Remote sensing of perennial crop stand duration and pre-crop identification

Authors: Mueller Warrant, George; Whittaker, Gerald; Trippe, Kristin

Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/26/2016
Publication Date: 11/3/2016
Citation: Mueller Warrant, G.W., Whittaker, G.W., Trippe, K.M. 2016. Remote sensing of perennial crop stand duration and pre-crop identification. Agronomy Journal. 108:2339-2354.

Interpretive Summary: Whether fields are disturbed for planting of new crops or remain in continued production of established perennials is vitally important to grass seed growers. The strong relationship between use of tillage in seedbed preparation and the subsequent movement of soil particles, nutrients, and agrichemicals from fields to streams and rivers interests the public. Crucial pieces in knowing how well grass seed growers are doing in maintaining perennial stands in production are the ability to measure stand ages and characterize crop rotation patterns across the entire region. Remote sensing classifications of crops and other landuses over the past 11 years provided this basis. Our first conclusion was the importance of improving classification accuracy, year-to-year consistency, and overall temporal extent of our landuse data. Our second conclusion was discovery of the highly localized nature of spatial variation in stand longevity of perennial grass seed crops. Fields whose perennial ryegrass or tall fescue stands only remained in production for two years were often directly juxtaposed with those of much longer durations. This implied a strong need for access to detailed crop management history if reasons for spatial variation in stand duration are to be eventually understood. Our third conclusion was the finding of general stability over time in key details of crop rotation practices. With the partial exception of cereal grain crops, the frequencies at which particular crops preceded (or followed) each other during rotations between perennial grass seed stands showed no consistent trends over time in the period from 2004 to 2014. While year-to-year variability in the frequency of specific rotational crops certainly did exist, popularity of crops tended to randomly wax and wane rather than develop into clear long-term trends. Our final conclusion was recognition that crop rotation patterns in perennial grass seed crops were numerous and complex, with at least a dozen ways to transition from some previous perennial grass seed stand into a newly established tall fescue.

Technical Abstract: Field to field variability in soil erosion and off-site transport of nutrients and pesticides in western Oregon in any single year is primarily driven by the question of whether individual fields were disturbed for planting of new crop stands or remained in production of established perennial crops such as tall fescue grown for seed. The issue of whether given perennial grass seed crop stands remain economically viable or must be terminated due problems with weeds, insects and other invertebrates, diseases, and yield declines is a central concern for most seed growers. We used 11 years of remote sensing classifications to identify multi-year stands of established perennial crops and the annually disturbed agricultural production occurring between such stands. Crucial requirements for successful measurement of perennial crop stand durations and characterization of crop rotation patterns included: (1) optimization of year-to-year landuse consistency and within-year landuse classification accuracy, (2) inclusion of a sufficiently large number of landuse classes to capture nearly all of the crops grown in the diverse agriculture of this region, (3) possession of enough years of data that a majority of fields had the opportunity to experience at least one full cycle of planting new stands of perennial grass seed crops, maintaining perennial stands in production until growers decided to terminate them, and subsequent planting of new crops. Averaged over all possible starting and ending years between 2004 and 2014, stand durations defined as continuous production of the same crop on the same field ranged from highs of 2.9, 4.6, 3.7, and 5.9 years for perennial ryegrass, orchardgrass, tall fescue, and Italian ryegrass grown for seed down to lows of 2.4 and 1.5 years for clovers and cereals, primarily winter wheat. Stand duration variability included a strong local component in which fields with much longer or shorter than average final ages were frequently found in close proximity to one another. Rotation patterns involving perennial grass seed crops were complex.