Our combined ‘research and extension approach’ includes a controlled experimental design (objective 1) comparing silvopasture access to grass pasture access for small flocks of broilers, executed at the Shenandoah Valley Agricultural Research and Extension Center (AREC). For objective 2, we will perform a field trial with a similar approach as objective 1, but applied at commercial farms. On each of the site (with possible exception of Wil Crombie’s farm), we will have a silvopasture-flock and a grass pasture-flock serving as a control to compare specific parameters related to welfare, economics and environment.

At the end of this study, an extension report will illustrate all these potential benefits and will contain photos and input from new and experienced silvopasture/broiler chicken producers than can serve as an educational document for industry stakeholders. Two field days will provide additional information on feasibility of the integrated system for interested parties and educate industry stakeholders on the system possibilities (objective 3).

Year 1 Objective 1 - Experimental trial: Compare broiler chicken production with access to existing silvopastures to broiler production with access to grass pasture, at the VT Shenandoah Valley AREC

Two experiments were conducted from April to May (Experiment 1; Exp 1) and June to August 2021 (Experiment 2; Exp 2). All procedures were approved by the Virginia Tech Institutional Animal Care and Use Committee (IACUC protocol 20-044).

Animals and housing 

In total, 886-day-old Ross 708 mixed-sex chicks in Exp 1 and 648 chicks in Exp 2 were obtained from a commercial hatchery (Harrisonburg, VA, USA). Birds were Marek’s vaccinated at the hatchery. Upon arrival, chicks were arbitrarily selected and housed in 12 identical pens (5.7 m²) with 73 or 74 birds per pen in Exp 1 and 53 or 54 birds per pen in Exp 2. Pens contained pine wood shavings (~5 cm depth), a heat lamp (day 1-8), a cardboard feed flat with feed (day 1-8), one bell drinkers (Plasson® Broiler Drinker complete, Or-Akiva, Israel), and one pet champion poultry drinker (Stout Stuff LLC, made in china), and one feeder (Superbowl poultry feeder, LaGrange, NC, USA). The chicks were fed commercial starter (day 0-15), grower (day 15-25), and finisher (day 25-42 or 43) diets meeting NCC recommendations (National Chicken Council 2017). Ambient temperatures were 35°C on day 1 and gradually reduced to 23°C on day 22 (Exp 1) or day 23 (Exp 2). Lighting was provided continuously for the first week and reduced to 12h light and 12h dark until day 22 (Exp 1) or day 23 (Exp 2).

In Exp 1, birds from each pen were equally but randomly allocated over 16 pasture-based treatments resulting in 53 birds per plot originating from all 12 pens. In Exp 2, complete pens (53-54 birds) were randomly allocated to 12 pasture-based plots with chicken coops. On day 22 (Exp 1) or day 23 (Exp 2), birds were transported for 1.5h to the pasture-based experimental site located at the Shenandoah Valley Agricultural Research and Extension Center (AREC) in Raphine, VA, USA. After transportation, birds were kept inside the coops for two days in Exp 1 and one day in Exp 2 to get acclimated to their new housing conditions. From days 24-41 or 42, coop doors in each plot were opened at approximately 8:00 am and closed at approximately 5:00 pm.

All pasture-based plots (125m², 16 in Exp 1, and 12 in Exp 2) contained a chicken coop  (6.55m²) constructed from wood, chicken wire, and tarp . The coops contained a wooded platform perch (0.05m x 0.1 m x 2.4 m) and the same feeder and bell drinker as when housed indoors. Coops were moved laterally across the plot each week. Plots were fenced with 1m-high and 50-meter-long FlexNet electric fences (PoultryNet®, Washington, IA, USA), connected to a 30-volt electric cattle fence.

Treatments

The silvopasture plots (8 replicates in two locations in Exp 1 and 6 replicates in two locations in Exp 2; Figure 1) contained a 10-12-year-old mixed hardwood stand of black walnut (Juglans nigrea L.) and hickory (Carya spp. Nutt.) trees and 30 newly planted saplings per plot (American hazelnut (Corylus americana), black walnut (Juglans nigra), persimmon (Diospyros virginiana L.), southern red oak (Quercus falcata Michx.), and southern pine (Pinus spp.) of approximately 30 cm height and 1 cm calipers. Saplings were planted in six rows with inter and intra-row spacing of 1.5 m. Canopy cover in all silvopasture and open pasture plots were calculated from photos using ImageJ software (1.5.3k, National Institute of Health, USA). The images were taken at ground-level in the center of the plot straight upwards. Photos were converted to 8-bit, binarized, and then the number of black (canopy) and white (sky) pixels were calculated as a percentage of total pixels. The canopy cover for silvopasture plots was (mean ± standard deviation) 31.7 ± 16.7% in Exp 1, and 33.3 ± 10.9% in Exp 2.

