Producing upscale pork for small-scale farmers: An Appalachian application

Final Report for FNE06-593

Project Type: Farmer
Funds awarded in 2006: $8,254.00
Projected End Date: 12/31/2007
Region: Northeast
State: West Virginia
Project Leader:
Chuck Talbott
Black Oak Holler Farm, LLC
Expand All

Project Information

Summary:

Note to readers, attached is the complete final report for FNE06-593

The goal of our project was to examine the impact of raising hogs in the woods to harvest mast and find ways to utilize on-farm feed resources that may enhance the flavor of pork. Sixteen market sized hogs (barrows and gilts averaging 6 months of age and weighing 224lbs) were randomly selected from contemporary littermates and placed in a 9.9 acre section in a predominate oak-hickory woodlot in Mason Co. WV. Another group of sixteen hogs (230 lbs average starting wt.) were placed on pastured lots and fed 16% corn-soy (free choice) ration and represented a production and meat sensory control. Hogs were introduced in the woodlots on 9/10/06 and moved to another fenced woodlot (Plot 2: 8.0 acres) on 10/24/06. Animal impact on the forest ecology is based on animals harvesting mast in Plot 1 for the six week period described. Woodlot hogs were offered 50 lbs/day (approximately 3 lbs/d per hog) whole corn and offered a mineral salt block free choice. Informal taste tests from two workshops were conducted and pork from four hogs consuming mast and the corn-soy diet where analyzed for meat quality and tested by a sensory panel at WVU.

Post grazing forest species richness (S) and species diversity (H) were approximately one-half pre-grazing values (S = 7.1 and 3.5, H’ = 1.5 and 0.9, pre- and post-grazing). Eight species that were tallied in the pre-grazing inventory were absent post-grazing. Estimates of percent soil disturbed averaged 3.3%; 11 of the 15 milacre plots showed no soil disturbance, and the range of disturbance was 2-20%. Japanese honeysuckle (Lonicera japonica) was noted on three pre-grazing sample plots with an average percent cover of 3.7%. This species possesses characteristics typical of many invasive species: it grows quickly, fruits within a single season, produces small but abundant seed that is very persistent in soils, has seeds that adhere to animal fur or clothing, is nonpalatable, and easily invades disturbed habitats. It is cause for concern in forested ecosystems in the region, because it is capable of invading forestlands and swiftly replacing natural communities with nearly monospecific stands. Post-gazing sampling occurred after this grass was senescent, so an evaluation of the potential spread of this species was inconclusive. A reassessment of the 15 milacre plots will be conducted again this August. No damage (girdling, buttressing etc.) to the timber stand, including saplings, pole-sized trees and sawlog-sized trees due to pig foraging was noted. There was no evidence of soil compaction.

Over the 56 day trial, the hogs consuming corn-soy gained 118 lbs while the pigs on tree mast gained an average of 8 lbs. The mast fed pigs became very attentive to the “role-call” for dinner while acorn crunching became more sporadic. We could not reschedule an earlier kill date with our packer and this became a problem. When the pigs were scheduled for harvest (first week in November), I predict they had lost 30-40 lbs. For this experiment, hogs should have been harvested at six weeks rather than eight weeks. At $0.10/lb for feed, the average corn soy pig consumed $59 worth of feed during the trial while the mast fed pig consumed $17. Based upon prices paid for by the processor ($.60/lb live wt. for corn-soy and $1/lb for acorn fed pigs), the income over feed costs (IOFC) for the corn-soy pigs is $49 while the IOFC for acorn fed hogs was $115 for mast fed hogs. Preliminary results suggest that hogs should be 275-300 lbs (with 1.25inches backfat) when placed in the woods. The objective is to incorporate the acorn flavor into the fat.

Scores from the informal taste tests at the Lowe Hotel suggested that the acorn pork roasts had more flavor than pork from corn-soy fed diets. Participants observed no differences in tenderness, juiciness or overall evaluation in the loins for the two diet treatments. On the contrary, pork from corn-soy fed hogs was considered more flavorful, juicy and tender than acorn fed hogs at the farmer workshop. Differences in weights and fat deposition of the shoulders due to diet and exercise may have influenced the taste test at the farm. The shoulder from the acorn diet was 11 lbs lighter than the corn-soy shoulder and may have dried out due to excessive cooking. Pork with higher levels of unsaturated fats have lower melting points and may require different cooking strategies to retain moisture and flavor. Professional cooks are recommended when cooking for taste tests.

