Final Report for OS03-015
Open pasture produced more forage annually (17,200 kg/ha in 2005) than any of the two silvopastoral systems studied. Similarly, the double tree-row silvopasture outperformed scattered tree silvopasture by 2,000 kg/ha dry matter yield, totaling 13,650 kg/ha in 2005. Adding ryegrass and crimson clover to forage plots significantly increased annual forage production for each added species. However, adding a second clover species did not produce further yield increases. Seasonal weighed forage quality indices were not affected by either tree canopy treatments, nor by forage species/varieties employed. Average tree volume was greater in the silvopastures than conventionally 5th row thinned pine plantation.
Silvopasture is an intentional integration of trees, improved forages and livestock into one management system (Sharrow 1999, Sharrow and Ismail 2004). Formerly known as tree-pasture or pine-pasture, silvopasture is one of several agroforestry practices steadily gaining popularity in the southeastern U.S. (Bendfeldt et al. 2001, Grado et al. 2001, Husak and Grado 2002, Workman et al. 2003). Well-managed silvopasture can improve cash flow of small family farm operations (e.g. Dangerfield and Harwell 1990) and provide many environmental benefits (Nowak and Blount 2002). Silvopastures are usually established by planting trees in existing pastures with enough open space left between the rows of trees to allow for forage production (Grado et al. 2001, Husak and Grado 2002). However, these tree-forage-livestock systems can also be established after commercial thinnings of pine plantations (Clason 1995, Clason 1999). Abundance of mid-rotation age pine plantations in the South creates ample opportunity for such conversions to take place.
We tested suitability of several forage combinations for silvopasture establishment after pine plantation thinning. Of particular interests were varieties of bahiagrass, ryegrass, crimson and red clovers. Tested forage varieties were grown in two different silvopastures under different tree canopy conditions created by thinning of 18-year old loblolly pine (Pinus taeda L) stands. Forage yield and quality results obtained in silvopastures were compared with performance of the same forage varieties grown in open pasture. The goal of this study was to identify a forage combination yielding the highest amount and quality of animal feed under thinned 18-year-old loblolly pine canopies in North Florida. In addition, we were interested in tree growth and crown characteristics in two silvopastoral systems created by thinning compared with the same tree characteristics in conventionally 5th row thinned pine plantation. The results of this study are applicable in those parts of the Southeast where soil and climatic conditions are similar to these in North Florida.
- Evaluate effects of tree canopy configuration and forage species and variety composition on monthly, seasonal and yearly forage yield and quality obtained in two silvopastoral systems with those obtained for the same forage species and variety combinations in open pasture.
Compare tree growth and tree crown characteristics in two silvopastoral systems with those obtained for conventionally 5th row thinned pine plantation.
The project was conducted on a private farm in Gadsden County located in panhandle region of north Florida. This area is located at latitude 30.55 degrees North and longitude 84.60 degrees West, with an average elevation of 61 m (200 ft) above see level. The climate is temperate with moderate winters and hot, humid summers. Annual rainfall averages 1676 mm (66 in), and the average daily minimum and maximum air temperatures are 12.8 and 26.1 oC (55 and 79 oF), with an average temperature of 15.6 oC (60 oF). The soils in the study site area are loamy fine sands (fine-loamy, kaolinitic, thermic Typic Kandiudults).
In July 2002, an 18-year-old loblolly pine plantation was thinned to two silvopastoral tree spacings before forage establishment, while control was conventionally 5th row thinned. The two silvopastoral tree configurations tested were: (1) heavy conventional 4th row thinning, and (2) double tree row with forage alleys. In double row configuration, trees were spaced 3.0 x 3.7 m (10 x 12 ft) and were separated by 15.2 m (50 ft) wide forage alley between each pair of double-row tree sets. The double tree row configuration was “an after thinning” version of the popular in Florida 1.2 x 2.4 x 12.2 m (4 x 8 x 40 ft) tree spacing used for silvopasture establishment ever since its inception (Lewis et al. 1985). Both silvopastoral treatments were thinned to approximately 200 trees per hectare (80 trees per acre, or TPA) and 6.9 square meter (30 square feet) of basal area (BA), whereas the control was thinned to approximately 400 trees per hectare (160 TPA) and 13.8 square meter or 60 BA. The response variables were tree height, diameter at breast height (DBH), height to first live branch, crown width in two perpendicular directions (N-S and E-W), crown length, and tree and stand wood volume.
