Regeneration of mahogany and Spanish cedar in gaps created by railroad tie extraction in Quintana Roo, Mexico

Available online at www.sciencedirect.com

Forest Ecology and Management 255 (2008) 308–312 www.elsevier.com/locate/foreco

Regeneration of mahogany and Spanish cedar in gaps created by railroad tie extraction in Quintana Roo, Mexico Patricia Negreros-Castillo a,*, Carl W. Mize b a

Centro de Investigaciones Tropicales, Universidad Veracruzana, Aptdo. Postal 525, 91019 Xalapa, Veracruz, Mexico b Natural Resource Ecology and Management Department, Iowa State University, Ames, IA, USA

Abstract Seeds of mahogany and Spanish cedar were planted in gaps created by harvesting railroad ties at two locations in the Yucatan Peninsula. Gaps averaged 40 m2 in area. Germination after 2 months was very low (2%) for Spanish cedar and moderately low (29%) for mahogany at the one location where it was measured, probably due to predation and fungal damage. Annual seedling mortality was 25–40%, and height growth was moderate to normal, with average height equal to 45 cm after 3 or 4 years, depending upon the location. We conclude that gaps from railroad tie harvesting are not adequate for the establishment of either species from seed. If, however, gaps created by harvesting could be concentrated into at least moderately open areas 400 m2 in area or larger, they may be useful as sites for regenerating both species. # 2007 Published by Elsevier B.V. Keywords: Seedling establishment; Tropical forestry; Tropical silviculture; Swietenia macrophylla; Cedrela odorata

1. Introduction Mahogany (Swietenia macrophylla King) and Spanish cedar (Cedrela odorata L.) are the two most valuable forest tree species in the Yucatan Peninsula, and hundreds of thousands of mahogany seedlings have been planted there during the last 50 years. Unfortunately, seedling survival has been very poor (Negreros-Castillo and Mize, 2003; Javier Chavelas, personal communication), due to a number of factors, such as poor planting stock, poor planting practices, inadequate knowledge of appropriate site conditions, and planting under forest canopies (Cuevas, 1947; Negreros-Castillo and Mize, 2003). Spanish cedar, another valuable species, is not often planted, but problems encountered with it are similar to those for mahogany. In Quintana Roo, most forest harvests focus on mahogany, but some other species are harvested. In 1997, 14,000 m3 of wood were harvested to produce railroad ties (Kierman and Freese, 1998; Shoch, 1999). Recently, harvest of small diameter trees (from 15 to 25 cm in diameter) has increased substantially to meet demand for tourism-related construction (Filiberto and Dionisio Yan, personal communication; Victoria Santos Jime´nez, personal communication).

* Corresponding author. Tel.: +52 228 810 8263; fax: +52 228 810 8263. E-mail address: [email protected] (P. Negreros-Castillo). 0378-1127/$ – see front matter # 2007 Published by Elsevier B.V. doi:10.1016/j.foreco.2007.09.052

Mahogany and Spanish cedar are shade intolerant species (Lamb, 1966), so relatively high light levels are necessary for survival and growth, and mahogany seedlings planted beneath canopies have had poor survival (Negreros-Castillo and Mize, 2003; Negreros-Castillo et al., 2003). Gaps created by harvesting for railroad ties and small diameter trees might create conditions needed for regenerating and growing both species to commercial size (Dupuy and Chazdon, 2006; D’Oliveira, 2000; Bongers et al., 1988). Mahogany has been successfully regenerated from seed and seedlings planted in large gaps, such as 0.5 ha (Negreros-Castillo et al., 2003; Snook and Negreros-Castillo, 2004), but a minimum gap size has not been identified (Dickinson et al., 2000; Whitman et al., 1997). Harvesting smaller trees could double as a tool for regenerating both species if the resultant gaps were larger than the unknown minimum gap size for acceptable growth. This study was done to evaluate survival and growth of mahogany and Spanish cedar seedlings that arose from seed planted in gaps created by harvesting trees for railroad tie production. 2. Material and methods The study was conducted near the center of Quintana Roo on land owned by X-Pichil and Naranjal, two communities that are approximately 20 km apart (Negreros-Castillo et al., 2003).