Untrained sensory panelists at the WVU meats lab tended to favor acorn-fed pork flavor over corn-soy pork and characterized the acorn pork flavor as nutty, sweet and earthy. Acorn pork was darker than corn-soy pork and may reflect the significantly higher intramuscular fat stores in the corn-soy pork over mast fed diets (5.92% vs. 4.07% respectively) or possibly the muscle fiber type (not analyzed). Saturated fats in the subcutaneous fat and muscle tissue from the acorn diets were lower (8.6% and 3.7% respectively) than from the corn-soy diets; unsaturated fats increased 6.1% in the back fat and 2.6% in the muscle when acorns were fed instead of corn-soy diets. More research is needed to assess the heart-health benefits of feeding a diet of acorns to market hogs. In the last two weeks of the experiment, the catabolism of fat of the forest hogs to support energy requirements and maintenance probably influenced the amount and type of fatty-acid accumulation in the muscles.

The predator fears that I had originally cited have been unfounded so far. Coyotes, mountain lions and bears are occasionally heard from at night or sited in the woods. The two strand electric fence proved effective in keeping 16 hogs in a 16 A wood lot. The secret to keeping pigs in electric fence lots is to provide adequate feed and water at all times. The unpredictable nature of mast availability as a reliable feed source adds to the challenge.

Flavor and (possible) health benefits of Appalachian locally and sustainably raised pork must be conveyed to the communities in the context of food security. We must convince our policy makers to invest in “off the grid” farming systems, lest we forget the traditional knowledge of Appalachia, i.e curing and preservation of meat and vegetables in root cellars. The next focus of our work will be to collaborate with existing programs (Slow Food Bluegrass, Collaborative for the 21st Century, WV and KY Extension Service) to document the contribution of the farmer and offer him real value ($) for his work.

Introduction:

Many people refer to pastured or forest raised pigs as “old timey” farming, straight out of the book, Fox Fire1. A better term (that farmers understand) might be profitable farming, especially if farmers can produce a unique product that stands out from the majority of corporate produced hogs that are characterized as the “other white meat” (essentially from the same lean genotype, fed the same diet, etc.). The goal of this project is to help small-scale/part time farmers find new markets and higher profit margins by raising swine in ways that may enhance the flavor of pork (diet, genetics and management systems) as well as the environment they are raised in.

Many species of swine evolved in sylvan environments where cooler temperatures provided by a natural canopy aid in thermoregulation during the warm seasons. Woodlands also supply a vast amount of nutrients in the form of nuts/mast from forest canopies, roots and tubers, and carrion. Appalachian farmers and homesteaders typically ear notched their 150-200 hogs to harvest chestnuts and acorns from the “open range” woods1. Chestnut fed hogs produced the “sweetest meat” but rendered the fat into dark oil instead of a white lard. Some acorns produced bitter tasting pork and altered fat consistency. The Iberian hog model offers credence to niche-market application whereby swine purposely harvest tree mast (acorns) to provide a unique flavor and fat profile of the meat.

Iberian hams are produced from Iberian hogs in the southwest region of Spain. At twelve months of age, the hogs graze acorns (2 hogs/hectare) and native grasses exclusively for 2-3 months or until they reach approximately 160 kg. The unique aromatic flavors of the Iberian hams are attributed to the variety of volatile compounds and high levels of amino acids, produced from both the feeding regime of the hogs and aging process of the hams2. Iberian hogs are given free range of the oak and pasture lands during which time, the considerable quantity of mast and late season grasses available in the fall enable the hogs to put on 40% of their slaughter weight, all of it high in monounsaturated fats (62%) and anti-oxidants3. When the Iberian hog is raised in this system, the locals call it the “four legged olive tree”, as the pork fat contains high levels of oleic acid which is found in olive oil. The resultant meat, full of intramuscular fat, produces hams that can age for two or more years without desiccation. Like a long aged wine, these hams develop deep flavor. Fetching prices in excess of 77 Euros per kilo ($35 US per lb.) in fall 2004, the Iberico Ham industry alone is a billion dollar business.

Forest-finished Ossabaw hogs fed peanuts, acorns and alfalfa (PAA) produced pork that was considered more flavorful by food critics and renowned chefs than that of hogs raised on dirt-lots fed a standard corn-soy diet. For health conscious consumers, unsaturated to saturated fatty acid ratios improved (P < 0.01) from 1.6 to 2.6 as a result of feeding the PAA diet4. Consumption of “heart healthy pork” (pork with higher levels of unsaturated fats in the fat profiles) lowered LDL plasma levels in studies with women5. Mast from hardwoods offers the possibility of enhancing pork flavor and fat quality for niche-markets while using a renewable forest resource as a food source.

