Yield monitor data allow for evaluation of both spatial and temporal yield variability for all fields, soil types, and management zones within a specific farm. Individual farms benefit from having proper yield records by knowing annual yield (bu/acre) at the whole field level; yield at the field level and soil type within field level; and yield at soil type level within the farm. With this knowledge comes an understanding of where you can invest your time and money. When multiple years of data are available, management zones can be derived. Such management zones can help with evaluation of and implementation of management decision that can increase yield and yield stability over time. In addition, multi-year yield data for specific fields can be used to quickly evaluate field or soil type specific yield potentials.
In the past year, work has focused on collecting soybean yield data as part of a regional project to evaluate soil type specific yield potentials on individual farms and to develop a yield potential database for soybean. Additional years of data are needed to create a critical mass of yield data for statewide assessment, and enable development of multi-year yield reports for farmers. In addition, recently a study measured the impact of headland area on whole field and farm corn silage and grain yield and found that in headland areas yield was, on average, 14% (grain) and 16% (silage) lower than in non-headland areas of the field (Sunoj et al. 2020). A similar assessment of the impact of headland yields on overall field and farm yield can help farmers make decision about investments in headland areas.
Here we propose to expand the number of years of soybean data included in the project by including 2020 yield data for already participating farms and expand with inclusion of data from five additional farms in western New York. In addition, we will evaluate yield hits on headland areas. As part of the evaluation, economic analysis of partial budgets for headlands versus main portions of the field will be conducted to answer the question: “Are headlands worth fixing?”
The grain yield data obtained with yield monitoring equipment contain a variety of errors due to machine and operating characteristics such as (1) rapid velocity changes; (2) travel time/ crop flow delay between harvest location and the location of the sensors that read volume and moisture content; (3) start pass delay, end pass delay as flow ramps up/down; (4) unknown harvester width, (5) overlapped data near end of rows; and (6) stops in fields: crop throughput near 0 speed => erroneously high yields. Thus, data cleaning is needed before yield data can be used to set yield potentials and develop management zones. A protocol for removing errors in an efficient and consistent manner from both silage and grain corn yield monitor data was developed to ensure high quality data from field to field, from farm to farm, and from year to year. A similar protocol is being developed for soybeans.
As we continue to evaluate and update the Cornell corn yield database (100,000 acres of corn grain and silage) through the collection of whole farm yield monitor data, we need to also build a database for soybean yields through the addition of more farms and acreage. It is still a challenge to bring a database together, even with the amount of acreage included for corn grain and silage. With increased participation, this project will become increasingly useful for both the individual farmers who participate and the larger group of soybean producers.

Part A: Data Collection and Individual Farm Reports
Data will be obtained for the 2020 growing season for 5 farms already participating in 2019, and five new farms will be included, with 2020 and previous years of data. We currently have 12 farms participating since the start with multiple years of data. Data cleaning will be done as we currently do with the 2019 yield data. Each of the participating farms will receive a farm-specific report with yield per harvested soybean field, yield per field without headland areas included, yield per soil type within a field, and yield per soil type on the farm. For each farm and field with at least 3 years of data, multi-year reports will be generated.

Part B: State Soil Type Based Yield Potentials
The data from all ten farms will be pulled together to determine soil type specific yield distribution histogram that show average yield per soil type and ranges in yield across farms and years. The results will be summarized in an extension document and be the start of a New York soybean yield database.

Part C: Economic Analyses of Headland Areas of Fields
Partial budgets will be done to determine the economic impact of a yield hit on headland areas within fields. This analyses is done to evaluate production costs versus yield on headland areas versus the main portions of the field, to determine economic soundness of decisions like reducing seed population or adding a tillage pass to break up compaction layers.

Talks will be given to stress the importance of data cleaning and standardization of methods across farming units (equipment, software use, etc.) and the information that can be gained when accurate soybean yield records are being kept for multiple years. Findings will be summarized in individual farm reports, extension presentations, and extension articles (local newsletters, What’s Cropping Up?).

Updated February 28, 2022:
Requesting a no cost extension due to my departure and return to extension.

Soybean farmers who know their actual yield, yield potential based on their soil type and understand the impact of headland yields on overall field and farm yield will be able to make better decisions on their investments and management practices.