Managing Agricultural Drainage Ditches for Conservation Biological Control on the Delmarva Peninsula

Progress report for LNE20-408R

Project Type: Research Only
Funds awarded in 2020: $197,728.00
Projected End Date: 11/30/2023
Grant Recipients: University of Maryland, College Park; University of Maryland, Eastern Shore; University of Delaware
Region: Northeast
State: Maryland
Project Leader:
William Lamp
University of Maryland, College Park
Expand All

Project Information


1) Problem, Novel Approach and Justification.

Agricultural drainage ditches serve a necessary function for water management on all farms. Especially for the flat Delmarva Peninsula (coastal plain portions of Delaware, Maryland, and Virginia) and portions of New Jersey, ditches are necessary to drain the high water table during the year. In addition, ditch habitats may also provide valuable ecosystem services to producers by enhancing natural enemies of pests and increasing their biocontrol. More broadly, ditches serve as examples of how uncultivated land in and around fields can provide ecological benefits. The goal of this proposal is to determine how to manage ditches to enhance populations of natural enemies, and to work with farmers to test management methods on farms.

2) Hypothesis and Research Plan.

The hypotheses are that ditches can be managed to increase numbers of natural enemies and to increase their biocontrol of pests in adjacent fields. The objectives are designed to complement each other and will be conducted over three years. During Year 1, we will focus on testing specific management practices to help determine which practices applied to drainage ditches will enhance natural enemies. Experiments will be conducted primarily on University farm ditches. During Years 2 and 3, we will test management practices as a way to enhance natural enemy populations and performance in ditches and adjacent crops. Experiments will be conducted on University and private farms. At the whole farm scale, the final objective will be conducted using two ditches on each of five farms, over three years of sampling, to determine if ditch management does indeed enhance natural enemies and suppress pest populations by biocontrol.

3) Outreach Plan.

Results from Objectives 1 and 2 will be delivered to farmers participating with Objective 3 as ditch modification practices are discussed and selected. Additionally, pamphlets detailing beneficial arthropods, pests controlled by them, and the habitats that favor beneficial arthropods will be produced and disseminated to farmers, extension personnel, and conservation groups. Project performance will be highlighted at other in-season events, such as research field days. Websites will be used to disseminate video clips, project summaries, and photographs of ditch management methods and natural enemies. Results will be shared at winter meetings, and project briefs will be prepared for trade journals.

4) Project Objective.

Our specific objectives include 1) to conduct experiments designed to enhance populations of natural enemies (e.g., spiders, predatory mites and beetles, and parasitoid wasps) in ditches by providing resources for shelter, food, and reproduction, 2) to compare natural enemy performance in ditch habitats and adjacent crop fields across a range of habitat conditions, and 3) to use ditches on farms to demonstrate the value of this enhancement of natural enemies for pest management.

Project Objective:

Our objectives are 1) to conduct experiments designed to enhance populations of natural enemies (e.g., spiders, predatory mites and beetles, and parasitoid wasps) in ditches by providing necessary resources for shelter, food, and reproduction, 2) to compare natural enemy performance in ditch habitats and adjacent crop fields across a range of habitat conditions, and 3) to use ditches on farms to demonstrate the value of this enhancement of natural enemies for pest management.


Agricultural fields are frequently disturbed, and natural enemies that serve as biological controls of pests must colonize the fields from elsewhere.  Drainage ditches, because they are frequently left undisturbed, may be a source of natural enemies.  Our long-term research is designed to determine what natural enemies are found in agricultural drainage ditches, and which ones move from ditches into fields during the production season.  Specifically for this project, our research aims to determine what ditch management practices will aid natural enemies to reproduce in ditches and to disperse from ditches into crop fields.  In this way, ditches may serve a valuable function of conservation biological control, whereby predators and parasites will suppress populations of arthropod pests.

An example of such a natural enemy from our research are jumping spiders, family Salticidae, in which certain species occur in drainage ditches in the spring.  As potential prey abundance decreases in the ditch during the summer, the spiders disperse into neighboring crop fields to locate new types of prey.  These predators follow a typical life history pattern of beneficial species: they reproduce in unmanaged areas and survive overwinter, then in late spring/summer they disperse into crop fields and locate injurious insect pests before reproducing in late summer and fall.  We seek to encourage natural enemies to use ditches located within crop fields, and thus become a source of dispersing biological control agents into the adjacent fields.