There are few references regarding managed studies of swine in sylvan ecosystems to improve an existing timber stand or help to establish a more marketable one. According to Dr. Accisioli, of the University of Florence, the main problem with maintaining swine in the woods in the Tuscany region of Italy (similar topography to Appalachia) results from erosion, predominately due to soil compaction and loss of ground cover from continued animal use and inadequate rotation. Because our interest is in enhancing the environment (as well as the meat produced) our animals will remain in their designated wood lots only during periods of peak mast accumulation (3 months) and then subsequently harvested. In order to avoid damage to the ecosystem, reliable estimates of stocking density are required. Carrying capacity of Chinta Senese pigs foraging in turkey-oak forests in the Tuscany region of Italy varied from 0.1 to 2.1 pig/ha, pending annual production of mast. (personal conversations with Dr. Anna Accisioli of the University of Florence). The amount of mast produced in oak, hickory and beech woodlands is not completely synchronized. Within oaks species, seed production is offset with different species having different intermast periods; for some oaks it’s two years, others three. This probably represents the time it takes the trees to accumulate the resources to produce an acorn crop. Other factors influence the timing of acorn crops; drought and ambient temperature during the spring flowering period are most notably the biggest factors (personal conversation with Todd Starlett, Dept of Forestry, University of South Carolina). These effects modify the inherent period of the mast cycle for a species.

Managing hogs in the woods is a dynamic “put and take” system and requires careful monitoring to realize the owner’s forest management plan. Tree growth response or destruction depends on the animal stocking rate and duration of their stay. Preliminary studies with gestating sows in woodlots during the warm season in North Carolina (two six-month seasons, April 1-September 30) suggests potential for improving hardwood timber stands by reducing vegetative competition. Depending on stocking density (12/Ha or 36/Ha) and duration of placement, sows tended to girdle soft woods and leave hard woods (except elms) intact; 70% of the under-story vegetation was removed 6. Similar to “Old Oak Savannas” grass production offers an additional feed source to tree mast as the canopy opens up and more light reaches the forest floor. By moving sows to new sylvan lots (both seasonally and/or when forest management goals are achieved), farmers may gradually improve their existing, non-marketable timber stands with wood products of higher market value.

Alternative diets and genetics to produce unique flavors in meat: The characteristic flavor and aroma of different meats is attributed to the unique combinations of water-soluble and lipid components found in each species7. The concentration of flavor precursors and their deposition into fat reserves are dependent on the particular feed resources and animal species8. Fat also serves as a reservoir for the volatile compounds that accumulate during cooking, and are known to enhance the flavor of meat. Intensive selection pressure over the last 20 years for lean meat composition in hogs, however has resulted in pork with noticeable reductions in intramuscular fat deposits (marbling) and taste. When Tamworth x Duroc hogs were fed barley or corn-soy diets, no differences in the sensory characteristics of the loin chops were observed, presumably due to low amounts of marbling (< 2%) reported in the loins9. In a subsequent experiment, when hogs with higher levels of intramuscular fat (>3.5%) were fed similar diets, barley fed hogs were preferred to corn based diets by sensory evaluation panels10. Additional research is needed to understand the effects of alternative feedstuffs on pork flavor with hogs that have higher levels of intramuscular fat. Farmers who have orchards or woodlots may be able to produce a “Pork of the Season” by gleaning fallen cherries in the spring, peaches in the summer, and apples or acorns in the fall. This preliminary study differs from previous studies reported in that it examines the environmental impact of swine stocking densities on woodlands with an inclined topography.

References cited:
1. Wigginton, Eliot. 1968. The Foxfire Book. Anchor Press, Garden City, NY.
2. Carrapiso, A.I., Jurudo, A. Timon, M.L. and Garcia, C. 2002. Odor-active compounds of Iberian hams with different aroma characteristics. Journal of Agricultural Food Chemistry 50(22):6453-8.
3. Martin, M., Carnicero, M. and Ruiz, J. 1992. The Iberian pig: an animal with a high content of monounsaturated fatty acids. Journal of Hospital Nutrition 7(5):329.
4. Talbott, C., T. See, P. Kaminsky, D. Bixby, M. Sturek, I. L. Brisbin and C. Kadzere. Enhancing Pork Flavor and Fat Quality with Swine Raised in Sylvan Systems: Potential Niche Market Application for the Ossabaw Hog. Renewable Agriculture and Food Systems. (in press).
5. Stewart, J. W., Kaplan, M. L. and Beitz, D. C. 2001. Pork with a high content of polyunsaturated fatty acids lowers LDL cholesterol in women. American Journal of Clinical Nutrition 74:179.
6. Talbott, C. W., Reddy, G. B., Raczkowsky, C., Barrios, T., Matlapudi, M., Coffee, A. and Andrews, J. 2004. Environmental impact of integrating crop and sylvan systems with swine. Journal of Animal Science 82(Suppl. 1):303. (Abstr.).
7. Wasserman, A.E., and Spinelli, A.M. 1972. Effect of some water-soluable
components on aroma of heated adipose tissue. Journal of Agricultural Food Chemistry 20:171.
8. Field, R.A., Williams, J. C. Ferrell, C. L., Crouse, J. D. and Kunsman, J.E.
1978. Dietary alteration of palatability and fatty acids in meat from light
and heavy weight ram lambs. Journal of Animal Science 47:858.
9. Talbott, C., T. See, M. Ahmedna, H. Fennell, G. Gunthorp and P. Willis. 2004. Potential for Small-Scale Farmers to Produce Niche Market Pork Using Alternative Diets, Breeds and Rearing Environments: Observations from North Carolina. Renewable Agriculture and Food Systems. 19(3):135-140.
10. Barrios, T. C. Talbott, T. See and R. Pfortmiller. 2004. Designer pork studies with small-scale farmers targeting niche markets. J. Anim. Sci. 82(Suppl. 1):3. (Abstr.).

