Posted on

burning a weed seed

In this project we aim to test the hypothesis that fall flaming can efficiently and effectively kill weed seeds on the soil surface. While flaming is widely used to kill small weed seedlings, and field stubble burning is known to kill weed seeds, we are not aware of any research in which flaming has been used to target seeds. This idea comes from the experience of Rob Johanson, a diversified organic farmer in Maine who, while flaming potato vines in particularly weedy fields observed smoldering weed residues and reasoned that he was likely killing weed seeds. He has further observed that flamed plots generally have lower weed pressure the following year. We aim to confirm this observation with a series of on-farm experiments evaluating three flame doses and their effect on the following season’s germinable weed seedbank and early-season weed seedling densities. Laboratory studies conducted with condiment mustard (Sinapis alba), indicated that a 900C propane flame required 3 sec. or more exposure to reach mortality levels of 80%; mortality was 100% at 5 sec. of exposure (Figures 1 and 2). Field studies will include higher temperatures to reduce the required exposure time, and will evaluate the combined stresses of flaming and overwinter mortality. Results will be shared with other growers during a Field Day and a session on “Managing the Seedbank for Improved Weed Management” at the 2012 Maine Organic Farmers and Gardeners Association Farmer to Farmer Conference. We will also produce a short video on fall flaming which will be posted to our YouTube Channel (zeroseedrain), and the national eXtension website.

This project will include field experiments conducted at one farm in Maine, and another in Vermont. The on-farm studies will be conducted on a uniformly cropped weedy field following cash crop harvest. At each farm, in the fall of 2011, weeds will be flail mowed and four replicated strips will be subject to the following treatments: no flaming; flaming at 40 kg propane per ha; flaming at 80 kg per ha; and flaming at 160 kg per ha (note: a 20-40 kg rate would kill sensitive weed seedlings). In the spring of 2012, soil sampling and subsequent greenhouse germination assays will measure the weed seedbank in each treatment. Ten soil cores (6.5 cm diam by 10 cm deep) will be randomly chosen from each plot, bulked, and spread in greenhouse flats, and watered to encourage weeds to germinate. Seedlings will be removed monthly, followed by drying and crumbling of soil and re-watering (four cycles in total). To measure the contribution of fall flaming on weed management in the subsequent crop, in June of 2012 we will record weed density after the final cultivation event at each farm. Specifically, eight 0.5 sq. m quadrats will be censused in each replicate of each treatment. Weeds will be identified by species and counted.

Project objectives from proposal:

Objective: To measure weed seed mortality in response to flaming in the field.

Propane usage will be recorded by weight using portable field scales, and will be used to estimate per ha costs for the flaming treatments. Thermocouples and a portable temperature recorder will be installed in each treatment to record the thermal dose.

Performance Target: We expect that the germinable weed seedbank density will be inversely related to the propane flaming rate. Further, the spring seedling census data will proportionally reflect the effects of flaming on the germinable seedbank, with the lowest surviving weed density in the 160 kg per ha propane treatment, and the greatest weed density in the control treatment. Adoption of this promising management practice will depend on the expected tradeoff between faming cost and weed control benefits measured the following year.

An integrated weed management approach requires alternative management practices to herbicide use such as tillage, crop rotations and cultural controls to reduce soil weed seed banks. The objective of this study was to examine the value of different tillage practices and stubble burning to exhaust the seed bank of common weeds from the northern grain region of Australia. Five tillage and burning treatments were incorporated in a field experiment, at Armidale (30 degrees 30’S, 151 degrees 40’E), New South Wales, Australia in July 2004 in a randomized block design replicated four times. The trial was continued and treatments repeated in July 2005 with all the mature plants from the first year being allowed to shed seed in their respective treatment plots. The treatments were (i) no tillage (NT), (ii) chisel ploughing (CP), (iii) mould board ploughing (MBP), (iv) wheat straw burning with no tillage (SBNT) and (v) wheat straw burning with chisel ploughing (SBC). Soil samples were collected before applying treatments and before the weeds flowered to establish the seed bank status of the various weeds in the soil. Wheat was sown after the tillage treatments. Burning treatments were only initiated in the second year, one month prior to tillage treatments. The major weeds present in the seed bank before initiating the trial were Polygonum aviculare, Sonchus oleraceus and Avena fatua. Tillage promoted the germination of other weeds like Hibiscus trionum, Medicago sativa, Vicia sp. and Phalaris paradoxa later in the season in 2004 and Convolvulus erubescens emerged as a new weed in 2005. The MBP treatment in 2004 reduced the weed biomass to a significantly lower level of 55 g/m2 than the other treatments of CP (118 g/m2) and NT plots (196 g/m2) (P < 0.05). However, in 2005 SBC and MBP treatments were similar in reducing the weed biomass. In 2004, the grain yield trend of wheat was significantly different between CP and NT, and MBP and NT (P < 0.05) with maximum yield of 5898 kg/ha in CP and 5731 kg/ha in MBP. Rainfall before the start of the second trial season promoted the germination of a large numbers of weeds. SBC and MBP treatments reduced the numbers of most of the individual weed species compared with CP, SBNT and NT. SBC was able to destroy a large proportion of seeds most likely through burning and burying some in the soil and was found to be the best treatment in exhausting the seed bank followed closely by MBP which probably buried large number of seeds deep in the soil and promoted others to germinate. CP might have buried some of the seeds in the top 5-10 cm but also promoted parts of the seed bank to germinate. SBNT and NT provided an ideal medium for weeds to germinate and resulted in heavy infestations of weeds.