Theobroma cacao on the regeneration of the riparian forest at La Selva Biological Station, Costa Rica

Carol C. Horvitz ^1,*, Mario A. Blanco ², Orlando Vargas ³

* Corresponding author, e-mail address: carolhorvitz@miami.edu

Cacao farms vary considerable in their shade management and geographical position with respect to distance from unmanaged forest (Beer 1997). Shade management systems in cacao form a gradient similar to that of coffee, where rustic systems utilize more local species than planted shade systems or zero shade systems. The systems previously in place at our field site was rustic cacao management, which is characterized by planting cacao under a thinned forest canopy. This system is used in West Africa and Latin America (Wood 1985). In planted shade systems vary in the diversitybut more often it is grown under a diverse canopy of shade trees (planted shade cacao). Due to fluctuations in market prices cacao can periodically suffer from periods of overproduction, which eventually lead to boom/bust cycles at the level of production (Rica and Greenberg 2000). The relative importance of northern Latin America to world cacao production has declined over the last 50 years due to persistent disease problems (Wood and Lass 1985). It is not known how abandoned cacao effects alluvial forest regeneration, and if removing cacao after abandonment can speed up succession.

Orchards are plantations managed solely for fruit production. Compared to timber plantations orchards are more intensively managed because they are harvested annually or biannually, are weeded, and are typically monocultures.

Their goal was to compare the forest structure of abandoned: pasture (<5yr), mixed plantings (<5yr), old mixed plantings (7-30yrs), cacao plantations (>25yrs), old forest, and forest with no history of human use. Even though the cacao plantation was abandoned 25 yr ago the diversity was low due to continued regeneration of this persistent crop. They have 2 a cacao sites. Tree density and basal area was significantly different between the land use sites. Cacao sites had intermediate basal area, but low density of stems. Species richness was the second lowest in cacao. They do not have parallel chronosequences in the different habitats, so they can’t compare. They did find seedlings and saplings recruiting in the plots. Cacao plots were similar in composition to the old mixed plantings.

The number of stems per hectare increased significantly before and after the hurricane in the cacao plantation and the young conuco.

Goals: measure/compare differences in secondary forest structure, tree diversity, soil properties between abandoned cacao plantations, abandoned palm plantations, pasture, and mature forests. They are comparing the human modified habitats to mature forests from which the human modified habitats were cut from, they say that the mature forest provides an estimate of the state of the system prior to human disturbance (they are substituting space for time). Sites were selected based on the period of land use, time of abandonment, and soil properties at the time of forest clearing. The sites were chosen so that the three adjacent land uses on a single soil series could be sampled. The sites were all on alluvial terrace soils at La Selva. The agricultural sites were cleared 30 years ago, and abandoned 7 years ago. The plots were 30 x 30m except for the cacao plot which was 45 x 20 m. The cacoa plot had the highest basal area and stem density. The tree Neea psychotrioides (46% of individuals), Cordia alliodora accounted for 18% of basal area, Neea was 14% of basal area. The cacoa plot had the lowest diversity and the least species in common with the primary forest.

River meandering

Rivers overflowing their banks during floods and depositing alluvium

Many rivers

Primary succession in the Amazonian forest is caused by later erosion and channel changes on meandering rivers. New deposits of riverine soils is a major mode of forest regeneration.

Floods create disturbance, and a distinct regeneration niche

Stem density is lower in riparian forests because mature trees

Effect of exotic species removal on regeneration

Regeneration after agriculture
Paper: Martin, P.H. 2004. Forty years of tropical forest recovery from agriculture: Structure and floristics of secondary and old-growth riparian forests in the Dominican Republic. Biotropica 36: 297-317.
Do protected areas of primary forests in the tropics act as effective sources of native plant species for the recolonization of abandoned lands?
Does the high diversity of life forms that typify mature tropical forests recover along the same trajectory as species richness?
They describe the floristic composition and structure of 40-year-old secondary forests and upstream old-growth forest in the riparian zone of two rivers in the Dominica Republic.
Hypothesis: assuming dispersal limitation of isolated rare species, species richness will be lower in the secondary forest although abundance and diversity of some plant life-forms (epiphytes) would remain lower in the post-agricultural forests even after 40 years.
Area was used for agriculture during 60 years.
Land use consisted on low-intensity agriculture, supplemented with introduced fruit trees, coffee, and small-scale animal husbandry.
The sampled the vegetation using circular plots, and measuring woody vegetation in three size classes: 1) trees, 2) saplings, 3) seedlings, the last two measured in a nested strip plot. Woody stems were classified by life-form. They compared structure, life-form, and species density between old and secondary forests.
Results

Species densities were not different between forests. Basal-area was higher in the old-forests. Introduced species was a high proportion of tree abundance in secondary forests, and were absent from old-growth forests. There was a higher number of species in the old-growth forest (213) than in the secondary forest (157). However, mean forest type and site-level comparisons indicated that the number of tree species was comparable between secondary and old-growth forests. When only considering native tree species, the mean number of species per plot was significantly lower in secondary plots.

