Subtropical forests

Subtropical forests, also known as tropical moist forests or subtropical moist broadleaf forests (SMBF), occur worldwide within the tropical humid forest biome. Variously sized tracts of subtropical forest remain in Mexico, Latin America, eastern Australia, New Zealand, east Asia, and southeast Africa. Subtropical forests flourish in humid regions with annual rainfall varying from about 315 to 512 inches (about 8,000 to 13,000 millimeters), with critical temperatures occurring in a window below 11 degrees F (below 24 degrees C).

In the scientific and popular literature, subtropical forests are frequently unspecified to type or geographic distribution, instead identified variously as “tropical forest,” “lowland forest, “tropical evergreen forest, “tropical humid forest, “tropical moist broadleaf forest, or “tropical rainforest.” Inconsistent terminology obscures the characteristic and measurable features differentiating among the three ecosystems comprising the tropical humid forest biome (after L. Holdridge): tropical rainforests, transitional wet forests, and subtropical moist broadleaf forests (SMBF) that transition into seasonal tropical forests. Confusing terminology and mapping contribute not only to the relative lack of scientific and media attention to SMBF but also to their critically endangered status compared to tropical rainforest landscapes. This comparison is exemplified by the conservation status of Brazil's Amazon rainforest versus the seriously threatened moist tropical Mata Atlântica (Atlantic Forest) along Brazil's eastern coast.

The two annual rainfall peaks occurring in subtropical forests are more seasonally differentiated (wet season, dry season) than those in wetter, tropical rainforests, occasioned by characteristic and measurable differences in annual precipitation and temperature between the two forest types. SMBF flourish in humid tropical regions with annual rainfall varying from about 315 to 512 inches (about 8,000 to 13,000 millimeters), with critical temperature occurring in a window below 24 degrees C. On the other hand, annual rainfall in tropical rainforests varies from about 512 to 630 inches (about 13,000 to 16,000 millimeters), with critical temperatures occurring in a window above 24 degrees C. This article not only summarizes information about SMBF available in the scientific literature but also highlights research questions and topics required for assembly of quantitative data based upon modeling and experimentation that will permit differentiation of SMBF from other tropical humid forest ecosystems. Tropical humid forests occurring at high altitude (e.g., tropical “cloud” forest) are not covered in this article.

Ecosystem Variability

P. M. Vitousek and R. L. Sanford discussed the within- and between-ecosystem variability of biomass, production, and nutrient cycling in tropical humid forests, concluding that these variations “nonetheless follow coherent, explicable patterns.” Tropical humid forest biomes feature relatively higher productivity compared to other terrestrial ecosystems, and biodiversity of tropical rainforests is second only to temperate rainforests. Supporting these evidence-based generalizations about the positive association between rainfall and productivity in tropical forests, C. B. Jones found a wider range of group sizes for mantled howler monkeys (Alouatta palliata Gray) in Central American SMBF (Panama) than for groups of the same species inhabiting Mesoamerican seasonal tropical forest (Costa Rica). Jones concluded that higher standard deviations in group size found in Panama were explained by higher carrying capacity in SMBF, the wetter site.

In SMBF, seasonal patterns in plant (primary producers) and animal (primary and secondary consumers) life-history features (survival and fecundity) are less pronounced than those in tropical rainforests and more pronounced than those in seasonal tropical forests. Temporal patterns of biotic life histories (e.g., flowering, fruiting, dispersal) occurring in response to water stress and water availability, in addition to light reflectance, are the most significant factors determining tropical tree species life-history trajectories and diversity. Analyses of species compositions when tropical rainforests, wet forests, subtropical forests, and seasonal tropical forests are compared regionally demonstrate overlap for some plants and animals whose wide habitat ranges result from one or more adaptable biochemical, developmental, or phenotypic traits. On the other hand, some organisms are specialized for or limited to the restricted abiotic or biotic features of tropical humid forests, in some cases because of endemism (forest inhabitants found in one ecosystem type), in others because of physical factors limited to the particular parameters associated with the subtropics (e.g., restricted ranges of thermal tolerances). Barring cryptic biochemical or phenotypic plasticity, species constrained by adaptations to tropical humid forest conditions are particularly vulnerable to anthropogenic effects (climate change, deforestation, habitat fragmentation) as a result of their relatively limited resilience to environmental perturbations and other stressors (e.g., disease).