Project Objectives:

The goal of this project is to help small-scale/part time farmers find new markets and higher profit margins by raising swine in ways that may enhance the flavor of pork (diet, genetics and management systems) as well as the environment they are raised in.
Objective 1. Determine impact of swine stocking density on sylvan ecology: Sixteen market sized hogs (average age and weight of barrows and gilts at start of trial was 6 months and 224lbs, respectively) were randomly selected from contemporary littermates and placed in a 9.9 acre section (Plot 1; see appendix) in a predominate oak-hickory woodlot in Mason Co. WV. Another group of sixteen hogs (230 lbs average starting wt.) were placed on pastured lots and fed 16% corn-soy (free choice) ration and represented a production and meat sensory control (study below). Hogs were introduced in the woodlots on 9/10/06 (see appendix) and moved to another fenced woodlot (Plot 2: 8.0 acres) on 10/24/06. Animal impact on the forest ecology is based on animals harvesting mast in Plot 1 for the six week period described. Hogs were weighed again on 11/06/06 and harvested two days later at a USDA plant in VA. Woodlot hogs were fed 50 lbs/d (approximately 3 lbs/d per hog) of whole corn and offered a mineral salt block free choice. A 50’x50’ hog panel sacrifice area contained a water trough and facilitated loading the hogs at harvest. Woodland parameters studied included: a) prevalence of rooting, scratching, and marking behavior on timber species, b) effect of swine stocking density on survival, composition and condition of under-story vegetation and ground cover, and c) effect of stocking density on soil compaction and erosion.
Objective 2. Examine potential for woodland mast to improve pork flavor, reduce feed costs and increase revenue from forest ecology. Pork from four hogs consuming mast and four from those consuming the corn-soy diet (control) where tested by a sensory panel at WVU to compare sensory characteristics of chops taken from the two treatments. Fatty acid profiles and pork quality characteristics where conducted by the Food Science Laboratory at WVU. Additionally, an informal pork sensory assessment by farmers, consumers and local chefs where conducted at a Field Day in December 2005 or January 2006.
Objective 3. Demonstrate infrastructure required, animal management strategies utilized, and niche-market applications available to promote small-scale sustainable hog farms in Appalachia. Solicit interest from other small-scale farmers in WV to participate in a more comprehensive project through Workshops and Field Days.

Cooperators

Click linked name(s) to expand
  • Rodney Wallbrown

Research

Materials and methods:

Farm and Woodland Test Plot Profile: I have owned my 267 A tobacco farm in Mason Co. WV since 1978. My farm produced and sold tobacco up until I participated in the USDA tobacco quota buy-out. There are approximately 10 acres of bottom land, 50 acres of pasture and the rest in woodlands. On July 11, 2006, Valerie Hannon and Scott Eggerud used a GPS satellite tracking device to survey an area of woodlands suitable to the stocking rates prescribed (1/A and 2/A). The stocking rate component of the experiment was changed (see below) to one woodlot area for the entire group (16 animals or 1.6 hogs per acre for six weeks then moved for two weeks to plot #2). Plot #1 was approximately 9.9 acres in size and plot #2 was approximately 8.0 acres in size. See attached map for locations. Both areas have similar physical and cultural characteristics therefore the resulting timber stands are almost identical and will be described together.

