Are weeds a concern when dredged ditch sediments are applied to agricultural fields?

Abstract

Agricultural ditches constructed to provide an outlet for improved subsurface drainage systems (i.e., tile drains) are ubiquitous in the poorly drained but highly productive soils of the midwestern United States. Many ditches require regular maintenance to remove deposited sediments due to reduced hydraulic capacity and blockage of subsurface tile outlets. Dredge material is often spoiled along the edge-of-field as an economical means of disposal. Placement of dredge materials at the top of a ditch can impede surface drainage, creating depressional areas in fields that can negatively impact crop growth.

An alternative approach to spoil management would be to place dredge materials further into fields to level the ground or as a soil amendment. Studies have shown the potential beneficial effects of lake and river dredge soil amendments on crop yields ( Brigham et al., 2021; Darmody & Diaz, 2017). However, farmers may be reluctant to spread dredge materials in fields due to concerns over redistribution of weed seed banks accumulated in deposited ditch sediments. In this study, we collected sediments from ditches in three regions of Ohio and germinated the weed seed banks in a greenhouse mesocosm study. We tested for regional differences in number of weed species and dry biomass per mesocosm. Additionally, we assessed potential controllability of the weeds using common chemical herbicides.

We collected sediment samples from ditches at three sites from each of three regions (NE, northeast; NW, northwest; and W, west) in Ohio (Figure 1). Sediment samples were collected in December 2020 following a series of freeze–thaw events. Sediments were allowed to dry, then packed into mesocosms (13.0 inch × 8.0 inch plastic containers) to a depth of 4.0 inches and arranged on a table in the greenhouse using a randomized complete block design with four replications (Figure 2). Sediment mesocosms were watered using a mist irrigation system and soil moisture levels adequate for germination were maintained throughout the duration of the study. During the first 45 days, greenhouse temperatures were maintained between 50–60°F for germination of winter annuals. Ambient temperatures were then raised to 78–90°F for an additional 45 days for germination of summer annual weeds. At the termination of the study, aboveground biomass was collected, identified, and dried for the determination of biomass for each species by mesocosm. Differences in means were tested with one-way analysis of variance and pairwise comparison of means using Student's t and least significant differences (α = 0.05) in JMP Pro 15.2.0 software (SAS Institute, Inc.). Dry biomass and number of weed species were response variables, region was the factor, and data were blocked by replication. To determine potential controllability of germinated weed species, we consulted the labels of several common herbicides used on agronomic crops in Ohio, including glyphosate, glufosinate, 2,4-D, atrazine, and mesotrione.

In total, 50 weed species were germinated from ditch sediments collected from across the three regions of Ohio (Table 1; includes weed scientific names). Ditch sediments from the W region had the highest number of weed species (38 species), followed by the NW (27 species) and NE (25 species) regions. Thirteen weed species were present in sediments from all three regions; these included: annual sow thistle (Sonchus oleraceus L.), barnyard grass [Echinocloa crus-galli (L.) P. Beauv], black medic (Medicago lupulina L.), common chickweed [Stellaria media (L.) Vill.], corn speedwell (Veronica arvensis L.), eastern black nightshade (Solanum nigrum L.), hairy bittercress (Cardamine hirsuta L.), knotweed (Polygonum cuspidatum Siebold & Zucc.), lambsquarter (Chenopodium album L.), purselane speedwell [Veronica peregrina L. ssp. xalapensis (Kunth) Pennell], quackgrass [Agropyron repens (L.) P. Beauv], wild carrot (Daucus carota L. var. sativus Hoffm.), and yellow wood sorrel (Oxalis stricta L.). Sixteen weed species were identified in two regions and 21 species were found in only a single region. The W region had the highest mean number of species per mesocosm (7.8 ± 3.5 per mesocosm) followed by the NW (5.8 ± 1.6 per mesocosm) and NE (4.8 ± 1.5 per mesocosm) regions. There were differences in the means for the number of species between the W and NE regions, but no statistical differences were detected between the NW and other regions.

The NE region had the highest weed biomass (2.68 ± 1.07 g per mesocosm) with quackgrass, black medic, Canada thistle [Cirsium arvense (L.) Scop.], Pennsylvania smartweed (Polygonum pensylvanicum L.), and stinging nettle (Urtica dioica L.) constituting 63% of the total. The NW region produced the second highest total weed biomass (2.28 ± 1.61 g per mesocosm), with quackgrass, wild carrot, and giant ragweed (Ambrosia trifida L.) accounting for ∼40% of the total. Weed biomass in the W region was lowest (1.97 ± 1.73 g per mesocosm) and was composed predominantly of knotweed and quackgrass, which accounted for ∼30% of total biomass. Across all regions, quackgrass, black medic, and knotweed were the most dominant species and accounted for more than one-third of the total biomass. Overall, there were 41 species of broadleaf weeds totaling 78% of biomass compared with nine grass species with 22% of biomass (Tables 1 and 2). Results of the statistical analysis indicated no differences in mean total biomass per mesocosm by region.

These findings demonstrate that dredged ditch sediments are likely to be a source of weeds if applied to fields as a soil amendment for agronomic crops, though most species identified are either commonly found in production environments or are not typically problematic. Both perennial weeds, Canada thistle and quackgrass, could be concerning if establishment occurs, though currently available herbicide options and common cultural practices (i.e., tillage) may limit establishment. Among the species identified, pigweed (Amaranthus L.) and ragweed are two that could present trouble to producers if not controlled quickly after their presence is identified. These species are common in corn (Zea mays L.) and cotton (Gossypium hirsutum L.) fields and can develop tolerance against weed chemicals such as glyphosate (Webster & Nichols, 2017).

This study suggests that the application of dredged ditch sediments to fields used for agronomic crops is likely to introduce seeds to the weed seed bank. However, the introduced species were identified as common for the production environments in the region. In terms of herbicide control options, all weed species were treatable by preemergence and postemergence herbicides, such as glyphosate, glufosinate, atrazine, mesotrione, and 2,4-D (Table 2). Herbicide resistance of the identified species was not tested but may be an issue to consider when evaluating this practice. Knowledge of nearby production fields, herbicide use history, and the presence of resistant populations may need to be considered when determining if this is a viable practice. With concerns about the many types of weeds that could be introduced, we confirmed that the majority of them were broadleaves, which are treatable with common chemicals. Lastly, we could conclude that the type of weeds would vary depending on where the sediment was sourced. Obtaining sediment samples from three different regions of Ohio demonstrated weed types will not always be consistent, which will need to be considered when creating a weed management plan. Future work should include combined greenhouse and field experiments of weed establishment from dredge materials to determine if greenhouse studies are a viable means to test which weed species are likely to establish under field conditions. Further research may examine ways to minimize the viability of weed seeds in dredge sediment prior to application to reduce this concern in the future.

Mathew Simmons: Investigation; writing—original draft; writing—review and editing; Jonathan Witter: Conceptualization; funding acquisition; investigation; supervision; writing—review and editing; Daniel Voltz: Investigation; writing—review and editing; Ryan Haden—Conceptualization; investigation; supervision; writing–review and editing; Alexander Lindsey: Supervision; writing—review and editing.

The authors declare no conflict of interest.