Forage Treatment Combinations
Argentine bahiagrass (Paspalum notatum Flugge), Jumbo ryegrass (Lolium multiflorum Lam.), Dixie crimson clover (Trifolium incarnatum L.), and Cherokee red clover (Trifolium pretense L.) were identified for this project based on their relative shade and drought tolerance. These forage varieties showed good performance under varied photoperiod in previous research (Blount, personal communication). Forage combinations were assigned at random to four replicated 18.3 x 20.1 m (60 x 66 ft) plots within each silvopasture and a nearby open pasture. The following combinations were used: (1) bahiagrass alone, (2) bahiagrass plus ryegrass, (3) bahiagrass plus ryegrass plus crimson clover, (4) bahiagrass plus ryegrass plus crimson clover plus red clover. Bahiagrass was seeded in all forage plots at 22.4 kg/ha (20 lbs/acre) in July 2003, and winter forages were overseeded in November 2003. Seeding rate was 15.7 kg/ha (14 lbs/acre) for ryegrass and red clover, and 16.8 kg/ha (15 lbs/acre) for crimson clover. The appropriate Rhizobium trifolii inoculum was applied to crimson clover seeds immediately prior to seeding. The red clover seeds were inoculated by seed distributor. The forages were established on open pasture (control), in the 15 m forage alleys in the double tree row silvopasture, and among the trees in the 4th row heavily thinned silvopasture. Before forage establishment, site preparation in both silvopastures included prescribed fire to remove logging debris, and herbicide (glyphosate) to control herbaceous weeds. Residual sweetgum and cherry trees were cut down and the stumps treated with herbicide (triclopyr) to prevent sprouting. All forage containing treatments were disk-harrowed prior to seeding bahiagrass. Bahiagrass was mowed and plots cultipacked after overseeding winter forages. The response variables were forage yield per hectare (acre) and forage quality (in vitro organic matter digestibility, crude protein and phosphorus content).
Forage sampling and analyses
During 2004 and 2005, herbage mass and nutritive value of each forage combination were measured on four replicated plots 18.3 x 20.1 m (60 x 66 ft) under two tree canopy conditions and in open pasture. During warm and cool growing seasons, respective forage production was sampled every four to five weeks. Two samples per forage plot were collected, each from within an exclusion cage (at the beginning of each growing season, two such enclosures were randomly placed on each plot). Exclusion cages protected sampling spots from wildlife browsing. They were constructed of wire mesh (2.5 x 5 cm) and shaped as cylindrical enclosures 1 m in diameter and 1.2 m tall. On sampling date, forage from an area of 0.25 square meters in the center of each exclusion cage was clipped off and taken to the laboratory for further processing. Subsequently, forage samples were dried at 60oC to constant mass and weighed. Dried samples were ground to pass a 1-mm screen, and analyzed for crude protein (CP) and phosphorus concentrations, and in vitro organic matter digestibility (IVOMD). To accomplish that, first samples were digested for N determination using a modification of the aluminum block digestion procedure of Gallaher et al., (1975). This method is a modification of the standard Kjeldahl procedure yielding total N (organic plus inorganic fractions). The N in the digest was determined by semi automated colorimetry (Hambleton, 1977). The CP was calculated as N x 6.25. The IVOMD was determined using the two-stage procedure of Tilley and Terry (1963) as modified by Moore and Mott (1974). To evaluate seasonal distribution of yield, month to month harvests were plotted. Seasonal and yearly forage production were calculated by summing across monthly harvests. Seasonal weighed average forage quality indices were computed by summing across winter or summer growing season products of multiplications of sample-determined index by monthly forage dry matter yield, and dividing the summation result by a total seasonal forage dry matter yield harvested. For IVOMD, the percent of forage organic matter determined for each monthly harvest had to be factored in before the seasonal weighed index could be computed.