Both plots, plot #1 to the south and plot #2, adjacent and to the north, have an entirely western aspect with slopes ranging from 15 to 45% with most of the area between 25 to 35% slopes. Gilpin-Upshur silty loams (soil types) dominate the area. Some rock outcrops occur, however most of the area is a long, gentle to moderate slope, with about 220 feet of relief. Elevation ranges from approximately 600 feet in the northwestern corner of Plot #2 to approximately 820 feet along the eastern side of Plot #2.

Most of this area, along with most of the tract, was cleared for agriculture approximately 70 to 80 years ago. This area reverted back to an oak hickory forest type, dominated by white oak, chestnut oak, black oak, scarlet oak, Virginia pine, hickory, red oak and sugar maple (in descending order). In 1992 the sawlog-sized trees for Stand #2 consisted of white oak 24%, Virginia pine 20%, black oak 14%, scarlet oak 8%, chestnut oak 7%, red oak 7% and sugar maple 7%. White ash, yellow poplar, hickory, basswood, red maple and white pine were also tallied as sawlog-sized trees in 1992. Most of the timber 16 inches DBH (diameter breast height) and larger were harvested in 1994.

Once the plot was marked, a two strand electric fence (one strand at 12 inches and another 20 inches) was built using step-in posts and insulators around standing trees. The fence was charged by a solar powered charger and tested (on average) at 4,000 volts. Orange ribbons where tied every 16 ft on the top wire for (animal) visibility.

Impact assessment study areas: In August 2006, prior to hog placement, 15-0.001 ac (milacre, radius = 3.74 ft.) sampling plots were established along three transects (Figure 1.). Transects were 100 ft. apart, and sample plots were 150 ft. apart along each transect. Plots were permanently marked with a plastic stake and the general location was flagged for future identification. Within each plot, all woody seedlings were tallied by species in one of three size classes: 0-12 in. tall (small), 12-48 in. tall (medium), and > 48 in. tall and < 1 in. diameter breast height (large). In addition, percent cover of non-woody vegetation (ferns, forbs, grasses, sedges, and mosses) as well as percent bare ground and exposed rock were estimated and recorded. Percent cover of woody non-tree species such as greenbrier, Japanese honeysuckle, and blueberries were also estimated. Sample plots were re-inventoried on November 17, 2006, following removal of hogs. During this resample, we also estimated percent of plots where soil was disturbed.

Because non-woody species structure was recorded as percent cover and woody species were tallied by abundance (e.g., 1000s/acre), we examined differences between before and after for each vegetation category separately. For non-woody percent cover, several data deletions were performed due to seasonal differences in plant presence. Thus, since the post-sample occurred during the dormant season, we deleted forbs and grasses from the analysis. For woody species, we entered data as total heights (Fei et al. 2003) of each species by sample plot, using the midpoint of the height class (e.g., small = 0.5 ft., medium = 2 ft., and large = 4 ft.). Finally we contrasted plot-level species richness and diversity using a two sample t-test (? = 0.05).

Research results and discussion:

Objective 1, Animal impact on forest ecology: Total woody seedling density before grazing was 26,533/acre (Tables 1 and 2). Eighty-six percent were small seedlings (< 12 in.), and average seedling height was 0.8 ft. The most abundant species was red maple (7,667/acre), 96% of which were in the small size-class. Eastern hophornbeam, flowering dogwood, and white ash were abundant in all size classes. The site was moderately stocked with oak seedlings (1,067/acre), however 89% of these were also in the small size-class; no large size-class oaks were tallied. Following one month of hog grazing, small declines were noted for eastern hophornbeam, flowering dogwood, and white ash, however most of these declines occurred in small size-class stems. Among the oaks, a similar pattern occurred for black and white oak. Much larger declines occurred for red maple, sugar maple, and Virginia pine, almost exclusively in the smallest seedling size-class. We suspect that the bulk of these losses were germinal-sized seedlings either buried in leaf-litter or killed from fall frost. Consistent with the mortality of smallest seedlings, post-grazing average height/seedling increased to 1.4 ft.

Table 1. Average percent cover of non-woody vegetation, before and after hog grazing.
% Before % After
Grass 7.5 1.2
Fern 2.6 1.4
Sedges 0.9 0.7
Forbs 5.7 1.3
Moss 2.5 2.3
Bare/rock 7.5 6.9
Disturbed soil na 3.3
Greenbrier 0.3 0.6
Japanese honeysuckle 0.7 0.4
Blueberries 0.9 0.7

Statistical comparison of of the non-woody percent cover before and after grazing yielded conflicting results. The MRPP test suggested no difference in composition or structure (p = 0.23, see Table 3. in appendix). Conversely, the Mantel test yielded a positive correlation value (r = 0.203), however this was not significant. Neither the MRPP or the Mantel indicated any significant differences in vegetative composition of composition and total height of woody tree seedling species when the two time periods were compared (Table 3.and Table 4.) values were negative, indicating a difference, p values were > 0.05. Similarly, r- and t-values of the Mantel test was positive and statistically significant (p < 0.001). Changes in species richness (S) and species diversity (H’) showed the opposite pattern. Post grazing S and H’ by sample plot were approximately one-half pre-grazing values (S = 7.1 and 3.5, H’ = 1.5 and 0.9, pre- and post-grazing, respectively, p < 0.0001). Eight species that were tallied in the pre-grazing inventory were absent post-grazing. Virtually all of these seedlings were small and the same caveat suggested for the perceived decline of red and sugar maple likely applies here.