Testing Whittaker's Ideas

Quantitative and experimental tests of Robert Whittaker's ideas have also been conducted. Investigating the spatial dynamics of species assembly rules with theoretical models, S. Nee and R. M. May modeled metacommunity (“secondary”) effects for two competing species while decreasing the amount of available habitat (“patch removal”). These researchers found that patch removal may decrease the distribution and abundance of superior competitors while increasing the same parameters for inferior competitors, counterintuitive effects dependent upon rates of dispersal or colonization and rates of patch extinction. For plants and animals, patterns of movement are of fundamental importance as they may counteract effects of genetic drift by maintaining connectivity among subpopulations and populations, decreasing likelihoods of isolation and extinction. Consistent with Whittaker's formulations about how ecosystem networks are assembled and change, Nee and May showed that patch removal has the potential to modify the makeup of communities that received no simulated alterations. Gradients (microscale networks), then, may form, in essence, from perturbed locations to patches that have experienced minimal “intrinsic” perturbations of their own, and the previous authors' generalized mathematical treatments showed that these nonlinear landscape effects are difficult to predict. Again, the forgoing research has not been tested in SMBF, studies that should be prioritized for clear distinctions of the previous biome from tropical rainforest and wet forest ecosystems.

Studying a disturbance gradient (primary to old-growth secondary to plantation) in experimentally fragmented central African SMBF transects, J. H. Lawton and his collaborators documented an edge to interior gradient in invertebrate and vertebrate species assemblies, consistent with propositions advanced by Whittaker, whereby ecological structures and processes result from networks of relatively independent species functions and effects. As expected, species richness declined with degree of disturbance. However, as Whittaker predicted, responses to different perturbation states by one species were poor indicators of responses by others. By manipulating spatial dynamics, Lawton's group showed that species in assemblages were more likely to respond individualistically than holistically and interdependently. These findings addressed fundamental questions concerning micro- and macroscale transitions from one patch, habitat, or forest type to another in SMBF. In addition, Lawton's group studied “boundary dynamics” and “edge effects” across changing plant communities from seasonal tropical, to subtropical, to rainforest ecosystems. These results are not only of fundamental importance to basic and applied ecology but also the research programs discussed herein contribute to conservation biology databases, including information required to protect subtropical and seasonal tropical forests buffering rainforests, reservoirs of Earth's richest biodiversity assemblages.

Studying the Causes of Gradients in Tropical Tree Diversity

T. J. Givnish's classic theoretical study “on the causes of gradients in tropical tree diversity” reviewed hypotheses found in the ecological literature advanced to explain variations in tropical forest microscale variations (e.g., precipitation, forest architecture, soil fertility). Within regions, density-dependent plant mortality in tropical wet forests promoted greater forest diversity, a characteristic with significant implications for plant life-history “trajectories.” Two other factors were strongly associated with diversity in tropical forests: (1) a “shifting balance” among tree mortality, competition, and recruitment, and (2) facilitation of diversity by high productivity. The latter relationship is expected if higher tree mortality yields a higher proportion of young, rapidly growing plants, a “fast” life-history strategy, as detailed by S. C. Stearns.

Givnish modeled the aforementioned principal components, accounting for “trends in woody plant diversity along ecological gradients in the tropics.” Since (1) and (2) varied by forest type, precipitation was a principal component in combination with dispersal of seeds by vertebrate consumers. Finally, Givnish was unable to exclude the possibility that random drift accounted for certain of his theoretical results concerning the causes of forest gradients (e.g., “the repeated dominance of particular taxa in separate but ecologically similar sites”). These fundamental studies have not been conducted for the discrete traits of SMBF, highlighting, again, the ecological literature's relative lack of focus on this ecosystem as one separate from other humid tropical forest landscapes.

Variations in Nutrient Cycling

Vitousek and Sanford reviewed variations in worldwide SMBF nutrient cycling, emphasizing physiology and community ecology, as well as plant and animal population biology. These authors reported that, within and between SMBF, variations in nutrient cycling were explained by differences in soil characteristics, in addition to duration and intensity of dry seasons as well as altitudinal gradients. These authors highlighted the importance of evaluating species composition as a source of micro- and macroscale variations in SMBF. However, Vitousek and Sanford noted that these variables are difficult to measure without experimental tests. These researchers found that soil fertility and nutrient cycling are closely linked in SMBF. However, biomass varied little across space even though differences in nutrient concentrations of primary producers varied with “soil fertility.” This research program is of particular significance because results describe distinguishing features of SMBF that permit comparative analyses.