P. Negreros-Castillo, C.W. Mize / Forest Ecology and Management 255 (2008) 308–312

Central and southern Quintana Roo forests are classified as seasonal tropical forests, the most widespread type of tropical forest in Central America (Murphy and Lugo, 1986). Hurricanes, fires, and agricultural activities have had an important effect on the vegetation for centuries (Go´mezPompa, 1987). The climate of the region is classified as Aw in the Ko¨ppen classification system, as modified by Garcı´a (1973), which is defined as warm and submoist with abundant rains in summer and dry winters. Annual rainfall averages 1200 mm year 1, usually falling between June and October (INEGI, 1994), with a 5–6 month dry season, when rainfall is less than 45 mm month 1. Soils are derived from limestone, and Rendolls (USA Soil Taxonomy; Foth and Schafer, 1980) are the most common type. In Naranjal and X-Pichil railroad ties were harvested in May 1996 and June 1997, respectively. Approximately 2 weeks after harvest, 50 harvest-created gaps at Naranjal and 100 at X-Pichil were located and mapped. After each gap was located in Naranjal, the point on the ground that would receive the most direct sunlight when the sun was at its zenith (12:00) was located and marked, and four mahogany and four Spanish cedar seeds were planted around each marked site. In X-Pichil planting sites were similarly selected, five mahogany seeds were planted around the marked site in each of 50 gaps, and five Spanish cedar seeds were similarly planted in the other 50 gaps. Seeds were buried 2 cm deep and 5 cm apart. Seeds for planting had been collected from designated seed trees 2–3 months before seeding, at the forestry research station San Felipe Bacalar of the National Institute for Research in Forestry, Agriculture and Animal Husbandry (INIFAP) in Quintana Roo, and kept in cloth bags at room temperature until planting. When we established the experiment, we believed that planted seedlings would have higher survival and growth rates than seedlings arising from planted seed. Seeds were used, however, because people harvesting trees can easily carry enough to plant all gaps created in a day, whereas seedlings would be inconvenient to carry and likely to be damaged during transport (Negreros-Castillo and Mize, 2003). We thought that if seedlings from seed would grow fast enough to reach the canopy, low survival rates could be compensated for by planting more seed in each gap. When the seeds were planted in X-Pichil, the following measurements were taken: maximum and minimum gap width as measured from the edges of the canopies surrounding the opening created by harvesting; canopy cover above the gap as measured by a Forestry Suppliers spherical crown densitometer held 70 cm above the ground; and the species and stump diameter of the harvested tree. Gap edges were visually estimated in a manner that could be used by harvest crews. The same measurements were made in Naranjal in February 1997, about 6 months after seeds were planted, but only gaps with established seedlings were measured (28 of the 50). Sites in X-Pichil were visited 2 months after planting. Naranjal sites were not checked because labor was not available. Gaps with live seedlings were measured at X-Pichil in April 1999 and May 2000, and at Naranjal in February 1997, April 1999, and June 2000. Measurements included canopy

309

coverage and the number of live seedlings and their total heights. No measurements were made on gaps lacking live seedlings. Statistical analyses (paired and unpaired t-tests and Pearson correlation analyses) were done using the SAS system (SAS Institute Inc., 1989). 3. Results and discussion At X-Pichil maximum and minimum gap width averaged 12 and 4 m, respectively, with an average area, assuming the gap is an ellipse, of 38 m2 (range 13–99 m2). At Naranjal maximum and minimum widths also averaged 12 and 4 m, respectively, with an average area of 39 m2 (range 8–155 m2). Average stump diameter was approximately 40 cm at both locations, and gap area was not correlated with stump diameter for either location ( p  0.50), partially because additional trees were often cut to facilitate the dropping of harvested trees. In a community about 50 km from both locations, the (estimated) average size for naturally occurring gaps was 55 m2, and 80 m2 for gaps created by harvesting commercial-sized mahogany (Dickinson et al., 2000). In northern Belize in a forest with larger trees, the average gap size for felled mahogany was 583 m2, and 130 m2 for naturally occurring gaps (Whitman et al., 1997). For X-Pichil in 1997, average canopy coverage at the center of each gap was 79% for the 50 mahogany plots (range 48– 98%), and 90% for the 50 Spanish cedar plots (range 28–98%). Canopy coverage of mahogany plots with seedlings increased from an average of 77% in 1997 to 84% in 1999 ( p < 0.001, S.E. of difference = 0.7%, n = 23), and change in canopy coverage was not correlated with most gap characteristics (Table 1). For Naranjal in 1997, average canopy coverage averaged 79% (range 49–94%, n = 28). There canopy coverage of plots with seedlings increased from an average of 71% in 1997 to 83% in 1999 ( p = 0.005, S.E. of difference = 3.4%, n = 11) and change in coverage was strongly related to some gap characteristics (Table 1). From measurements made in 1997, canopy coverage and gap area were negatively correlated for X-Pichil (r = 0.31, p = 0.027, n = 50) and Naranjal (r = 0.45, p = 0.017, n = 28). Graphs of change in canopy coverage versus initial coverage (not included) showed that when canopy coverage was about 85%, there was no net change over time, indicating that the canopy coverage for the undisturbed forest, which was not measured, probably averaged about 85%. It is reasonable to assume that canopy coverage had returned to pre-harvest levels within a few years for the gaps Table 1 Correlation of change in canopy coverage (1997–1999) with initial coverage, initial gap area, and minimum and maximum gap widths Characteristic Initial coverage Initial gap area Minimum gap width Maximum gap width

