Search

We present a Holocene relative sea-level (RSL) database for the Caribbean region (5°N to 25°N and 55°W to 90°W) that consists of 499 sea-level index points and 238 limiting dates. The database was compiled from multiple sea-level indicators (mangrove peat, microbial mats, beach rock and acroporid and massive corals). We subdivided the database into 20 regions to investigate the influence of tectonics and glacial isostatic adjustment on RSL. We account for the local-scale processes of sediment compaction and tidal range change using the stratigraphic position (overburden thickness) of index points and paleotidal modeling, respectively. We use a spatio-temporal empirical hierarchical model to estimate RSL position and its rates of change in the Caribbean over 1-ka time slices. Because of meltwater input, the rates of RSL change were highest during the early Holocene, with a maximum of 10.9 ± 0.6 m/ka in Suriname and Guyana and minimum of 7.4 ± 0.7 m/ka in south Florida from 12 to 8 ka. Following complete deglaciation of the Laurentide Ice Sheet (LIS) by ∼7 ka, mid-to late-Holocene rates slowed to

Barriers within streams can affect riverine species' ability to access habitats and may reduce their population viability. Connectivity metrics attempt to quantify the impacts of barriers; however, little is known about their functioning when applied to dendritic habitats such as watersheds. Several graph-theoretic connectivity metrics were calculated on rivers originating in the Luquillo Mountains of Puerto Rico. These metrics were classified into two primary groups: metrics that count weighted paths through the stream network and metrics that predict the flow of organisms through a stream reach. Representative metrics from each of these categories were suggested to model the effects of dams and water intakes, respectively. Copyright © 2014 John Wiley & Sons, Ltd.

Humid tropical forests play a dominant role in many global biogeochemical cycles, yet long-term records of tropical stream chemistry and its response to disturbance events such as severe storms and droughts are rare. Here we document the long-term variability in chemistry of two streams in the Luquillo Mountains, Puerto Rico over a period of 27 years. Our two focal study watersheds, the Rio Icacos and Quebrada Sonadora, both drain several hundred hectares of tropical wet forests, and each received direct hits from Hurricanes Hugo (1989) and Georges (1998). They differ primarily in lithology (granitic vs. volcaniclastic) and elevation. Changes in major cations, anions, silica, and dissolved organic carbon were minimal over the study period, but the concentrations of nitrate show a strong response to hurricane disturbance and the longest time to recovery. Potassium also showed a large, although less consistent, response to disturbance. In the granitic terrain, nitrate concentrations exceeded long-term pre-hurricane background levels for over a decade, but were elevated in the volcaniclastic terrain for only 1–2 years. Lithology appears to be the primary driver explaining the different response trajectories of the two watersheds. In the granitic terrain, which showed slow recovery to pre-hurricane conditions, the quartz diorite bedrock weathers to produce coarser soils, deeper groundwater flow paths, and riparian zones with sharp spatial variation in redox conditions and very high nitrogen levels immediately adjacent to the stream. Groundwater flow paths are shallow and the levels of N in streamside groundwater are much lower in the volcaniclastic terrain. The recovery of vegetation following hurricane disturbance appears similar in the two watersheds, suggesting that the extent of structural damage to canopy trees determines the magnitude of NO3 increases, but that the duration of elevated concentrations in stream water is a function of lithology.

This study investigated the landscape characteristics that influence C and N in unsaturated surface soils of riparian zones along 1st to 3rd order streams in the Atlantic Coastal Plain of the Delaware River Basin. Unsaturated surface soils (0–30 cm) were sampled in forested and non-forested sites at 29 locations throughout S New Jersey and SE Pennsylvania. Overall, the soil %C and %N in forested and non-forested riparian sites studied in this investigation were comparable to similar riparian zone soils in eastern North America. However, the soil C and N contents of these Atlantic Coastal Plain soils were 3 to 8-fold greater which underscores the value of these riparian soils as C pools. Soil C content (100.3 ± 15.0 Mg ha−1) in forested riparian sites was consistently higher but not statistically different (P > 0.05) from soil C content (90.6 ± 12.1 Mg ha−1) in non-forested riparian sites. Likewise, neither soil N storage or the C:N ratio were different between the contrasting land covers but forested sites with forest floor organic horizons had significantly greater (82%, P = 0.004) soil C storage than the non-forested sites. Of the forested sites, 70% did not have organic horizons. All of the forested sites without organic horizons had abundant earthworms and comparisons of sites with and without forest floor suggests that earthworms and the removal of native forest cover may be responsible for a loss of 75–93 Mg ha−1 of soil C from these riparian zones. Multivariate regression tree analysis was able to explain ⩾50% of the variability in soil C and N and as much as 68% of the variability in the C:N ratio. The analysis indicated that watershed-scale land cover, local soil series, and elevation above the active channel had the greatest influence on C and N storage. Moreover, this analysis indicated that a combination of easily measured, reach-scale characteristics and GIS-based watershed-scale variables can be used to estimate regional riparian soil C pools and identify restoration sites with the potential to store soil C.