Fertilization treatments were based on soil testing results and forage needs. In 2004, bahiagrass plots in the open pasture received (rates in kg of element per hectare, in parentheses in lbs of element per acre) 168.3 (150) N and 129.0 (115) K; bahiagrass plots in the silvopastures 168.3 (150) N, 44.9 (40) P and 163.8 (146) K; and the conventionally 5th row thinned pine plantation plots 168.3 (150) N, 44.9 (40) P and 167.2 (149) K. Winter forage plots in the open pasture received 112.2 (100) N and 65.1 (58) K; and winter forage plots in silvopastures 112.2 (100) N, 44.9 (40) P and 154.8 (138) K. These varied fertilization regimes were meant to equalize the soil nutritional status from the standpoint of forage growth. After that all bahiagrass and winter forages, and thinned pine plantation plots received 168.3 (150) N, 39.3 (35) K, 11.2 (10) Mg and 22.4 (20) S in the following fashion. All winter forage plots received 56.1 (50) N on December 28, 2004 and 56.1 (50) N on March 14, 2005, the remainder consisting of 56.1 (50) N, 39.3 (35) K, 11.2 (10) Mg and 22.4 (20) S was applied on July 1, 2005. All bahiagrass forage plots and thinned pine plantation plots received 112.2 (100) N on April 21, 2005 and the remainder consisting of 56.1 (50) N plus 39.3 (35) K, 11.2 (10) Mg and 22.4 (20) S on July 1, 2005. The following mineral fertilizers alone or in appropriate mixtures with each other were used to achieve desired amounts of elemental fertilizer applied on each date: ammonium nitrate (NH4NO3), ammonium sulfate [(NH4)2SO4], urea [CO(NH2)2], triple superphosphate (TSP), muriate of potash (KCl), K-Mag [K2Mg2(SO4)].
The experiment was analyzed as a modified split plot design by the general linear procedures of SAS statistical package (SAS Institute, Inc., Cary, NC). Four tree canopy plots within each thinning treatment and within the open pasture treatment served as the main plots. Forage species combination sub-plots were assigned using a random procedure within each main plot. Tree growth parameters were analyzed by analysis of covariance to account for tree growth differences before the onset of the experiment.
As expected we found that open pasture produced more total herbage mass (14,900 kg/ha in 2004) and (17,200 kg/ha in 2005) than any of the two silvopastoral systems studied. Similarly, double tree-row silvopasture outperformed the scattered tree silvopasture in both years by about 2,000 kg/ha dry matter yield. The total forage production in the double tree-row silvopasture reached 13,650 kg/ha in 2005. Also, on a seasonal basis, open pasture produced more winter and summer forage mass than any of the two silvopastoral systems in both years. In winter 2004, silvopastures did not differ in the amount of forage produced, but in the winter 2005, the double tree-row silvopasture (6,800 kg/ha) outperformed the scattered tree silvopasture (5,200 kg/ha). In the summer forage production, double tree-row silvopasture produced significantly more forage (8,200 kg/ha) than the scattered tree silvopasture (6,900 kg/ha) in 2004, but in 2005 the differences in forage production (9,400 vs. 8,700 kg/ha, respectively) were statistically insignificant. From 2004 to 2005, open pasture forage production rose from 6,500 kg/ha to 8,300 kg/ha for winter forages, and from 10,000 kg/ha to 12,000 kg/ha for summer forages.
Forage species combination had a significant effect on the amount of herbage mass produced during winter 2005, and both summer and total production during 2004 and 2005. In 2005, winter forage production reached 7,100 kg/ha in ryegrass plus crimson clover plus red clover species combination plots, and was only insignificantly lower when crimson clover was the only leguminous species present in the system. During both summers, bahiagrass only plots (no winter forages planted), produced more herbage mass than plots containing ryegrass and one or two clovers in the cool seasons. Plots containing clovers in winter did not differ from each other in bahiagrass production, and yielded more bahiagrass than plots containing ryegrass as the only cool season forage. However, total forage production, both in 2004 and 2005, was the lowest in bahiagrass only plots, higher in plots containing ryegrass, and the highest in plots containing clovers. There were no significant differences in total yearly forage dry mass produced due to the number of clover species planted, i.e., plots with crimson clover only, and plots containing crimson clover plus red clover yielded similar amounts of total forage dry mass. Total forage production peaked at 13,800 kg/ha in 2004 and 16,100 kg/ha in 2005 for plots containing all four forage species.