The open pasture plots (8 replicates in two locations in Exp 1 and 6 replicates in two locations in Exp 2; Figure 1) contained ground vegetation. Ground vegetation in the open pasture and silvopasture plots were similar and consisted of tall fescue (Festuca arundinacea Schreb.), orchardgrass (Dactylis glomerata L.), honeysuckle (Lonicera japonica Thunb.), greenbrier (Smilax spp. L.), Virginia creepers (Parthenocissus quinquefolia L. Planch.), horsenettle (Solanum carolinese L.), common milkweed (Asclepias syriaca L.), and thistle (Sonchus spp. L.). Canopy cover in open pasture plots was 0 ± 0% in Exp 1 and Exp 2.

Experimental setup for objective 1 at the VT AREC. The orange rectangles indicate 8 silvopasture plots that will hold 50 broilers each. The maroon rectangles indicate the 8 grass plots that will hold 50 birds each.

Data will be collected, focusing on aspects of animal welfare, economics and environment.

Animal welfare. Range use may benefit leg health, footpad health and fearfulness. For all sixteen 50-bird flocks, range use will be assessed via two approaches; cameras (e.g. Bushnell Trophy Cam) and live observations. One photograph per hour will be taken within the chicken coop and used to calculate the percentage of birds outdoors each hour. Hourly live observations will be performed at plot level to count birds in the range between 8 AM and 5 PM on days 29, 30, 34, 35, 40, and 41. The observer will sit down 5 m from the plot to prevent changes in the birds’ behavior. The percentage of birds near, middle, and far from the coop, and the total percentage of the flock in the range were calculated.

Assessment of how the management systems affect bird fearfulness will be determined with a Tonic Immobility test. The test consists of placing a chicken on its back in a U-shaped cradle and restraining it for 15 sec (Stadig et al. 2017). The chicken is then released and the time until righting (latency) is recorded, with longer durations of tonic immobility associated with greater fear. The test will be performed in week 5, after birds have had sufficient time to adopt to their respective environments, and 10 birds per coop will be assessed.

On the following day, leg condition will be assessed via a latency-to-lie test, in which birds are placed in a tote with a small amount of water. The time (latency) until the birds sit down will be recorded as an indicator of leg health or strength. In addition, footpad dermatitis and hock burns will be scored on a 0-4 categorical scale, with increasing scores indicative of worse lesions (Welfare Quality® Network 2009).

Economics. Economic productivity of the poultry system will be assessed by measuring animal growth (coop-level n = 50 per coop), feed conversion (coop-level, n = 8 per treatment) and whole-carcass yield (individual level n = 10 per coop). Feed intake will be recorded per group, starting from day 1 by weighing provided and left-over feed (difference = consumed feed). All birds will be weighed prior to processing on day 42 and carcass weights will be recorded thereafter for a sample of 10 birds per coop.

Environment. Temperature and humidity will be monitored in the range and within the coops using temperature-humidity sensors. We will assess soil quality on each plot twice to assess potential benefits from silvopasture/broiler production integration for the environment. Outdoor range soil quality will be assessed at three locations in each range (n = 16). To determine soil chemical properties, samples will be systematically collected at four distances from the coop (close, middle, end of range), resulting in 4 samples per flock per sampling time (baseline or post-experiment; total sample n(16*4*2)=128). At each location the surface litter will be removed and a sample collected with bucket soil augur. All samples will be homogenized separately by depth in a bucket onsite to obtain a single 50-g sample for each depth. Samples will be air dried and sieved with a 2-mm soil sieve. Soil sample analyses will be done to assess C and N, soil microbial biomass carbon (MBC), microbial biomass nitrogen (MBN), C and N fractions, and phospholipid fatty acid content (PLFA). The latter is used as a biological indicator of overall soil quality (Quideau et al. 2016).

We propose to quantify insect diversity using a combination of direct observations, pitfall traps and sticky cards. Flowering plants will be observed directly to assess the quantity and diversity of pollinators, which will be grouped into the following categories: lepidoptera (butterflies), bumblebees, honeybees, other native bees, hoverflies, and wasps. Sticky cards will be used to assess flying insects, especially lady beetles. Pitfall traps will be used to assess ground beetles and orthoptera (especially crickets). We expect insect abundance and diversity to be greater in complex habitats with trees and native grasses than in conventional grass pastures.