Variations in Soil Characteristics

A review by A. R. Townsend and his colleagues highlighted the importance of variations in soil characteristics as determinants of tropical forest diversity (e.g., soil age, chemical composition). Feedback processes occurred at microscale from soils to plants, on the one hand, and from plants to soils, on the other, operations including effects from plant diversity to variations in “chemical and structural traits” influencing local productivity (e.g., growth rates), decomposition (e.g., leaf litter conditions), and nutrient cycling. At regional levels in tropical ecosystems, biogeochemical variables are strongly influenced by “landscape dynamics” increasing unpredictability in the limiting resources available to plants, organisms upon which animals directly or indirectly depend. This review discussed advances in remote sensing capable of “capturing” features of tropical forest diversity beyond the power of satellite and other “large-scale estimates,” methods with the potential to differentiate one category of tropical humid forest from another.

Soils were also the focus of study by J. N. Price and her colleagues. This empirical work focused on the role of small-scale competition among roots and rhizomes below soil surfaces as determinants of above ground, nonrandom assembly patterns. These authors found that root and rhizome assembly in grassland ecosystems was driven by abiotic factors, data having fundamental import for future investigations in humid forest biomes. Investigating the characteristic dynamics and consequences of inorganic components of materials below SMBF soil surfaces will add an important dimension to the existing body of research documenting critical “soil services” (e.g., regulation and buffering of hydrologic and nutrient cycles as well as detritus).

Studies of Predation

Newer research questions are also advanced in a paper published by O. Schmitz and his colleagues. Field experiments demonstrated that predation, a stressful and, possibly, lethal environmental event, caused herbivorous, temperate-forest grasshoppers to switch from a nitrogen-rich to a carbohydrate-rich diet with correlated changes in soil carbon cycles. Price and Schmitz published their papers in 2012, suggesting that novel mechanisms associated with consumer-producer interactions remain to be discovered and investigated. Scientists conducting research within and across humid tropical forest landscapes are certain to intensify efforts to increase our understanding of known biogeochemical mechanisms as well as formulate hypotheses concerning dynamic abiotic and biotic processes characterizing SMBF.

Geochemical and Biophysical Dynamics and the Biodiversity Crisis

A case study based on research conducted in one Belizean SMBF will serve as an approximate model of geochemical and biophysical dynamics in this ecosystem, highlighting their current biodiversity crisis. According to the Forest Resources Assessment Programme, Belize is a biologically rich Mesoamerican nation with up to 60 percent of original tropical humid forest cover still remaining. Belize's population density is among the lowest in the world, with approximately seventeen people per square kilometer in 2022, according to the World Bank. Black howlers (Alouatta pigra) and Central American spider monkeys (Ateles geoffroyi yucatanensis), both members of the herbivorous primate family Atelidae, are the only nonhuman primates inhabiting Belizean forests, although populations may be fragmented, locally endangered, or locally extinct. In Belize, large tracts of forest and their attendant biogeochemical components are under private oversight (e.g., nongovernmental organizations, farmers). Following R. H. Horwich and J. Lyon, clay and fertile limestone soils are most characteristic of Belizean SMBF, supporting ecosystem gradients from more productive to less productive habitats, respectively. Most SMBF occur as a mosaic of variably sized tracts, from relatively intact stretches of landscape to fragments and patches of habitats formed by anthropogenic effects, particularly selective cutting, deforestation, rural poverty, external markets, unsustainable patterns of resource exploitation, poor management, and “milpa” (“slash-and-burn”) agricultural practices.

One suggestive pattern of results is that, according to geographic surveys, black howlers, folivore-frugivores, entered Belize from the north, moving southward across the Maya Mountains and Cockscomb Range in southern Belize. On the other hand, the mountains appear to have been a geographic barrier for the primarily frugivorous spider monkeys spreading from south to north. The ability to process leaves (black howlers) as a nutritional source is thought to provide a “fallback” dietary source permitting flexible switching of food sources, enhancing capacities for dispersal and colonization.