Monthly harvests from February until May, revealed that winter forage production was increasing gradually in all production systems toward the end of the cool growing season, both in 2004 and 2005. The open pasture outperformed the silvopastoral systems every month in both years, except for May 2005, when the double tree-row silvopasture produced almost as much forage (2,600 kg/ha) as the open pasture (2,800 kg/ha), with the difference between the two being statistically insignificant. The differences in monthly winter forage production between the two silvopastoral systems were all insignificant for every harvest reported here, except for May 2005 when the scattered tree silvopasture produced less forage (2010 kg/ha) than the double tree-row silvopasture (2600 kg/ha). Also winter forage species composition did not have a significant effect on forage yield produced until May 2005 when plots containing one (2500 kg/ha) or both (2,700 kg/ha) clover species outperformed the plots with ryegrass planted as the only cool season forage.
Monthly bahiagrass forage production during warm growing seasons of 2004 and 2005 was consistently higher in the open pasture than the silvopastoral systems. Double tree-row silvopasture outperformed scattered tree silvopasture in the summer of 2004 and in the last sampling of October 2005. Plots where only bahiagrass was planted produced more dry matter yield than plots with clovers during the first harvest of each warm season, and more dry matter yield than ryegrass containing plots during the first two harvests of both 2004 and 2005 and the last harvest of 2005 warm growing season. These results suggest that presence of clovers during cool growing season interfered with bahiagrass production during warm growing seasons less than presence of ryegrass.
For winter forages, monthly in vitro organic matter digestibility (IVOMD) was the highest at the beginning of each cool growing season during both 2004 and 2005. The monthly values of IVOMD for open pasture were significantly higher than those for the silvopastoral systems in 2004 alone. During that year, there were no differences in monthly IVOMD values due to forage species composition. However, in 2005, the highest IVOMD values were recorded for ryegrass plots in February through April harvests, but for May harvest, the plots containing one or two clover species had significantly higher IVOMD values than those containing ryegrass only. In 2004, average weighed IVOMD across all treatments and sampling times was 71.7%, while in 2005 the average IVOMD was 80.0%.
Crude protein content (CP) for winter forages grown in the open pasture was higher than in the silvopastures in April 2004 and February through March 2005. Later in the 2005 cool growing season for April and May samplings, forages in the double tree-row silvopasture had the highest CP content among the three systems under study. However the late season CP values (16.2%) were only about half of the peak values (29.2%) reached in February 2005. The CP values where higher in clover than ryegrass containing plots only for March 2004 and May 2005 samplings. In 2004, average weighed CP across all treatments and sampling times was 17.4%, while in 2005 the average CP was 22.1%.
Percent phosphorus (P) in winter forages was variable among the sampling times and among the tree canopy treatments in 2004 but not in 2005. In the first year, the highest P value reached 0.44% for scattered tree silvopasture in March 2004, while in the second year, the highest value of 0.47 was recorded for open pasture in April 2005. Winter forage P content did not differ due to forage species composition in neither 2004 nor 2005. Across all treatments and sampling times average weighed P was 0.31% in 2004 and 0.38% in 2005.
Summer forage percent in vitro organic matter digestibility (IVOMD) was higher in open pasture than in one or both silvopastoral systems in July and September 2004, and then again in June and October 2005. The same was true for plots containing all four forage species for July 2004, and again for June and October 2005 samplings. Summer (bahiagrass) forage IVOMD was only about 62% of that for winter forages. Across all treatments and sampling times average weighed summer forage IVOMD was 44.6% in 2004 and 49.3% in 2005.