JACOBS and GRAD STUDENT will be responsible for animal measures and soil collection with the technician, PENT will be responsible for silvopasture related measures, the environmental measures will be coordinated by JACOBS and soil samples will be analyzed through the South Dakota State University Soil Analysis Service Lab led by Dr. Sandeep Kumar (please see the letter of commitment for this contractual service here: Letter of Commitment SD STATE). 

Year 1-3 Objective 2A - Field trial: Compare broiler chicken production with access to a newly planted silvopasture on a large-scale commercial farm to grass pasture access on the same farm

Our cooperating large-scale VA-based poultry producer (SVO farmer Pam Miller) operates 3 commercial broiler chicken houses for approx. 90,000 birds. Each of the houses used in their system has pop holes along its length; these holes provide birds with access to grass pasture. On one of these three pastures we will establish a new silvopasture incorporating woody species and native grasses. Targeted woody species include Red Mulberry (Morus rubra), hazelnuts, pawpaw, and serviceberry. These species grow fairly rapidly and can provide shrubby cover and long-term feed resources. In addition to shrub cover, we will plant switchgrass (a native warm-season species) using sprigs (rather than seed) to generate rapid cover. Switchgrass is of interest given its potential to trap and reduce dust, to store carbon in deep root systems, and to serve as a source of bedding material or heating fuel for poultry operations.

The silvopasture system with trees, shrubs and grasses will be planted in rows parallel to the length of the house. The row-planting and space in between grasses, trees and shrubs (within rows) will allow for birds to move freely throughout the pasture. We will work with the landowners to design and install a mix of woody species and switchgrass based on their preference of vegetation and the physical layout of the space. The flocks with access to this newly established silvopasture plot will be compared to a control flock in one of the two adjacent houses.

For three production rounds spread over two years, we will collect data on animal welfare parameters (described for objective 1: fearfulness, leg health, footpad dermatitis and hock burns) in a sample of 80 birds per flock (two flocks per round) at week 5 of age. In addition, farmer and processing plant records will provide information on mortality, animal yield and carcass rejections. Litter moisture and soil quality data (6 samples per treatment (2), per round (3) = 72 soil samples total) will be assessed as described for objective 1.

Year 1-3 Objective 2B - Field trial: Compare broiler chicken production with access to established silvopastures on two small-scale commercial farms to grass pasture access on the same farms

 Similar to objective 2a, animal welfare, economic and environmental parameters will be assessed on two small-scale broiler operations (VA and MN). At these sites, producers already have integrated chicken production with silvopastures, or they have trees and shrubs as part of the outdoor range for broilers. The setup on the VA farm (small-scale producer Brent Wills) will be similar to the large-scale farm in that two flocks will be simultaneously grown, one with and one without access to shrubs and trees in their outdoor range (pasture versus silvopasture). All their birds have access to an outdoor range when temperatures allow. Broilers have full access to pasture and/or woods at all times once they come out of the brooder, usually at 2-3 weeks old, depending on time of year and weather. They grow a commercial hybrid breed (Freedom Rangers). All of their pastures border woods and fringe areas, so the birds have access to woods, trees and shrubby type vegetation, typically oak, hickory, maple, sourwood, pine thickets and the associated brambles and brushy vegetation. Their flock sizes depend on demand. Their pens allow for 80 birds per pen. They use Salatin-style pasture pens that offer 120 sf of “indoor” space per pen (10’x12’) as well as an A-frame pen with the same floor dimensions. For this study we will collect data during three rounds of production, with birds either with access to pasture (conventional outdoor range), or to vegetation (silvopasture).

On the MN farm (small-scale producer Wil Crombie), they grow birds according to the Regeneration Farms system/design for 5 years now. They have 40 acres and planted 20,000 hazelnut shrubs, around 5,000 elderberry bushes and a few thousand tree species, including oak, sugar maple, chestnut, basswood and Honey Locust. They currently have two active chicken barns. Their birds have outdoor access to grasses, Comfrey, Hazelnut and Sugar maple leaves and other naturally occurring plant species that are inside the paddocks. They also sprout grains within the paddocks. A single flock consists of 1,500 birds and is housed in a steel structure barn (18'x90'). The birds have outdoor access to circa 1.5 acres. At this point, we are not sure yet whether we can simultaneously assess birds with and without vegetation access as on the other farms and at the AREC.