Recent work has suggested that weathering processes occurring in the subsurface produce the majority of silicate weathering products discharged to the world s oceans, thereby exerting a primary control on global temperature via the well-known positive feedback between silicate weathering and CO2. In addition, chemical and physical weathering processes deep within the critical zone create aquifers and control groundwater chemistry, watershed geometry and regolith formation rates. Despite this, most weathering studies are restricted to the shallow critical zone (e.g., soils, outcrops). Here we investigate the chemical weathering, fracturing and geomorphology of the deep critical zone in the Bisley watershed in the Luquillo Critical Zone Observatory, Puerto Rico, from two boreholes drilled to 37.2 and 27.0 m depth, from which continuous core samples were taken. Corestones exposed aboveground were also sampled. Weathered rinds developed on exposed corestones and along fracture surfaces on subsurface rocks slough off of exposed corestones once rinds attain a thickness up to ~1 cm, preventing the corestones from rounding due to diffusion limitation. Such corestones at the land surface are assumed to be what remains after exhumation of similar, fractured bedrock pieces that were observed in the drilled cores between thick layers of regolith. Some of thesemore » subsurface corestones are massive and others are highly fractured, whereas aboveground corestones are generally massive with little to no apparent fracturing. Subsurface corestones are larger and less fractured in the borehole drilled on a road where it crosses a ridge compared to the borehole drilled where the road crosses the stream channel. Both borehole profiles indicate that the weathering zone extends to well below the stream channel in this upland catchment; hence weathering depth is not controlled by the stream level within the catchment and not all of the water in the watershed is discharged to the stream.« less

Cloud- and rain-water samples collected between 1984 and 2007 in the Luquillo Experimental Forest, Puerto Rico, were analyzed in order to understand the main processes and sources that control their chemistry. Three sites were used: El Verde Field Station (380 m asl), Bisley (361 m asl), and East Peak (1051 m asl). Bulk rainwater samples were collected from all sites, and cloud water was also collected on East Peak. All samples were analyzed for pH, conductivity, and concentrations of Cl−, SO42−, NO3−, NH4+, Na+, K+, Ca2+ and Mg2+. Similar patterns in overall chemistry were observed for both cloud- and rain-water samples. The majority of samples had low acidity (average pH of 4.4–5.0), similar to other remote sites. Sea salt (Na+ and Cl−) had a large influence on rain and cloud chemistry and accounted for approximately 70% of the total mass of solutes, followed by SO4=, which controls the acidity of the clouds and rainwater. Calcium accounted for 6–8% of the total cations and dominated neutralization processes. The highest concentrations of Ca2+ and NO3− in both cloud- and rain-water were observed in the summertime when large amounts of dust from the African continent reached the sites. Enrichment Factor and Principal Component Analyses showed that Na+, Cl−, and Mg2+ in the cloud- and rain-water were primarily of marine origin, while most of the Ca2+ was from crustal sources; and NO3− was predominantly anthropogenic, presumably from both local and long-range sources. In general, the results of this study suggested that cloud- and rain-water chemistry in northeastern Puerto Rico is strongly influenced by natural and marine sources rather than local anthropogenic sources. The pollutant species in the samples were mainly derived from long distance transport.

Urbanization through the addition of impervious cover can alter catchment hydrology, often resulting in increased peak flows during floods. This phenomenon and the resulting impact on stream channel morphology is well documented in temperate climatic regions, but not well documented in the humid tropics where urbanization is rapidly occurring. This study investigates the long-term effects of urbanization on channel morphology in the humid sub-tropical region of Puerto Rico, an area characterized by frequent high-magnitude flows, and steep coarse-grained rivers. Grain size, low-flow channel roughness, and the hydraulic geometry of streams across a land-use gradient that ranges from pristine forest to high density urbanized catchments are compared. In areas that have been urbanized for several decades changes in channel features were measurable, but were smaller than those reported for comparable temperate streams. Decades of development has resulted in increased fine sediment and anthropogenic debris in urbanized catchments. Materials of anthropogenic origin comprise an average of 6% of the bed material in streams with catchments with 15% or greater impervious cover. At-a-station hydraulic geometry shows that velocity makes up a larger component of discharge for rural channels, while depth contributes a larger component of discharge in urban catchments. The average bank-full cross-sectional area of urbanized reaches was 1.5 times larger than comparable forested reaches, and less than the world average increase of 2.5. On average, stream width at bank-full height did not change with urbanization while the world average increase is 1.5 times. Overall, this study indicates that the morphologic changes that occur in response to urban runoff are less in channels that are already subject to frequent large magnitude storms. Furthermore, this study suggests that developing regions in the humid tropics shouldn't rely on temperate analogues to determine the magnitude of impact of urbanization on stream morphology. Copyright © 2012 John Wiley & Sons, Ltd.

