Averages 0.5 F above Lettuce soil temp, Presumably this is the difference between the bare soil (corn) and white plastic. However, I had expected a much bigger difference due to the white plastic. The difference in early June is much larger, this is when the corn is small and does not shade the bare soil, when the corn has grown taller, the difference is not as large as the soil on the corn is shaded, and the lettuce is still largely in the sun but on white plastic.
See attached
The data suggests that the GDD*HD (Growing Degree Days * Day Light Hours) is as follows
• Everest Broccoli, 34,000 GDDf*HD (40 F base), from Seeding.
• Mystique Corn, 27,000 GDDf*HD (50F base) ), from Seeding
• Greenstar Lettuce, 25,000 to 30,000 GDDf*HD (40 F base), from Seeding.
• Vulcan Lettuce, about 28,000 GDDf*HD (40 F base), from Seeding.
So, this data suggests that if one had a greenhouse with lighting and heat so that Light and Temperature were controlled, And if the Light was on 16 hours per day, and the temperature was a constant 75F, that Vulcan Lettuce should grow from seed to a full head in 50 days (assuming a 40 F base). Similarly, if to save fuel, the greenhouse was at 65F, the same lettuce should take 70 days to grow to full size.
Education & Outreach Activities and Participation Summary
Participation Summary:
Summary report submitted to Becky Sideman for inclusion in her monthly Vegetable and Berry newsletter.
Project Outcomes
Future Recommendations
Verify that the data is repeatable. Testing for another year would be very useful.
Develope an Excel spreadsheet / tool that makes the forecasting easier. The graphical technique is good, but the spreadsheet would make the forecasting faster and more precise. Such a model would help the farmer look at the effects of warmer and cooler years as well.
Extend the testing into winter growing to see how the model stands up. The approach and testing seems that it would be easier to test in heated greenhouses in the winter. This model presents a means to calculate the growth response to business decisions such as what temperature should the greenhouse be set to, and how economic is adding lights to a greenhouse.