For both farms, measurement include fearfulness, leg health, footpad dermatitis, hock burns on 80 birds per flock (n = 2 per farm, per round), assessed a week prior to processing. In addition, farmer and processing records will provide information on mortality and animal yield. Litter moisture and soil quality (6 samples per treatment (2), per round (3) = 72 soil samples total per farm) will be assessed as described for objective 1.

JACOBS and GRAD STUDENT will be responsible for animal measures, and collection of soil samples together with the technician, FIKE and MUNSELL will be responsible for silvopasture related measures. The environmental measures will be supervised by JACOBS and soil samples will be analyzed through the South Dakota State University Soil Analysis Service Lab lead by Dr. Sandeep Kumar.

Year 1-3 Objective 3: Increase adoption of poultry-based silvopasture practices

3A. Surveying silvopasture- and traditional-system poultry producers about their experiences, concerns and opportunities for applying these practices

Online search plus network search of southeastern US-based small and large scale silvopasture producers in combination with poultry. In addition, the director of the American Pastured Poultry Producers Association (APPPA) is willing to assist in the search for volunteers for the interviews among their members. We aim to perform at least six quantitative in-depth interviews in person, over the phone, or via email depending on geographical location and producer preference. Questions will relate to productivity, animal welfare, and environment. Additionally, we will ask about particular challenges and how they were overcome, including potential exposure to predators. We will collect photo material for the extension publication; a case-study report (b) (online and hard copies). Role of farmers: partake in interviews, allow or provide photos at/from farms. GRAD STUDENT will interview, perform farm visits, supervised by JACOBS.

3B. Creating and disseminating case studies from abovementioned and other established silvopasture producers through extension

Outcomes from interviews and farm visits will be structured and organized into an extension publication with photos and text. JACOBS will coordinate, GRAD STUDENT will write, collate, and disseminate (online and hard copies). MUNSELL, DOWNING, PENT and FIKE will advise on extension documents. Consultants and farmers will be involved with development and dissemination.

3C. Showcasing poultry silvopastures through research center and on-farm field days and web-based delivery tools

Two field days will be organized during which interested stakeholders are welcomed to visit at least one of the participating farms. The large-scale farmer already indicated willingness to assist with this field day and allow for visitors. The Shenandoah Valley AREC will be the location of a second field day for interested parties. These days will provide interested parties to see and experience a silvopasture approach and ask questions about this system. Some research outcomes will be presented to attendees for education purposes. Role of farmer: allow access to farm and talk about experiences. JACOBS will coordinate, FIKE and PENT will lead actual field days.

3D. Disseminating technical and budget information to producer and agency communities

Based on data collected for objective 1 and 2, we will be able to formulate some poultry-silvopasture system scenarios and calculate associated budgets for those scenarios. These will be disseminated as technical information with targeted audience being industry stakeholders. MUNSELL and DOWNING will coordinate silvopasture aspects of the technical and budget outcomes, JACOBS will coordinate animal-related aspects. Consultants and farmers will be involved with development and dissemination.

For objective 1 and 2, we will repeatedly assess the same measures at four different sites: (1) the Shenandoah Valley AREC, (2) a large-scale farm where we will plant a new silvopasture, (3) and (4) two small-scale broiler farms with established silvopastures. At all sites, we aim to compare measures in flocks with and without access to silvopastures (meaning grass range versus silvopasture range). This results in flocks as the experimental unit, with at (1) 16 flocks (8 with and 8 without silvopasture access), (2) 6 large-scale flocks (3 with and 3 without silvopasture access), (3) 6 small-scale flocks (3 with and 3 without silvopasture access), and (4) 3 small-scale flocks with silvopasture access. The commercial farms differ in farm size (small-scale versus large-scale), “age” of silvopasture system (newly established (year 0) versus well-established (~year 5)), and flock access to vegetation within a farm (flocks either have access to grass pasture or access to silvopasture). The first factor will illustrate the feasibility of integration vegetation production and broiler chicken production on a large scale compared to a smaller scale farm. The well-established silvopasture systems will be on small-scale farms, as the approach has not yet been used by large poultry producers. Our approach, in which we will plant a silvopasture system on a large farm, will provide insight regarding benefits associated with new silvopasture systems. The established systems (small-scale farms and the experimental trial from objective 1) will serve as “proof of concept” for a more commercial application, and will illustrate potential benefits on a longer term. Thus, the large-scale production system will illustrate the potential benefits on the short term, and the small-scale farms benefits on the long term.