This study was conducted between 1996 and 2005 at La Selva Biological Station. La Selva is a tropical humid forest located in the Caribbean lowlands of Costa Rica, on the confluence of the Sarapiquí and Puerto Viejo rivers (10۫ 26’ N, 83۫ 59’ W). The elevation ranges from 35 – 150 m. and the annual mean rainfall is 3993 mm, with a mean monthly temperature of 24 – 27 ۫C (McDade and Hartshorn, 1994). La Selva is a mosaic of mature forest as well as secondary forest, old pastures and abandoned plantations in different stages of regeneration. In the year 1953, cacao trees (Theobroma cacao, Malvaceae) were planted on most of the alluvial soils along the Sarapiquí and Puerto Viejo rivers (Cite here). Planting of trees was done after felling some of the canopy trees and clearing all of the understory (Fernandes and Sandford, 1995). The alluvial soils at La Selva are located along the plain terrace (Fig. 1). For more details see McDade and Hartshorn (1994).

We selected four sites with abandoned cacao plantations in the riparian forest of the Puerto Viejo and Sarapiquí rivers, and the Sábalo creek (Fig. 1). The minimum distance among sites was _______. Cacao plantations in these sites were abandoned between 22 to 39 years ago (Table 1). Sites were located at similar altitude, about 36 msnm (Table 1). Although all sites have alluvial soils, edaphic characteristics of Site 1 differed from the soils at other sites (Table 1). This is a possible a consequence of the position of Site 1 on alluvial soils originated from the Sarapiquí River whereas the other sites are located in sites with alluvial sediments from common origin, either next to the Puerto Viejo River or its affluent, the Sábalo Creek (Fig. 1).

The criterium to select these four sites were to represent different land use histories and alluvial soil types. Sites 1 to 3 where planted only with cacao whereas Site 4 was originally a mixed plantation of cacao and Cordia alliodora (Boraginaceae), a tree extensively planted in Costa Rica for timber (Montagnin, 1994).

To determine if there is an effect of the removal of cacao trees on the processes of vegetation succession in the riparian forest, we performed the following experiment. In each site we placed ten 18 x 18 m contiguous plots (Fig 1). In five plots, all cacao trees were eliminated (treatment plots). The cacao trees in the other five plots (control) were not removed.

In ____ (it must have been after March 1996) we started the elimination of cacao trees from the treatment plots by ringing (i.e stripping a band of bark around the trunk near its base) and applying diesel fuel to the wound. After one year 48.8% of the treated cacao trees were still alive. For this reason we repeated the ringing in these trees and added herbicide to the wound. As part of a program for the eradication of non-native species at La Selva, all the cacao trees present outside of the plots were also ringed and herbicide was added to the wounds, although some of them were still alive in 2006.

The sampling area within each plot was a 5 m radius circle at the center of each plot (total area = 78.54 m², Fig. 1). We sampled an area smaller than the whole plot to avoid edge effects such as branches from adjacent cacao trees intersecting the treatment plots.

Within each sampling circle we recorded the number of individuals per species in the following size categories: 2 to 6 cm, and greater than 6 cm of diameter at breast height (dbh). We also recorded the number of smaller individuals per species (less than 2 cm dbh) in a one-quarter of the sampling circle (total area of each subplot = 19.64 m², Fig. 1). To determine the original species composition and the number of individuals per stem size category, we performed the first census on March 1996, before removing the cacao trees. After initiation of the treatment, all plots and subplots were surveyed annually from 1997 to 2005.

Plant species were initially identified by collecting a small branch, which was determined using the reference herbarium at La Selva and the station botany staff. Most of the plants were not fertile at the time of the censuses and therefore were not suitable to prepare herbarium vouchers. In subsequent censuses, we identified species in the field. We only collected a small sample for identification when the plant identity was not evident.

To understand the general effect of cacao removal on the regeneration of the entire riparian forest community as well as of specialist riparian species, we performed separate statistical analyses for the whole community, and for the subset of species specialists of riparian forest. We define riparian specialists as those species that at our study area require alluvial soils and riparian habitats that currently or historically were affected by periodic river floods. This classification is based on the known distribution of the species and habitat preferences at La Selva and surrounding areas. Although this classification is valid in the context of our study sites, habitat use of species included in this research may be different in other localities within their geographic range (References here).