Worldwide concern for continuing loss of montane forest cover in the tropics usually focuses on adverse ecological consequences. Less recognized, but equally important to inhabitants of these affected regions, is an increasing susceptibility to rainfall-induced debris flows and their associated impacts. The same high rainfall rates that sustain tropical forest cover can often serve as the triggering mechanism for debris flows. The natural rate of debris flow occurrence on steep slopes subject to episodic, intense rainfall is dependent on the stabilizing effect of tropical forests. Either loss or significant reduction in forest cover can weaken this natural defense. Information from postdisaster observations and research on the November 1988 storm event in southern Thailand provides a case study illustrating the potential impacts of increased debris flow susceptibility resulting from conversion of forest cover to rubber tree crops. Development resulting in the loss of tropical forest cover may be accompanied by local increase in population, property development, and infrastructure. Consequently, the potentially disastrous consequences of increased debris flow occurrence are amplified by the greater vulnerability of local populations. Preserving the tropical forest cover is an obvious and often difficult means of retaining this natural protection. Effective policy should capitalize on the values of tropical forests as part of the strategy for retaining adequate forest cover. Policy should also seek to avoid creating pressures that foster forest removal or their conversion to other types of land cover in steep terrain. Areas where tropical forests were converted to other cover types can be restored to secondary forests to avoid a permanent state of increased debris flow susceptibility. Restoration of secondary tropical forests can successfully re-establish the forest characteristics that limit debris flow occurrence. Experience in Central America and the Caribbean demonstrates that successful restoration is possible but requires a significant commitment of both time and resources. In addition to the cost and technical difficulties involved, the increased susceptibility to debris flow occurrence persists through many years until successful restoration is achieved. Both retention of existing tropical forests and restoration of forest cover where loss has occurred are often justified by the reduced risk of debris flow impacts to vulnerable populations and infrastructure.

article i nfo Soil organic carbon (SOC) in a humid subtropical forest in Puerto Rico is higher at ridge locations compared to valleys, and therefore opposite to what is commonly observed in other forested hillslope catenas. To better understand the spatial distribution of SOC in this system, plots previously characterized by topographic position, vegetation type and stand age were related to soil depth and SOC. Additional factors were also investigated, including topographically-related differences in litter dynamics and soil chemistry. To investigate the influence of litter dynamics, the Century soil organic model was parameterized to simulate the effect of substituting valley species for ridge species. Soil chemical controls on C concentrations were investigated with multiple linear regression models using iron, aluminum and clay variables. Deeper soils were associated with indicators of higher landscape stability (older tabonuco stands established on ridges and slopes), while shallower soils persisted in more disturbed areas (younger non-tabonuco stands in valleys and on slopes). Soil depth alone accounted for 77% of the observed difference in the mean 0 to 60 cm SOC between ridge soils (deeper) and valley soils (shallower). The remaining differences in SOC were due to additional factors that lowered C concentrations at valley locations in the 0 to 10 cm pool. Model simulations showed a slight decrease in SOC when lower litter C:N was substituted for higher litter C:N, but the effects of different woody inputs on SOC were unclear. Multiple linear regression models with ammonium oxalate extractable iron and aluminum, dithionite-citrate-extractable iron and aluminum, and clay contents explained as much as 74% of the variation in C concentrations, and indicated that organo-mineral complexation may be more limited in poorly developed valley soils. Thus, topography both directly and indirectly affects SOC pools through a variety of inter-related processes that are often not quantified or captured in terrestrial carbon models.