X-Pichil

Naranjal

0.73 0.12 0.10 0.13

0.96 0.58 0.71 0.10

(<0.001) (0.64) (0.69) (0.62)

(<0.001) (0.061) (0.015) (0.78)

p values in parentheses and based on 18 plots in X-Pichil and 11 plots at Naranjal.

310

P. Negreros-Castillo, C.W. Mize / Forest Ecology and Management 255 (2008) 308–312

Table 2 Number of plots with seedlings, seedling count, and mean height of mahogany and Spanish cedar in X-Pichil and Naranjal from 1997 to 2000 Year

X-Pichil Mahogany

1997 (2 months) 1999 (2 years) 2000 (3 years) Year

Spanish cedar

No. plots with seedlings

Total no. of seedlings

Height (cm)

No. plots with seedlings

Total no. of seedlings

Height (cm)

– 23 19

72 39 32

– 31 (1.8) 45 (4.2)

– 1 0

5 1 0

– – –

Naranjal Mahogany

1997 (6 months) 1999 (3 years) 2000 (4 years)

Spanish cedar

No. plots with seedlings

Total no. of seedlings

Height (cm)

No. plots with seedlings

Total no. of seedlings

Height (cm)

15 7 4

15 7 4

20 (1.3) 41 (11) 45 (16)

23 7 5

39 12 7

12 (0.6) 29 (2.8) 32 (2.8)

Standard error in parentheses. In 1997 at X-Pichil counts of germinated seedlings but no plot data were recorded.

with high post-harvest canopy coverage. Although gap closure seems to have been seldom measured, in a moist tropical forest at Barro Colorado Island, Panama, gaps ranging from 80 to 210 m2 closed in 24–50 months (Yavitt et al., 1995). At the same location, for gaps ranging from 138 to 368 m2, light levels returned to pre-gap levels within 48 months (Fraver et al., 1998). Two months after planting at X-Pichil, five (2%) of the 250 Spanish cedar seed and 72 (29%) of the 250 mahogany seed had germinated. Three years after seeding, there were no Spanish cedar and 32 mahogany seedlings (Table 2). Assuming a constant annual mortality rate, about 25% of mahogany died each year in X-Pichil. At Naranjal by 4 years after planting, there were seven Spanish cedar and four mahogany seedlings (Table 2). Assuming a constant annual mortality rate, about 40% of seedlings of both species died each year in Naranjal. t-Tests were used to compare initial canopy coverage, gap area, and minimum and maximum gap widths between plots with no surviving seedlings and plots with one or more surviving seedlings as measured in 1997, 1999, and 2000. For Naranjal no differences were found for any of the four variables for mahogany and Spanish cedar for 1997, 1999, and 2000 ( p  0.10). For X-Pichil there also were no differences between the two groups of plots as measured in 1999 and 2000 ( p  0.15). Other studies also have found no relation between gap size and survival of mahogany seedlings (Snook et al., 2005a,b). Assuming similar germination rates for both locations, seed loss due to predation by insects and small mammals or fungal attack was higher at X-Pichil, particularly for Spanish cedar. At both locations seed loss might have been higher than normal because seeds were planted relatively close together, making it more likely that multiple seeds would be predated if one was found. Previous studies of direct seeding of mahogany under canopy gaps in Mexico showed little predation (NegrerosCastillo et al., 2005; Negreros-Castillo and Hall, 1996), while in Belize extremely high levels of seed predation were observed (Stevenson, 1927). If seed loss is a major limitation to regeneration via direct seeding under some conditions, it might