Table 2. Abundance of woody seedlings in three size classes, and total height of woody seedlings, before and after hog grazing.
Before (Ave. seedlings/ac) After (Ave. seedlings/ac)
Small Med. Large Total ht (ft.) Small Med. Large Total ht (ft.)
Viburnum spp. 67 0 0 33 0 0 0 0
Am. beech 67 67 0 167 0 67 67 400
Autumn olive 0 0 0 0 0 67 0 133
Black cherry 133 0 0 67 0 0 0 0
Black gum 200 67 367 0 0 0 0
Blue beech 0 0 133 533 0 0 0 0
Buckeye 67 67 67 433 0 67 67 400
Black oak 400 0 67 467 67 133 67 567
Chestnut oak 133 67 0 200 0 0 0 0
White oak 533 0 0 267 333 0 0 167
Crab apple 67 0 0 33 0 0 0 0
E.hophornbeam 3333 667 400 4600 1333 600 800 5067
Elm 1333 0 0 667 0 0 0 0
Fl. Dogwood 2067 200 200 2233 1467 333 200 2200
Hawthorne 333 0 0 167 0 0 0 0
Hickory 933 0 67 733 400 0 67 467
Paw paw 0 67 0 133 0 0 0 0
Red bud 0 0 67 267 0 0 67 267
Red maple 7333 133 200 4733 133 0 333 1400
Serviceberry 533 200 133 1200 333 133 67 700
Sugar maple 1733 0 0 867 0 0 67 267
Virginia pine 867 0 0 433 67 0 0 33
White ash 2667 867 0 3067 2400 200 0 1600
Totals 22800 2333 1400 6733 1600 1867

Estimates of percent soil disturbed averaged 3.3% (SE = 1.33). Eleven of 15 milacre plots showed no soil disturbance, and the range of disturbance was 2-20%. Soil disturbance patterns varied somewhat according to whether the sample plot was intersected by a trail. In the case where disturbance was greatest, the sample plot was adjacent to a persimmon tree, a fruit favored by swine. An additional concern is the relationship between hog rooting, soil disturbance, and spread of invasive species. Japanese honeysuckle (Lonicera japonica) was noted on three pre-grazing sample plots with an average percent cover of 3.7%. Of greater concern is the invasive Japanese stilt grass (Microstegium vimineum). It occurred on three pre-grazing samples; coverage on one sample plot was 45%. This species possesses characteristics typical of

Table 3. Summary statistics for the multiple response permutation procedures (MRPP) for differences in non-woody percent cover, and total height of tree seedlings before and after hog grazing.
Obs. Sorenson Distance Exp. Sorenson Distance T p
Non-woody % cover 0.751 0.756 -0.635 0.229
Tree seedlings 0.782 0.791 -1.234 0.114
The T statistic is the weighted mean within-group distance; the p value indicates the likelihood of achieving the observed difference (T) by chance

Table 4 Summary statistics for the Mantel test for differences in non-woody percent cover, and total height of tree seedlings before and after hog grazing.
r t p
Non-woody % cover 0.203 1.458 0.154
Tree seedlings 0.451 3.600 0.0003
The r-statistic is a measure of association between matrices, and the p value indicates whether or not to reject the null hypothesis of no association between matrices.

many invasive species: it grows quickly, fruits within a single season, produces small but abundant seed that is very persistent in soils, has seeds that adhere to
animal fur or clothing, is nonpalatable, and easily invades disturbed habitats (Tu, no date). It is cause for considerable concern in forested ecosystems in the region, because it is capable of invading forestlands and swiftly replacing natural communities with nearly monospecific stands. Post-gazing sampling occurred after this grass was senescent, so an evaluation of the potential spread of this species was inconclusive (Rentch and McGill, 12/06 in appendix).
The roads constructed during the last timber harvest, most of which ran along natural benches, provided excellent access to the pig plots. No damage to the timber stand, including saplings, pole-sized trees and sawlog-sized trees due to pig foraging was noted (Eggerud, 12/06).