Crude protein (CP) content of summer forages varied little among the thinning treatments both in 2004 and 2005. Addition of one or two clovers to forage mixes in winter increased CP in summer forage (bahiagrass) significantly for July 2004 harvest. Presence of two clover species had the same effect on bahiagrass CP from June to October 2005. Across all treatments and sampling times average weighed summer forage CP was 11.4% in 2004 and 11.5% in 2005. The summer CP contents constitute on average 66% and 52% of CP measured in winter forages in 2004 and 2005, respectively.
Phosphorus (P) content in summer forage was higher in open pasture than other tree canopy treatments between July and October 2004 and again in June 2005. Increasing the number of forage species tended to increase summer forage P content during entire 2004 warm growing season, and then again for the first two harvests of 2005 warm growing season. Across all treatments and sampling times average weighed summer forage P was 0.27% in 2004 and 0.26% in 2005. These values represent 87% and 68% of the winter forage P contents in 2004 and 2005, respectively.
Tree and stand growth
Overall tree and stand growth characteristics measured in the two silvopastures and the conventionally 5th row thinned pine plantation conformed to the expectations. Average tree height in the conventionally 5th row thinned pine plantation was larger than in the silvopastures, both in 2004 and 2005. The opposite was true with regard to DBH, i.e., an average tree in the conventionally 5th row thinned pine plantation was thinner than those in the silvopastures throughout the study. Similarly, as expected, average height to the first live branch was less in the silvopastures than in the conventionally 5th row thinned pine plantation in the same time period. Consequently, average crown length was the largest in the double tree-row silvopasture, intermediate in the scattered tree silvopasture and the smallest in the conventionally 5th row thinned pine plantation for the duration of the study. Also average crown widths were larger in the silvopastures than in the conventionally 5th row thinned pine plantation in 2003 and at the end of the study in 2005. During the time of the experiment, both silvopastures did not differ with respect to average tree height, DBH, height to first live branch, and average crown widths. As a consequence of the tree dimensions, an average tree volume was higher in the silvopastures (and not different between the two) than in the conventionally 5th row thinned pine plantation between 2003 and 2005. However, wood volume per unit of land area was also higher in the silvopastures than conventionally 5th row thinned pine plantation during the same time period. This was likely an effect of higher tree mortality in the conventionally 5th row thinned pine plantation than in the silvopastures after a prescribed fire applied to all systems in April 2003. During the experiment between 2003 and 2005 the overall, across thinning treatments average tree growth variables changed as follows: height from 17.95 m to 18.80 m, DBH from 23.3 cm to 26.4 cm, height to first live branch from 9.67 m to 10.39 m, crown length from 8.27 m to 8.41 m, crown width from 3.52 m to 4.19 m, tree volume from 0.26 to 0.35 cubic meter, and wood volume per unit land area from 56.98 to 76.60 cubic meter per hectare.
The results of this study indicate that open pasture produced more forage dry matter yield than any of the two silvopastoral systems studied, and that the double tree-row silvopasture with 15 m wide forage alleys outperformed the scattered tree silvopasture with regard to forage yield. Therefore, if one is interested only in forage production, then open pasture should be the production system of choice. If one desires to establish silvopasture for reasons other than the highest forage yield possible, then double tree-row silvopasture is a better choice than scattered tree silvopasture because of intermediate forage yields and ease of management operations. More bahiagrass was produced during warm seasons on plots without winter forages both in 2004 and 2005. However, reduction in bahiagrass production was more than compensated by presence of ryegrass alone during cool growing seasons, increasing significantly the total forage mass produced. Adding crimson clover to the forage species mix significantly increased total yearly forage mass produced even further. However, adding second clover species (red clover) had no significant effect on total annual forage dry matter yield in neither 2004 nor 2005.
Despite occasionally significant monthly differences in IVOMD, CP, or P content due to either tree canopy thinning treatment or forage species composition, the weighed seasonal forage quality indices differed very little among the open pasture and silvopastures, as well as among plots with varying forage species composition, for either the winter or summer forages in both 2004 and 2005. This suggests that forage quality was not substantially affected by the employed experimental treatments.