Although our research funding began in spring, 2020, the pandemic stopped any potential research until June.  As a consequence, this year we only performed preliminary research, and postponed most of our research to be conducted in 2021-2023.  The Lamp lab established a study on an agricultural drainage ditch at the University of Maryland Wye Research and Education Center as a randomized complete block design with four treatments and four blocks.  During summer and fall, we used pitfall traps, sweep samples, and sticky card traps to identify species of natural enemies and the pattern of their occurrence.  For example, from the pitfall traps, we identified lycosid and linyphiid spiders, carabid and staphylinid beetles, formicid ants, gryllid crickets, and several families of parasitoid wasps.  We have initiated treatments on the plots to determine how ditch management practices such as mowing or the addition of straw may help beneficial natural enemy species.

In another study, the Zebelo lab performed research to determine the movement of natural enemies from ditches to adjacent crops, and to measure predation rates of those natural enemies on insect pests.  Measurements were taken in plots in which vegetation was cleared using chemicals while adjacent plots were left uncleared.  Natural enemy abundance was higher adjacent to uncleared ditches, suggesting that they are a source of natural enemies such as parasitic wasps, spiny stink bugs, ladybird beetles, and pirate bugs.  Using caterpillars as prey, they measured predation rates on cabbage adjacent to the ditch plots.  They measured high predation rates, however no differences were found between cleared and uncleared plots.  The experiment will be repeated next year.

Studies planned by the Owens lab were cancelled because of the pandemic.  The start delay to June prevented the planting of various flowering plants, designed to attract natural enemies.  In addition, restrictions by the University of Delaware prevented the recruitment of wage labor.  The project will be discussed during the 2021 AgWeek Agronomy session and additional farms will be located for studies.  In addition, a survey is being prepared for use during the winter meetings to determine how agricultural drainage ditches are managed. 


Click linked name(s) to expand/collapse or show everyone's info
  • Dr. Simon Zebelo (Educator and Researcher)
  • Dr. David Owens (Educator and Researcher)


Materials and methods:

Our proposal has two hypotheses. First, we hypothesize that ditches can be managed to increase numbers of natural enemies by providing needed resources for shelter, food, and reproduction of beneficial species. Second, we hypothesize that ditches can be managed by planting flowering and non-flowering species to increase their biological suppression of pests in fields adjacent to ditches, thereby lowering pest populations and decreasing the need for insecticide applications.

1. To determine how ditches can be managed to enhance the number of natural enemies in ditches, an experiment was begun on an agricultural drainage ditch at the University of Maryland Wye Research and Education Center.  Sixteen plots were created, set up in four blocks of four plots, and each plot was assigned a treatment.  No treatment was applied during the summer, 2020, so the sampling of natural enemies was conducted to identify species and their abundance before any treatment was started.  One treatment was started in fall: the addition of straw to the sides of the ditch.  Other treatments will start in spring, 2021.  Sampling consists of pitfall traps, sweep samples, and sticky card samples.  We hypothesize that the addition of straw, and perhaps other management practices, will enhance the numbers of natural enemies compared to the control plots.

In 2021, to determine the effect of different timings of mowing and the spreading of straw along ditch banks on ground beetle abundance, sixteen plots were set up in four blocks along a drainage ditch at the Wye Research and Education Center. One of four treatments was randomly assigned to each plot within each block. The four treatments consisted of the combinations of two states of two variables: mowing in the fall vs. mowing in the spring, and straw spreading vs. no straw spreading. The plots were sampled every three weeks during the 2020 field season as a control before treatments were applied. Sampling consisted of pitfall trap, sweep net, and sticky card samples. A layer of straw was spread over half of the plots in the fall of 2020, and half of the plots were mowed in the spring of 2021. The plots were then sampled in the same manner during the 2021 field season. Abundances of natural enemies in the 2021 samples will be compared to the samples from 2020. We hypothesize that the addition of straw in the fall and delaying mowing until the spring will have provided greater overwintering habitat for natural enemies such as ground beetles, leading to enhanced numbers during the field season.

The above photo shows a drainage ditch at Wye Research and Education Center in the process of spreading straw during fall 2021, effectively illustrating three of the four treatments. The nearest plot has been mowed for the fall, while the plot towards the back will not be mowed until spring 2022. The nearer ditch bank on the left has had straw spread over it, while the bank on the right had not yet been covered with straw. 