Low street tree survival rates and the resulting short lifespans are frequently discussed but inadequately quantified in urban forestry literature. This study addresses street tree survival rates with a meta-analysis of previous literature and a case study assessment of street trees in Philadelphia, USA 2–10 years after planting. Reported survivorship rates from 16 previous studies were compiled. Estimated annual survival rates for individual past studies were mostly above 91.0%. To estimate the annual street tree survival rate across multiple studies, a subset of previous studies was pooled for regression analysis of ln transformed cumulative survivorship vs. time since planting. Lifespan metrics were estimated using the annual survival rates determined from regression analysis. Based on the meta-analysis, we estimated that street tree annual survival rates ranged from 94.9 to 96.5%, and street tree population half-life ranged from 13 to 20 years. Estimated mean life expectancy ranged from 19 to 28 years, which is considerably longer than the 7- or 13-year street tree average lifespan reported in previous studies. Estimated annual survival rates and lifespan metrics were similar in the Philadelphia case study. Urban forest researchers are encouraged to use demographic concepts and analyses in the study of tree survival and mortality, and to monitor tree survival at repeated time intervals every few years.

Tropical montane cloud forests (TMCF) typically experience conditions of frequent to persistent fog. On the basis of the altitudinal limits between which TMCF generally occur (800–3500 m.a.s.l. depending on mountain size and distance to coast) their current areal extent is estimated at ∼215 000 km2 or 6·6% of all montane tropical forests. Alternatively, on the basis of remotely sensed frequencies of cloud occurrence, fog-affected forest may occupy as much as 2·21 Mkm2. Four hydrologically distinct montane forest types may be distinguished, viz. lower montane rain forest below the cloud belt (LMRF), tall lower montane cloud forest (LMCF), upper montane cloud forest (UMCF) of intermediate stature and a group that combines stunted sub-alpine cloud forest (SACF) and ‘elfin’ cloud forest (ECF). Average throughfall to precipitation ratios increase from 0·72 ± 0·07 in LMRF (n = 15) to 0·81 ± 0·11 in LMCF (n = 23), to 1·0 ± 0·27 (n = 18) and 1·04 ± 0·25 (n = 8) in UMCF and SACF–ECF, respectively. Average stemflow fractions increase from LMRF to UMCF and ECF, whereas leaf area index (LAI) and annual evapotranspiration (ET) decrease along the same sequence. Although the data sets for UMCF (n = 3) and ECF (n = 2) are very limited, the ET from UMCF (783 ± 112 mm) and ECF (547 ± 25 mm) is distinctly lower than that from LMCF (1188 ± 239 mm, n = 9) and LMRF (1280 ± 72 mm; n = 7). Field-measured annual ‘cloud-water’ interception (CWI) totals determined with the wet-canopy water budget method (WCWB) vary widely between locations and range between 22 and 1990 mm (n = 15). Field measured values also tend to be much larger than modelled amounts of fog interception, particularly at exposed sites. This is thought to reflect a combination of potential model limitations, a mismatch between the scale at which the model was applied (1 × 1 km) and the scale of the measurements (small plots), as well as the inclusion of near-horizontal wind-driven precipitation in the WCWB-based estimate of CWI. Regional maps of modelled amounts of fog interception across the tropics are presented, showing major spatial variability. Modelled contributions by CWI make up less than 5% of total precipitation in wet areas to more than 75% in low-rainfall areas. Catchment water yields typically increase from LMRF to UMCF and SACF–ECF reflecting concurrent increases in incident precipitation and decreases in evaporative losses. The conversion of LMCF (or LMRF) to pasture likely results in substantial increases in water yield. Changes in water yield after UMCF conversion are probably modest due to trade-offs between concurrent changes in ET and CWI. General circulation model (GCM)-projected rates of climatic drying under SRES greenhouse gas scenarios to the year 2050 are considered to have a profound effect on TMCF hydrological functioning and ecology, although different GCMs produce different and sometimes opposing results. Whilst there have been substantial increases in our understanding of the hydrological processes operating in TMCF, additional research is needed to improve the quantification of occult precipitation inputs (CWI and wind-driven precipitation), and to better understand the hydrological impacts of climate- and land-use change. Copyright © 2010 John Wiley & Sons, Ltd.

Between 5 March and 10 May 2001, the performance of three types of passive fog gauges (wire harp WH, standard fog collector SC and Juvik gauge JU) was compared at a wind-exposed Puerto Rican elfin cloud forest site. The gauges were used to determine the timing and duration of fog, as well as fog-induced crown drip using gauge to canopy conversions factors. The latter were calculated as the average ratio between throughfall, and fog collected by the respective gauges during periods with fog-only and negligible evaporation loss. The fog gauges indicated very similar timing and fog duration, presumably because dense fog prevailed for 75–80% of the time whereas winds were generally sufficiently strong (4–5 m s−1) and from a nearly constant direction. Resulting gauge to canopy factors were 0·12 for WH and SC, and 0·10 for JU. Estimating fog interception by the elfin forest canopy using these conversion factors gave very similar values for the three gauge types (mean rates of 0·15 ± 0·01 mm h−1). Throughfall typically started about 5 h after the fog gauges indicated the beginning of a fog event. This time lag probably reflects the filling of the canopy storage capacity. Integrating the estimated fog interception rates over the time lag suggested an average value of 0·4 mm for the canopy storage capacity during fog-only events. The present results are expected to apply to sites with comparable wind and fog conditions. Copyright © 2010 John Wiley & Sons, Ltd.