be overcome by planting more seeds in each gap and planting them further apart. Mahogany seedlings averaged 45 cm in height after 4 years at Naranjal and 45 cm after 3 years at X-Pichil. Spanish cedar averaged 32 cm tall at Naranjal after 4 years, and there was none alive at X-Pichil (Table 2). By the end of the study, the tallest individual mahoganies were 93 and 103 cm at the respective locations, and the tallest Spanish cedar at Naranjal was 68 cm. The tallest seedlings were found in gaps of about average size. While seedling survival was marginally influenced by gap characteristics, seedling height was more clearly affected at X-Pichil, the site with sufficient seedlings for analysis, particularly by 2000 (Table 3). Average height growth per plot from 1999 to 2000 (Fig. 1) was positively correlated with initial gap area (r = 0.57, p = 0.011, n = 19). Growth rate for mahogany also increased with gap size in a nearby community (Dickinson and Whigham, 1999). We detected no Hypsipyla grandella attack on any seedlings during any measurement times. Incident light affects mahogany growth regardless of gap size (Ramos and del Amo, 1992). As gap sizes and soils were similar at both locations, the greater growth and higher survival at X-Pichil might result from the greater number of gaps per hectare (by a factor of about 3 to 1) at X-Pichil compared to Naranjal. X-Pichil has often been harvested for railroad ties and more recently for tourism construction, producing more gaps than at Naranjal, which has a better stock of high-value timber species such as mahogany, so railroad tie harvesting is infrequent. A higher number of gaps per hectare should result Table 3 Correlation of average seedling height in 2000 by plot with gap area, maximum and minimum gap widths, and canopy coverage and p value (in parentheses) for mahogany in X-Pichil, based on 18 plots Characteristic Gap area Maximum gap width Minimum gap width Canopy coverage

Correlation 0.62 0.63 0.32 0.50

(0.006) (0.005) (0.20) (0.033)

P. Negreros-Castillo, C.W. Mize / Forest Ecology and Management 255 (2008) 308–312

311

number of studies have shown that mahogany grows well under partial shade (Holdridge and Marrero, 1940; Chable, 1967; Ramos and del Amo, 1992). Also, as Mayhew and Newton (1998) summarized from a variety of studies, young mahogany plantings, no matter how they are established, need tending for at least a few years to remove vines and cut back immediate competition. Acknowledgements

Fig. 1. Mahogany height growth (1999–2000) vs. initial gap area in X-Pichil.

in increased light levels in the understory and perhaps reduced moisture competition (Becker et al., 1988). At a site near the two locations of this study, in large openings (0.5 ha) prepared by felling and leaving trees, mahogany grown from buried seed amidst one-year-old vegetation averaged about 45 cm in height after almost 3 years (Negreros-Castillo et al., 2003), similar to our results in X-Pichil with much smaller gaps. Although we found mahogany height growth to increase with increasing gap size, it is that clear most of the gaps we studied will close relatively quickly, and that few, if any, mahogany will reach the forest canopy. 4. Implications for management Given that gaps around 40 m2 in area are not adequate for regenerating mahogany and Spanish cedar, and gaps about 5000 m2 (0.5 ha) are adequate (Negreros-Castillo et al., 2003), what is the smallest area needed for successful regeneration? Growth of mahogany seedlings in Belize increased with gap size in natural gaps from less than 100 m2 to more than 400 m2 (Snook et al., 2005a). In a forest with similar structure, stature, and site quality to the location of our study (Grogan and Galvao, 2006), Grogan et al. (2005, their Table 1) examined mahogany growth in seven gaps ranging from 241 to 824 m2. We found that their maximum height growth in each gap over 80 months increased with gap size and approached an asymptote for gaps >400 m2. Average height growth over at least 2 years showed a similar trend, implying that 400 m2 might be the minimum gap size needed for establishing mahogany in a comparable forest (original data from Grogan et al., 2005, used with permission of senior author). Trees grown from seed planted at the center of a 20  20 m gap may have sufficient time for growth before canopy closure. This needs to be verified with a study of mahogany growing across a range of gap sizes centered on 400 m2. Operationally, forest managers in the Yucatan may be able to concentrate harvests of small diameter and commercial-sized trees into areas that would create sufficiently large openings. Openings would not need 100% canopy removal because a