2. Also in 2021, we set up farmer-assisted research on five farms.  We identified two sections of ditches that were adjacent to the same crop field.  The two sections did not differ in 2021, but over the winter, 2021-22, one of the ditches was mowed as usual in the fall, whereas the other section was left unmowed or partially mowed to help conserve natural enemies.  Within each section, we established three transects including a sampling location in the ditch, 5 ft into the crop (corn or soybeans), and 30 ft into the crop.  At each sampling location, we collected natural enemies by three trapping methods: sweep net, sticky trap, and pitfall trap.  We sampled all farms for each month of the summer (June, July, August).  These data will be used to compare the overwintering practice tested on each farm. 

At the end of the 2021 season, we discussed alternatives for how each farmer could preserve natural enemies in their ditches.  All the farmers in the study normally mow their ditches close to the ground in the fall to reduce weed growth (especially woody species).  To help preserve natural enemies, the farmers chose one of their ditch sections to either wait until spring to mow the ditch, or only partially mow the ditches to about 6 inches of height.  The other ditch was mowed as usual.  All ditch sections and transects will be sampled for natural enemies next summer, and the mowing treatments will be reapplied during the fall-winter of 2022-23.

3. In DE, to compare various treatments of land adjacent to ditches to enhance natural enemies, at four locations, 50-ft plots were arranged in a randomized complete block design with four treatments and four replicates along the length of the ditch. At the Carvel REC, plots were 12-ft wide and installed on the NW side of a tax ditch. Replicates were separated by a 50’ft buffer. Wheat and soybean were planted adjacent to the plots. At the Gumboro location, plots were 5-ft wide but installed on each side of a field drainage ditch for a total plot width of 10-ft. Corn was planted on each side of the ditch. At the Princess Anne site, plots were 10-ft wide but also installed on each side of a field drainage ditch for a total plot width of 20-ft. A 5-ft buffer separated each plot. At the Middletown site, plots were installed into a large grassy swale cutting through a corn field. Plots were 20-ft wide and separated from each other by 5-ft.

Wildflower, buckwheat, and mulch plots were prepared by trimming existing vegetation and residue. Straw mulch was added in a 2-in layer. One standard sized bale covered approximately 200 sq ft. Wildflower and buckwheat plots were treated with glyphosate (Makaze, 5 qts/acre) with a CO2-pressurized backpack sprayer fitted to a 9-ft boom equipped with 6 8003 nozzles and calibrated to deliver 18 GPA at 26 PSI.  Plots were treated twice prior to residue incorporation via Mantis tiller to prepare the seed bed. Wildflower seed was broadcast using a hand spinner at a rate of 9.7lbs /acre. Buckwheat seed was spread using the same method, this time at a rate of 50lbs /acre. After seeding was completed, the soil was raked to incorporate the seed.

Each plot was sampled using three methods: yellow sticky card traps, pitfall traps, and sweep nets. Two sticky cards (3-in x 5-in) were placed in each plot (opposite sides of ditches in locations where both sides are used) approximately at the height of the existing vegetation. These cards are placed 5-ft from each end of the plot. A pitfall trap, consisting of two 16 fl oz solo cups nested in each other, was installed in each plot and partially filled with a propylene glycol solution (RV Antifreeze). A rain cover (1-sqft) covered each pitfall trap but provided a couple inches of clearance so as to not inhibit arthropod movement. Plots were also sampled with a 15-in diameter sweep net (10 sweeps per plot) at each sample date. The adjacent crop was sampled using pitfall traps approximately 25-ft into the field from the center of each plot, and 10 row-ft were visually assessed for insect pest presence and injury at the edge of each plot and 25-ft interior. Sweep samples were placed in paper and Ziploc bags and placed into a freezer for storage until samples could be processed. Pitfall trap contents were placed in 70% ethanol-filled vials for storage.

A 2-3 week period of dry weather following plot preparation and seeding negatively impacted wildflower germination and growth. Buckwheat performed better than wildflowers. Samples are currently being processed, but anecdotally, buckwheat plots seemed to attract a greater number and diversity of beneficial insects. Weed and grass control in the plots was a challenge for buckwheat and wildflower seedling establishment. Several pitfall trap samples were lost due to vertebrate activity (raccoons/foxes) and at two locations, farm equipment.