As M. L. Cody pointed out, frugivores (spider monkeys), on the other hand, are more sensitive and vulnerable to environmental stress, including variations in the predictability of fruit in time and space, conditions that may limit or retard the geographical spread of species. Colonizing abilities and frugivory are generally unrelated, unless fruit is evenly rather than patchily dispersed. A satellite image of forest cover created an overlay for the species distribution map, showing that, consistent with expectation, monkeys were least likely to occur along Belize's coastline where deforestation is most severe. A. pigra persist in more deforested areas, reflecting robustness due to their tolerance of leaves. A. geoffroyi yucatanensis was more likely to occur in larger tracts of SMBF, similar to distribution patterns observed by C. A. Jost for howler and spider monkeys in Costa Rica.

Terminology and Definitions

Conventional terminology and definitions of tropical humid ecosystems have not been standardized in the scientific literature. However, in 2002 D. M. Olson and E. Dinerstein (World Wildlife Fund) published a global classification of 238 “ecoregions,” defined as “regional-scale (continental-scale) units of biodiversity.” These authors determined an ecoregion's vulnerability to extirpation based on qualitative measures of species richness and endemism, uniqueness of abiotic and biotic components, degree of fragmentation, as well as other factors. Of 238 ecoregions, 50 occurred in tropical or SMBF. Most of the latter ecoregions were found in the Paleotropics (Old World: 38/50 ecoregions), and, SMBF were most commonly represented overall (27/50). Among terrestrial ecoregions worldwide, species diversity of the SMBF, Atlantic Forest, was second only to Amazon rainforests.

Despite the conceptual utility of Olson and Dinerstein's ecoregion analysis, their model has not become the scientific standard, possibly because it is based on qualitative factors rather than quantitative parameters. Nonetheless, this paper provides important details about what is not known about humid tropical forests that would permit conservation biologists to develop quantitative, predictive models to facilitate the preservation of biodiversity in SMBF. Increased knowledge of SMBF will advance what E. O. Wilson has called “the key problem facing humanity in the coming century, how to bring a better quality of life for 8 billion or more people without wrecking the environment entirely in the attempt.”

Early Subtropical Forest Research

Biome and ecosystem concepts were developed initially by several investigators central to the establishment of ecology as a scientific discipline distinguishable from natural history. Since the 19th century, biomes and ecosystems have been understood as assemblages of organisms inhabiting characteristic environments. However, early researchers held different perspectives about what abiotic (geochemical) and biotic (organismal) regulatory mechanisms accounted for ecosystem functions, complexity, interspecific associations, and stability. As synthesized by S. A. Levin, broadly speaking, one perspective emphasized a reductionistic approach whereby biomes and ecosystems might be assembled and reassembled from basic abiotic and biotic components. Other ecologists thought of biomes and ecosystems holistically, whereby components interacted synergistically, a viewpoint emphasizing component interdependency more than the independent function and, possibly, changeability of component parts.

Though these formulations are not mutually exclusive in any strict sense, they characterize different schools of thought in contemporary ecology. The contributions of F. E. Clements, H. A. Gleason, A. G. Tansley, Charles Elton, Albert Lotka, R. Lindeman, and Eugene P. Odum represent classic treatments of ecosystem and biome concepts. Later, Whittaker and Holdridge independently introduced the first comprehensive maps and classification systems of terrestrial habitats, ecosystems, and biomes throughout the world.

A gradient of temperature and precipitation occurs across tropical humid forests. R. H. Whittaker introduced the gradient analysis concept to explain changes in the distribution and abundance of individual plant species along correlation curves in an attempt to document gradual and individualistic (i.e., species-typical) patterns of change along topological gradients. Whittaker's schema was supported in Argentina by G. E. Zunino's research team studying six ecological factors along biogeochemical gradients in five different ecosystems, including SMBF (“paranaense”). In 2004, J. B. Bastow and his colleagues reevaluated Whittaker's database, finding his methods and results unconvincing, primarily because of inappropriate sampling of plants. Despite their criticisms of Whittaker's methods and results, Bastow et al., consistent with other reports, empirically confirmed that Whittaker's gradient analysis paradigm is, effectively, correct.

Analyses of gradients are essential to landscape and metapopulation ecology, in particular, the causes and consequences of species' spatial dispersion. The aforementioned research questions emphasize patch and subpopulation dynamics, especially the ways in which these mechanisms, functions, and effects influence dispersal, colonization, and other plant and animal population and subpopulation phenomena. The “patch” network view of physical and biotic environments has received increased attention in recent years due to researchers' attempts to document the effects of habitat fragmentation and other landscape perturbations on populations. T. A. Guisan and his colleagues, among other theoretical ecologists, have developed conceptual and simulation models for the analysis of gradients and landscapes, methods in need of application to the distinctive quantitative traits differentiating SMBF from other tropical humid forests.