Thanks to Andres Canul, an extraordinary person from XPichil, for his participation and to Antonio Moo, a student from the local forestry school (Instituto Tecnolo´gico Agropecuario No. 16. Chetumal) who helped in many ways. Thanks to Manuel Ake´ Pat from Naranjal for his help with the experiment and to the communities of X-Pichil and Naranjal for allowing us to conduct this study in their forests. Nick Brokaw, Odilo´n Sa´nchez, Frank Wadsworth, and Sheila Ward, in particular, are warmly thanked for comments. We also thank Jimmy Grogan for allowing us access to his data and for valuable discussions on results and mahogany silviculture. This journal paper of the Iowa Agriculture and Home Economics Experiment Station, Ames, Iowa, Project No. 3705, was supported by Hatch Act and State of Iowa funds. References Becker, P., Rabenold, P.E., Idol, J.R., Smith, A.P., 1988. Water potential gradient for gaps and slopes in a Panamanian tropical moist forest’s dry season. J. Trop. Ecol. 4, 173–184. Bongers, F.J., Pompa, J., Iriarte-Vivar, S., 1988. Response of Cordia megalantha Blake seedlings to gap environments in tropical rain forest. Funct. Ecol. 2, 379–390. Chable, A.C., 1967. Reforestation in the Republic of Honduras, Central America. Ceiba 13, 1–56. Cuevas, L.A., 1947. Explotacio´n de tres especies forestales y propogacio´n artificial de caoba en Quintana Roo. Thesis. National Agriculture School, Chapingo, Me´xico. Dickinson, M.B., Whigham, D.F., 1999. Regeneration of mahogany (Swietenia macrophylla) in the Yucatan. Int. Forest Rev. 1, 35–39. Dickinson, M.B., Whigham, D.F., Herman, S.M., 2000. Tree regeneration in felling and natural treefall disturbances in a semideciduous tropical forest in Mexico. Forest Ecol. Manage. 134, 137–151. D’Oliveira, M.V.N., 2000. Artificial regeneration in gaps and skidding trails after mechanised forest exploitation in Acre, Brazil. Forest Ecol. Manage. 127, 67–76. Dupuy, J.M., Chazdon, R.L., 2006. Effects of vegetation cover on seedling and sapling dynamics in secondary tropical wet forests in Costa Rica. J. Trop. Ecol. 22, 65–76. Foth, H., Schafer, J.W., 1980. Soil Geography and Land Use. John Wiley and Sons, New York. Fraver, S., Brokaw, N.V.L., Smith, A.P., 1998. Delimiting the gap phase in the growth cycle of a Panamanian forest. J. Trop. Ecol. 14, 673–681. Garcı´a, E., 1973. Modificaciones al sistema de clasificacio´n clima´tica de Ko¨ppen (para adaptarlo a las condiciones de la Repu´blica Mexicana). Instituto de Geografı´a, UNAM. Go´mez-Pompa, A., 1987. On Maya Silviculture. Mexican Studies 3(1) Regents of the University of California, Berkeley, 17 pp. Grogan, J., Galvao, J., 2006. Physiographic and floristic gradients across topography in transitional seasonally dry evergreen forests of southeast Para´, Brazil. Acta Amazo´nica 36, 483–496. Grogan, J., Landis, R.M., Ashton, P.M.S., Galvao, J., 2005. Growth response of big-leaf mahogany (Swietenia macrophylla) advance seedling regeneration