Tree growth was affected by silvopastoral systems as expected, in that on average the trees were shorter, thicker, with longer and wider crowns, and shorter branch free logs than in the conventionally 5th row thinned pine plantation. Although an average tree volume was higher in the silvopastures than conventionally 5th row thinned pine plantation, quality of lumber produced from trees grown in silvopastures may be adversely affected by more and larger knots and shorter logs compared with conventionally managed pine plantations.
Educational & Outreach Activities
Bambo, S. (Ph.D. dissertation in progress). Loblolly pine growth and forage performance under thinned tree canopies in North Florida. School of Forest Resources and Conservation, University of Florida.
Bambo, S., Nowak, J.; Blount, A.; Osiecka, A. and Myer, R. 2004. Loblolly pine growth and warm/cool-season forage performance under thinned tree canopies in North Florida. Abstracts. 1st World Congress of Agroforestry, Orlando, FL, June 27-July 2, 2004.
Bambo, S.; Nowak, J.; Blount, A.; Osiecka, A. and Myer, R. 2004. Loblolly pine growth and forage performance under thinned tree canopies in North Florida. Poster, 1st World Congress of Agroforestry, University of Florida, IFAS, Orlando, FL, June 27 – July 2, 2004.
Nowak, J.; Bambo, S.; Blount, A.; Osiecka, A. and Myer, R. 2004. From a mid-rotation loblolly pine plantation to silvopasture – a North Florida case study. Pre-conference Workshop Presentation, 1st World Congress of Agroforestry, University of Florida, IFAS, Orlando, FL, June 27 – July 2, 2004. (Participants: ~30 conference participants from around the world)
Nowak, J., Bambo, S., Blount, A., Myer, R. and Mackowiak, C. 2004. Performance of various forage combinations under pine canopies in North Florida – Field Presentation, In-Service Training for County Agents: Center for Subtropical Agroforestry, Educator Curriculum Guide, University of Florida, IFAS, Cooperative Extension Service, School of Forest Resources and Conservation and North Florida Research and Education Center, Quincy, FL, 17-18 March 2004. (Participants: 10 FL county extension agents and 4 other natural resource professionals)
Nowak, J., Blount, A. and Myer, R. 2003. Planting and thinning pines for grass production at the Providence Farm. Field presentation, pilot in-service training, University of Florida, IFAS, Cooperative Extension Service, School of Forest Resources and Conservation and North Florida Research and Education Center, Quincy, FL, 11 December 2003. (Participants: 2 FL county extension agents, 6 other natural resource professionals, and 2 Florida landowners)
Summary of outreach activities
We had two farmers from Gadsden County, Florida participate in the design and conduct of the study. The same individuals were active participants of the field demonstrations conducted as part of this project on 11 December 2003 and again on 17-18 March 2004. The field presentations were attended by total or 24 individuals from the southeastern U.S. We also presented the initial results of this study at the 1st World Congress of Agroforestry in Orlando, FL, between 27 June and 2 July 2004, both as a poster and a pre-conference workshop. It is hard to know how many from about 500 Conference participants viewed the poster. However, about 30 persons attended our pre-conference presentation on the same subject, roughly half of them from the southeastern U.S.
To date, the project had limited impact on Florida landowners and county extension agents who are still learning about the project results and their applicability to farm operations and/or extension clientele. It is expected that the project will have greater impact in the future when cattle is incorporated into the silvopastoral systems created in the course of this project.
To date, we are aware of two North Florida farmers who are thinking about emulating our project. Both farmers plan to create silvopastoral systems after thinning mid-rotation pine plantations and planting forages under the thinned pine canopies. We probably reached another 20 to 30 farmers through conference poster and workshop presentations. Farmers planning conversion of mid-rotation pine plantations to silvopastures after first commercial thinning should consider grinding tree stumps after tree removal. We found that even stumps cut close to the ground, rot too slowly and were impeding mechanized management operations up to three years after thinning. Also, prompt introduction of livestock grazing in an established silvopasture may reduce a need for forage harvest and weed control. Unless small dimension harvesting equipment is available, haying in silvopastures might prove cumbersome.