Objective 2. Examine potential for woodland mast to reduce feed costs, improve pork flavor, and increase revenue from forest ecology.
Over the 56 day trial, the hogs consuming corn-soy gained 118 lbs (Table 5) and the pigs on mast gained an average of 8 lbs. Based on a conservative feed conversion rate of 4 lbs of feed to 1 lb of gain, all pigs consumed approximately 1000 lb of feed (costing $100 @ $.1/lb) to reach 225-250 lbs at the start of the trial. During the 8 week trial, the corn-soy pigs each consumed approximately 590 lbs of grain (5 x 118 lbs) vs. approximately 168 lbs (56d x 3) of shelled corn for each of the acorn fed pigs. A 5:1 feed to gain conversion factor is used during the 56 day trial to reflect the higher maintenance requirements (animals > 250 lbs) needed to deposit nearly one inch more subcutaneous back fat and two percent more intramuscular fat in the corn-soy fed hogs. Hogs consuming acorns travelled up and down 25 degree slopes to find their food and as discussed previously, presumably lost 20-30 lbs at the end of the trial.

At $0.10/lb for feed, the average corn soy pig consumed $59 worth of feed during the trial while the mast fed pig consumed $17. Based upon prices paid for by the processor @ $.60/lb live wt. (no live weight penalties) for corn-soy and $1/lb for acorn fed pigs, the income over feed costs (IOFC) for the corn-soy pigs is $49 (348 lbs. x $.6/lb = $208 gross sales less $159 estimated feed costs); the IOFC for acorn fed hogs was $115 for mast fed hogs ( 232 lbs * $1/lb = $232 less $117 estimated feed costs). As the demand for corn increases for ethanol production, the cost of feeding pigs will rise accordingly. In order to hedge against higher grain prices, farmers should consider utilizing alternative feed resources and raising as much on-farm nutrients as possible.

Table 5. Initial and Harvest Weights of Hogs Fed Predominately Acorns or Ad-libitum Corn-Soy Diets for 56 days.
Diet # obs. Days on Trial Start Trial Wt
9/11/06 Harvest Wt Wt Gain
± SE
Acorn*
Corn-soy 16
16 56
56 224
230 232
348 8 ± 3.8
118 ± 6.7
*Hogs consuming mast were offered 3 lbs/day shelled corn

Table 6. Mean back-fat depth and ribeye areas (in2) from Farmer’s Hybrid hogs fed two diets.
Backfat (in.) Ribeye area (in2)
Acorn 0.59 (0.06)a 4.96 (0.64)a
Corn-soy 1.56 (0.46)b 7.04 (1.87)b
P value 0.0045 0.0505
ab Means with different superscripts are different (P < .05)

Table 7. Mean % moisture, crude fat, ash and crude protein in 10th rib chops from Farmer’s Hybrid hogs fed two diets.
Moisture
(%) Crude fat
(%) Ash
(%) Crude protein (%)
Acorn lean 70.31 (1.35) 4.07 (0.88)a 1.07 (0.05) 24.38 (1.22)
Corn-soy lean 70.07 (1.14) 5.92 (1.18)b 1.05 (0.06) 23.51 (1.12)
P value 0.6143 0.0001 0.2822 0.0687
Acorn fat 15.45 (2.25)a 80.00 (3.55)a — —
Corn-soy fat 8.64 (2.00)b 89.98 (2.48)b — —
P value 0.0004 0.0002 — —
ab Means with different superscripts are different (P < .05)

Scores from the informal taste tests at the Lowe Hotel (n=16) and the Farmer Workshop (n=25) are in Table 8. Loin pork roasts served at the Lowe Hotel Dinner from barley-acorn fed hogs had more flavor (SE = 3; P<.1) than pork from corn-soy fed diets. Participants observed no differences in tenderness, juiciness or overall evaluation (rank) in the two loins for the two diet treatments. On the contrary, pork from corn-soy fed hogs was considered more flavorful, juicy and tender than acorn fed hogs at the BOHF farmer workshop. Differences in weights and fat deposition (Tables 5 and 6) of the shoulders due to diet may have influenced the taste test at BOHF. The shoulder from the acorn diet was 7 lbs lighter (corn-soy animals were 100lbs heavier at harvest) than the corn-soy shoulder and may have dried out due to excessive cooking. Unsaturated fats have lower melting points and possibly require different cooking strategies to retain moisture and flavor.