312

P. Negreros-Castillo, C.W. Mize / Forest Ecology and Management 255 (2008) 308–312

to overhead canopy release in southeast Para´, Brazil. Forest Ecol. Manage. 204, 399–412. Holdridge, L.R., Marrero, J., 1940. Preliminary notes on the silviculture of the big-leaf mahogany. Caribbean For. 2, 20–23. INEGI, 1994. Anuario estadı´stico del estado de Quintana Roo. Instituto Nacional de Estadı´stica Geografı´a e Informa´tica and the Government of the State of Quintana Roo. Kierman, M.J., Freese, C.H., 1998. Mexico’s plan piloto forestal: the search for balance between socioeconomic and ecological sustainability. In: Freese, C.H. (Ed.), Harvesting Wild Species: Implications for Biodiversity Conservation. Johns Hopkins University Press, Baltimore, pp. 93–131. Lamb, F.B., 1966. Mahogany of Tropical America: Its Ecology and Management. University of Michigan, Ann Arbor. Mayhew, J.E., Newton, A.C., 1998. The Silviculture of Mahogany. CABI Publishing. Murphy, P.G., Lugo, A.E., 1986. Ecology of a tropical dry forest. Ann. Rev. Ecol. Syst. 17, 67–88. Negreros-Castillo, P., Hall, R., 1996. First-year results of partial overstory removal and direct seeding of mahogany (Swietenia macrophylla) in Quintana Roo, Mexico. J. Sust. For. 3, 65–76. Negreros-Castillo, P., Mize, C., 2003. Enrichment planting of Swietenia macrophylla (big leaf mahogany) and Cedrela odorata (Spanish cedar) in Quintana Roo, Mexico. In: Lugo, A.E., Figueroa Colo´n, J.C., Alayo´n, M. (Eds.), Big-Leaf Mahogany: Genetics, Ecology, and Mangement, Ecological Studies, vol. 159. Springer-Verlag, New York, pp. 278–287. Negreros-Castillo, P., Snook, L., Mize, C.W., 2003. Regenerating mahogany (Swietenia macrophylla) from seed in Quintana Roo, Mexico: the effects of sowing method and clearing treatment. Forest Ecol. Manage. 183, 351–362.

Negreros-Castillo, P., Snook, L.K., Mize, C.W., 2005. Regeneracio´n de caoba a partir de siembra directa en aperturas creadas en un bosque natural en Me´xico. Recur. Nat. Amb. (CATIE) 44, 84–90. Ramos, J.M., del Amo, S., 1992. Enrichment planning in a tropical secondary forest in Veracruz, Mexico. Forest Ecol. Manage. 54, 289–304. SAS Institute Inc., 1989. SAS Language and Procedures, Version 6. SAS Institute Inc., Cary, NC. Shoch, D.T., 1999. An ecological and economic evaluation of railroad tie harvest in the ejido X-Pichil, Quintana Roo, Mexico. Master’s Thesis. Nicholas School of Environment, Duke University. Snook, L.K., Negreros-Castillo, P., 2004. Regenerating mahogany (Swietenia macrophylla King) on clearings in Mexico’s Maya Forest: the effects of clearing treatment and cleaning on seedling survival and growth. Forest Ecol. Manage. 189, 143–160. Snook, L.K., Iskandar, H., Chow, J., Cohen, J., O’Connor, J., 2005a. Supervivencia y crecimiento de caoba en claros post-extraccio´n en Belice, a partir de semillas y plantulas. Recur. Nat. Amb. (CATIE) 44, 76–83. Snook, L.K., Negreros-Castillo, P., O’Connor, J., 2005b. Supervivencia y crecimiento de pla´ntulas de caoba en aperturas creadas en la Selva Maya de Belice y Me´xico. Recur. Nat. Amb. (CATIE) 44, 76–83. Stevenson, D., 1927. Silvicultural treatment of mahogany forests in British Honduras. Emp. For. J. 6, 219–227. Whitman, A.A., Brokaw, N.V.L., Hagan, J.M., 1997. Forest damage caused by selection logging of mahogany (Swietenia macrophylla) in northern Belize. Forest Ecol. Manage. 92, 87–96. Yavitt, J.B., Battles, J.J., Lang, G.E., Knight, D.H., 1995. The canopy gap regime in a secondary Neotropical forest in Panama. J. Trop. Ecol. 11, 391– 402.