Areas needing additional study
Research demonstrating livestock integration and management in silvopastoral systems like ours is needed. Successful and profitable livestock integration and management in silvopastures would also help popularize this agroforestry practice among farmers and landowners.
There is a need to identify or design more specialized equipment for soil cultivation, fertilizer application and forage harvesting in silvopastoral systems. We have used standard farm equipment, which was cumbersome to operate among the trees. We also relied to large degree on graduate student and OPS labor, and used hand tools wherever we could not use standard farm equipment. We found silvopasture very labor and management intensive which might impede wide adoption by farmers. Therefore we also suggest that future research should address streamlining of silvopasture management.
Alavalapati, J.R.R.; Shrestha, R.K.; Stainback, G.A. and Matta, J.R. 2004. Agroforestry development: An environmental economic perspective. Agroforestry Systems. 61: 299-310.
Allen, S.C.; Jose, S.; Nair, P.K.R.; Brecke, B.J.; Nkedi-Kizza, P. and Ramsey, C.L. 2004. Safety-net role of tree roots: evidence from a pecan (Carya illinoensis K. Koch) – cotton (Gossypium hirsutum L.) alley cropping system in the southern United States. Forest Ecology and Management. 192: 395-407.
Bendfeldt, E.S.; Feldhake, C.M. and Burger, J.A. 2001. Establishing trees in an Appalachian silvopasture: response to shelters, grass control, mulch, and fertilization. Agroforestry Systems. 53: 291-295.
Clason, T.R. 1995. Economic implications of silvipastures on southern pine plantations. Agroforestry Systems. 29: 227-238.
Clason, T. R. 1999. Silvopastoral practices sustain timber and forage production in commercial loblolly pine plantations of northwest Louisiana, USA. Agroforestry Systems. 44: 293-303.
Dangerfield, C.W. and Harwell, R.L. 1990. An analysis of a silvopastoral system for the marginal land in the southeast United States. Agroforestry Systems. 10: 187-197.
Gallaher, R.N.; Weldon, C.O. and Futral, J.G. 1975. An aluminum block digestor for plant and soil analysis. Soil Science Society of America Proceedings. 39: 803-806.
Grado, S.C.; Hovermale, C.H. and St. Louis D.G. 2001. A financial analysis of silvopasture system in southern Mississippi. Agroforestry Systems, 53: 313-322.
Hambleton, L.G. 1977. Semiautomated method for simultaneous determination of phosphorus, calcium and crude protein in animal feeds. Journal of the Association of Official Analytical Chemists (JAOAC). 60: 845-852
Husak, A.L. and Grado, S.C. 2002. Monetary benefits in a southern silvopastoral system. Southern Journal of Applied Forestry. 26: 159-164.
Lewis, C.E.; Tanner, G.W.; Terry, W.S. 1985. Double vs. single-row pine plantations for wood and forage production. Southern Journal of Applied Forestry. 9(1): 55-61.
Moore, J.E. and Mott, G.O. 1974. Recovery of Residual organic matter from in vitro digestion of forages. Journal of Dairy Science. 57: 1258-1259.
Nowak, J. and Blount, A. 2002. Silvopasture: more than cattle grazing in pine plantations. Forest Landowner. 61: 10-13.
Sharrow, S.H. 1999. Silvopastoralism: competition and facilitation between trees, livestock and improved grass-clover pastures on temperate rainfed lands. In: Agroforestry in Sustainable Agricultural Systems. L.E. Buck, J.P. Lassoie and E.C.M. Fernandes Eds. CRC Press LLC. Boca Raton, FL. 416p.
Sharrow, S.H. and Ismail, S. 2004. Carbon and nitrogen storage in agroforests, tree plantations, and pastures in western Oregon, USA. Agroforestry Systems. 60: 123-130.
Tilley, J.A. and Terry, R.A. 1963. A two-stage technique for the in vitro digestion of forage crops. Journal of British Grassland Society. 18: 104-111.
Workman, S.W.; Bannister, M.E. and Nair, P.K.R. 2003. Agroforestry potential in the southeastern United States: perceptions of landowners and extension professionals. Agroforestry Systems. 59: 73-83.