For each stem size category we explored differences in 1) stem density and 2) species richness among the four sites included in this study, between treatment and control plots, and among different census years. Differences were tested with a two way repeated measures ANOVA. All data analyses were performed using the program SAS (insert here ref. of SAS).

3. Results
4. Discussion
5. Conclusions
Acknowledgements
References

Fernandes, D.N., Sandford Jr., R.L. 1995. Effects of recent land-use practices on soil nutrients and succession under Tropical Wet Forest in Costa Rica. Conserv. Biol. 9: 915-922.

Hartshorn, G.S., Hammel, B.E., 1994. Vegetation Types and Floristic Patterns. In: McDade, L.A., Bawa, K. S., Hespenheide, H.A. and Hartshorn G.S. (Eds.), La Selva: Ecology and Natural History of a Neotropical Rain Forest. University of Chicago Press, Chicago. pp. 73-89.

McDade, L.A., Hartshorn, G.S., 1994. La Selva Biological Station. In: McDade, L.A., Bawa, K. S., Hespenheide, H.A. and Hartshorn G.S. (Eds.), La Selva: Ecology and Natural History of a Neotropical Rain Forest. University of Chicago Press, Chicago. pp. 6-14.

Montagnini, F., 1994. Agricultural systems in the La Selva Region. In: McDade, L.A., Bawa, K. S., Hespenheide, H.A. and Hartshorn G.S. (Eds.), La Selva: Ecology and Natural History of a Neotropical Rain Forest. University of Chicago Press, Chicago. pp. 307-316.

Sanford Jr., R.L., Paaby, P., Luvall, J.C., Phillips, E., 1994. Climate, Geomorphology, and Aquatic Systems. In: McDade, L.A., Bawa, K. S., Hespenheide, H.A. and Hartshorn G.S. (Eds.), La Selva: Ecology and Natural History of a Neotropical Rain Forest. University of Chicago Press, Chicago. pp. 19-33.

Sollins., P., Sancho M, F., Mata Ch., R., Sanford Jr., R.L., 1994. Soils and Soil Process Research. In: McDade, L.A., Bawa, K. S., Hespenheide, H.A. and Hartshorn G.S. (Eds.), La Selva: Ecology and Natural History of a Neotropical Rain Forest. University of Chicago Press, Chicago. pp. 34-53.

Table 1

Fig. 1.Study plot distribution in abandoned cacao plantations at La Selva Biological Station. Note the detail showing how plots were arranged in each site. C = control, T = treatment

Beer, J.,1997. Shade management in coffee and cacao plantation. Agroforest. Syst. 4, 1-26.

Cavelier, J., Santos, C.,1999. Efectos de plantaciones abandonadas de especies exóticas y nativas sobre la regeneración natural de un bosque montano en Colombia (in Spanish with English abstract). Rev. Biol. Trop. 47, 775-784.

Colón, S.M., Lugo, A.E.,2006. Recovery of a subtropical dry forest after abandonment of different land uses. Biotropica 38, 354-364.

Fernandes, D.N., Sanford, R.L.,1995. Effects of recent land-use practices on soil nutrients and succession under tropical wet forest in Costa Rica. Conserv. Biol. 9, 915-922.

Lugo, A.E.,1992. Comparison of tropical tree plantations with secondary forests of similar age. Ecol. Monogr. 62, 1-41.

Martin, P.H.,2004. Forty years of tropical forest recovery from agriculture: Structure and floristics of secondary and old-growth riparian forests in the Dominican Republic. Biotropica 36, 297-317.

Rice, R.A., Greenberg, R.,2000. Cacao cultivation and the conservation of biological diversity. Ambio 29, 167-173.

Rolim, S.G., Chiarello, A.G.,2004. Slow death of Atlantic forest trees in cacoa agroforestry in Southeastern Brazil. Biodivers. Conserv. 13, 2679-2694.

Rivera, L.W., Zimmerman, J.K., Aide, T.A.,2000. Forest recovery in abandoned agricultural lands in a karst region of the Dominican Republic. Plant Ecol. 148, 115-125.

Salo, J., Kalliola, R., Hakkinen, I., Makinen, Y., Niemela, P., Puhakka, M., Coley, P.1986. River dynamics and diversity of Amazon lowland forest. Nature, 254-258.

Uriarte, M., Rivera, L.W., Zimmerman, J.K., Aide, T.M., Power, A.G., Flecker, A.S.,2004. Effects of land use history and hurricane damage and recovery in a neotropical forest. Plant Ecol. 174, 49-58.