4. At locations near UMES, to compare natural enemy performance in ditch habitats and adjacent crop fields across a range of habitat conditions. two Agricultural drainage ditches were selected in the Delmarva region in the Bowman road (Location-1) and Stewart Neck road (Location-2), Princess Anne, Maryland. In ditch in location-1 was between cornfields, and the ditch in location-2 was between two soybean fields. Within each location, there were uncleared ditches and cleared. In location-1, the cleared and uncleared ditches had three 6 ft apart raised beds where the Brassica was grown at 1.5ft spacing (Figure 1 A). At Location-2, potted Brassica plants lined the ditch. Only in location-2 in both cleared and uncleared ditches had Brassica plants positioned 0ft, 100ft, and 200ft from the ditch. In each bed or between potted plants, three 10-inch yellow sticky traps were mounted. Every week, traps were removed, and the number of beneficial insects counted and replaced. Pitfall traps were also used to trap bugs in and around ditches. Each cup received 200ml of diluted propylene glycol solution. Three cups per plot were emptied, counted, identified, and replenished weekly. The results were compared using a paired T-test between cleared and uncleared ditches. 

Evaluating the activity of natural enemies: Predation rates provided by ditch resident predators were quantified using corn earworm (Helicoverpa zea) and cabbage looper (Trichoplusia ni) larvae as sentinel prey. For estimating larval predation, ten third-instars were placed on the upper leaves of four randomly selected plants per plot. After 24 h of exposure in the field, the remaining larvae were counted to determine the number of larvae consumed by predators. Moreover, the sentinel larvae were reared for parasitoid emergence. To distinguish sentinel prey predation from unknown losses due to handling and rainfall, we enclosed one plant per site in a cage that excluded natural enemies. Plants in these cages were infested with ten sentinel prey and field-collected predatory wasp, Polistes spp (Figure 3) Net mortality due to predation was determined by assessing mortality from uncaged plants and subtracting it from mortality from caged plants. The experiment was repeated three times per plot as described in [2] (Figure 2). 

The number of beneficial insects trapped with sticky traps was higher in uncleared ditches than the cleared in both locations (Figure 6). The beneficial insects include soldier beetle, stinky bugs, lacewing, ladybird beetles, different types of bees, and unidentified wasps. The ticky and pitfall traps revealed that the abundance of insects is higher in uncleared ditches when compared with the cleared ditches (Figurer 5 and 6).

In the preliminary trial in location-2, the abundances of insects decrease as the distance increases from the cleared and uncleared ditches (Figure 4). However, more insects were recorded in uncleared ditches in all distance points (Figure 4). 

The number of prayed larvae was higher in the uncleared ditches in both locations, except the cabbage lopper larvae were prayed higher in the cleared ditches (Figure 7). Interestingly, the corn earworm larvae recovered from the larval predation trial in location-1 were parasitized by braconid wasp, Microplitis croceipes, 41 wasps, emerged from the corn earworm larvae collected from the cabbage plants near the uncleared ditch. 

In summary, The presence of vegetation in uncleared ditches created a habitat and a food source for insects, resulting in a higher insect population level. The level of natural enemies was higher in uncleared plots when compared with the cleared plot. The vegetation in the uncleared plot includes narrow-leaved grasses, mainly the Eragrotis family, and the broad-leaved flowering plants include gold star flowers, queen Ann lace, datura…etc.). This vegetation might support the insect population. In addition, if the agricultural ditches were maintained well with flowering plants, they might enhance the population of beneficial arthropods ( pollinators and natural enemies).

Research results and discussion:

From samples from ditches, to date, we have identified 47 families of arthropod natural enemies.  A large number of the families are common in ditches, and also can be found in the adjacent crops.

The photo above shows parasitoids recovered from corn earworm larvae collected from field (18 families of hymenopteran parasitoids have been recorded from ditches in 2021) (A). Polistes spp predator prying on cabbage looper and corn earworm larvae (B).    

Planting of areas adjacent to ditches led to mixed results, largely because of rainfall patterns.  We have learned to plant earlier in the year for better results.

Samples are continuing to be processed, so many results are not yet clear.

Participation Summary
5 Farmers participating in research
Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.