Table 8. Informal Evaluation of Pork Flavor at Lowe Hotel on 12/7/06 and Black Oak Holl’r Farm (BOHF) on 12/8/07 and WVU Meat Science Lab on 12/5/07.
Observation:
Diet # obs. Tender-ness Juiciness Flavor Overall
Lowe Hotel: (Loins)
Barley-acorn
Corn-soy

BOHF (Shoulders)
Barley-acorn
Acorn
Corn-soy
16
16

25
25
25
7.8
7.8

6.4 b
6.5 b
7.3 a
7.4
7.6

7.3 a
6.1 b
7.6 a
8.1a
7.6b

7.0 a
6.4 b
7.5 a
7.6
7.5

7.3
7.2
7.6
Different superscripts indicate differences (P<.1)

Six untrained faculty volunteers from the School of Agriculture agreed to sample 1 cm3 samples of pork loin (cooked to internal temperatures of 165? F) from the BOHF corn-soy and barley diets. There were no statistical differences in sensory characteristics reported between diet treatments (see figure below). Participants at WVU lab tended to favor acorn-fed pork flavor over corn-soy pork and characterized the acorn pork flavor as nutty, sweet and earthy. Minolta scores are listed in Table 9. Acorn pork was darker (P<.01) than corn-soy pork and may reflect the higher intramuscular fat stores in the corn-soy pork (Table 7.). With the considerable exertion in finding mast, different muscle fiber types (red vs. white) could explain the darker muscle color in the acorn-fed hogs (not analyzed).

Table 9. Mean Minolta scores (SE) of 10th rib loin chops from Farmer’s Hybrid hogs fed two diets.
L* value a* value b* value
Acorn 55.71 (3.53)a 8.60 (2.42) 9.65 (1.58)
Corn-soy 60.60 (3.81)b 9.48 (1.20) 9.29 (1.59)
P value 0.0098 0.3129 0.6206
ab Means with different superscripts are different (P < .05)

Table 10. Mean cooking yield and shear force from Farmer’s Hybrid hogs fed two diets.
Cook yield (%) Shear force (g)
Acorn 70.08 (1.86) 3504.06 (741.31)
Corn-soy 72.01 (3.56) 3640.36 (742.74)
P value 0.4276 0.4778
ab Means with different superscripts are different (P < .05)

No differences were observed in cooking yield or shear force (degree of muscle fiber strength and meat tenderness) due to diet (Table 10).

Table 11. Mean fatty-acid profiles from Farmer’s Hybrid hogs fed corn-soy vs. acorn diets.
Fatty acid 2 Corn-soy Acorn
P value Corn-soy Acorn
P value
Subcutaneous fat Lean muscle
C14:0 1.43 (0.15)b1 1.25 (0.12)a 0.0311 1.51 (0.14)b 1.40 (0.08)a 0.0463
C16:0 24.89 (1.04)b 23.37 (0.62)a 0.0001 27.35 (0.65)b 26.05 (0.56)a 0.0017
C16:1 2.39 (0.43) 2.46 (0.29) 0.7284 4.36 (0.52) 4.07 (0.42 0.2517
C18:0 15.22 (1.96) 13.34 (1.02) 0.0686 12.16 (1.35) 12.06 (0.89) 0.8855
C18:1n9c 44.22 (0.88) 44.36 (1.51) 0.7076 44.50 (1.51) 43.78 (0.63) 0.2431
C18:2n6c 10.19 (1.03)a 12.86 (0.72)b 0.0002 7.54 (1.49)a 9.18 (0.47)b 0.0114
C20:1 0.75 (0.44) 1.01 (0.04) 0.1025 0.55 (0.20) 0.52 (0.20) 0.7654
C18:3n3 0.36 (0.20) 0.49 (0.04) 0.0913 0.12 (0.11) 0.23 (0.12) 0.1338
C20:2 0.37 (0.27)a 0.68 (0.13)b 0.0068 0.22 (0.17) 0.14 (0.20)3 0.3010
C20:4n6 0.17 (0.03) 0.18 (0.04) 0.3815 1.70 (0.42)a 2.58 (0.55)b 0.0113
1 Mean % relative peak area (standard deviation)
2 C14:0 = myristic acid; C16:0 = palmitic acid; C16:1 = palmitoleic acid; C18:0 = stearic acid; C18:1n9c = oleic acid; C18:2n6c = linoleic acid; C20:1 = cis-11-eicosenoic acid; C18:3n3 = linolenic acid; C20:2 = cis-11,14-eicosadienoic acid; C20:4n6 = arachidonic acid.
3 Standard deviation due to non-measurable amounts in several samples.

Objective 3: Demonstrate infrastructure required, animal management strategies utilized, and niche-market applications available to promote small-scale sustainable hog farms in Appalachia. (see section 10. Outreach)

Participation Summary
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.