Questions related to spatial dynamics in tropical and temperate forests have been addressed by numerous ecologists, notably R. H. MacArthur, R. C. Lewontin, J. H. Connell, S. Harrison, S. Nee, R. M. May, J. H. Brown, J. Clobert, A. A. Dhondt, E. Danchin, S. A. Levin, I. A. Hanski, M. E. Gilpin, and other scientific associates. Though research on spatial dynamics in tropical forests is less common in SMBF compared to tropical rainforests and seasonal tropical forests, descriptive research has been conducted in SMBF by L. Poorter (trees, Bolivia), G. H. Adler (rodents, Panama), J. H. Lawton (beetles, Central Amazon), and J. E. Fa (mammals, western and central Africa), among others. Despite theoretical and empirical support for the robustness of gradient analysis at the level of microclimates, Vitousek and Sanford cautioned that attempts to apply the paradigm as a continuum across macroclimates (across forest types) are suspect because of significant variability and discontinuity of tropical forest characteristics, termed “state factors” by H. Jenny (biotic and substrate components, climate, topography, and time).

South America's Atlantic Forest

One of the world's most ecologically important, critically threatened SMBF is South America's Atlantic Forest (Mata Atlântica), approximately 330 million acres stretching from northeastern Brazil to northeast Argentina and eastern Paraguay. According to the Worldwide Fund for Nature, as of 2020, only 7 of the original forest existed. The evolution of biotic diversity (e.g., species richness), community assembly patterns, as well as abiotic, biogeochemical gradients (e.g., variations in below- and aboveground soil compositions) have been facilitated by Pleistocene forest shrinkage patterns in addition to factors discussed elsewhere in this article. This United Nations Educational, Scientific and Cultural Organization World Heritage Site exhibits biodiversity rivaling that of the Brazilian Amazon, including many threatened or endangered endemic plant and animal species, as well as two indigenous human tribal groups, the Tupi and the Guarani. One source quoted the Brazilian conservation biologist, A. R. Mendes Pontes: “Species in the Mata Atlântica are the living dead.” Gradually, efforts have been made to reverse some of the negative human impact on the area. According to the World Bank, by 2017, its sponsored Paraguay Biodiversity Project had succeeded in the restoration of over three hundred thousand hectares of the forest's protected lands in the country. In 2018, an outbreak of yellow fever led to the deaths of hundreds of the forest's howler monkeys; scientists speculated that the virus was able to spread quickly throughout the population because of the shrinkage of the monkeys' habitat due to longtime deforestation practices, and concern grew over the future of animal and plant life in the threatened forest.

Bibliography

"Brazil: Restoration of the Atlantic Forest (Mata Atlântica)." Society for Ecological Restoration, 2018, www.ser-rrc.org/project/brazil-restoration-of-the-atlantic-forest-mata-atlantica/. Accessed 30 July 2018.

Duchiade, André. "Brazilian Forests Fall Silent as Yellow Fever Decimates Threatened Monkeys." Scientific American, 5 July 2018, www.scientificamerican.com/article/brazilian-forests-fall-silent-as-yellow-fever-decimates-threatened-monkeys/. Accessed 30 July 2018.

"Population Density (People per Sq. Km of Land Area)." The World Bank, data.worldbank.org/indicator/EN.POP.DNST. Accessed 30 July 2018.

"Protecting the Atlantic Forest: Creating a Biodiversity Corridor in Eastern Paraguay." The World Bank, 30 Oct. 2017, www.worldbank.org/en/results/2017/10/30/protecting-the-atlantic-forest-creating-a-biodiversity-corridor-in-eastern-paraguay. Accessed 30 July 2018.

Seibt, Tais, and Duda Menegassi. "Brazil's Atlantic Forest Gets a Chance at a Fresh Start Through Restoration." Mongabay, 7 Mar. 2022, news.mongabay.com/2022/03/brazils-atlantic-forest-gets-a-chance-at-a-fresh-start-through-restoration/. Accessed 16 July 2022.

"Tropical and Subtropical Moist Broadleaf Forests." World Wildlife Fund, www.worldwildlife.org/biomes/tropical-and-subtropical-moist-broadleaf-forests. Accessed 30 July 2018.