Summary 1. Overlapping river and road networks provide a framework for studying the complex interactions between natural and human systems, with river-road intersections as focal areas of study. Roads can alter the morphology of stream channels, pose barriers to freshwater fauna, provide easy access to streams for humans and non-native species and accelerate the expansion of urban development. 2. We determined what variables control the structure of diadromous fish and shrimp communities and assessed whether particular road crossings altered community structure in north-eastern Puerto Rico. We identified 24 sites that represented a range of river and road sizes across two catchments that drain El Yunque National Forest in Puerto Rico. 3. The location of natural barriers and the size of stream pools were the most important variables for predicting six of fifteen fish and shrimp distributions. Predatory fishes were predicted to be limited to areas in the river network below large, steep waterfalls, whereas adult shrimp Atya lanipes (Atyidae) were predicted to be present above these waterfalls. The fish Awaous banana was predicted to be present in pools >11.6 m wide, whereas the shrimp Xiphocaris elongata was predicted to be present in pools

Starting in the 1950s Luna B. Leopold and M. Gordon Wolman transformed the field of geomorphology with quantitative and process-orientated studies designed to understand landscape adjustments to temporal and spatial changes in sediment supply and streamflow. By integrating fundamental science with keen observations they developed the first comprehensive and mechanistic understanding of rivers and floodplains. Their research has not only provided a quantitative framework for understanding landscape evolution, it is a cornerstone of modern water resource management and environmental impact analysis. Specific research areas included quantifying: (1) the “hydraulic geometry” of rivers; (2) the morphology and processes of rivers; (3) channel networks and the longitudinal profiles of rivers; (4) processes controlling floodplain formation; (5) the magnitude and frequency of geomorphic processes; and (6) the cycle of sedimentation in response to urban development. Much of this research was published in seven co-authored articles and in a widely used 1964 book that they co-authored with John Miller, Fluvial Processes in Geomorphology. While their contributions are synergistically linked and stem from their co-authored papers, their individual contributions are distinct and extend over 50 years. For these accomplishments, Luna Leopold and M. Gordon Wolman were awarded the 2006 Benjamin Franklin Medal in Earth and Environmental Science.

Summary Many field indicators have been used to approximate the magnitude and frequency of flows in a variety of streams and rivers, yet due to a scarcity of long-term flow records in tropical mountain streams, little to no work has been done to establish such relationships between field features and the flow regime in these environments. Furthermore, the transition between the active channel of a river and the adjacent flood zone (i.e. bankfull) is an important geomorphologic and ecological boundary, but is rarely identifiable in steep mountain channels that lack alluvial flood plains. This study (a) quantifies relationships between field indicators and flow frequency in alluvial and steepland channels in a tropical mountain stream network and (b) identifies a reference active channel boundary in these channels, based on statistically defined combinations of riparian features, that corresponds to the same flow frequency of the bankfull stage and the effective discharge in adjacent alluvial channels. The relative elevation of transitions in riparian vegetation, soil, and substrate characteristics were first surveyed at nine stream gages in and around the Luquillo Experimental Forest in Northeastern Puerto Rico. The corresponding discharge, flow frequency, and recurrence intervals associated with these features was then determined from long-term 15-min discharge records and a partial duration series analysis. Survey data indicate that mosses and short grasses dominate at a stage often inundated by sub-effective flows. Herbaceous vegetation is associated with intermediate discharges that correspond to the threshold for sediment mobilization. Near-channel woody shrubs and trees establish at elevations along the channel margin inundated by a less frequent discharge that is coincident with the effective discharge of bed load sediment transport. Our data demonstrate that in alluvial channels in the study, both the bankfull stage (as marked by a flood plain) and the channel-forming (effective) discharge are associated with the presence of fine-grained substrate and soil, and tall, mature woody vegetation. In montane reaches that lack a flood plain, a boundary that is characterized by the incipient presence of soil, woody shrubs, and trees corresponds to the same flow frequency as the bankfull discharge of nearby alluvial channels. The reference discharge based on these riparian features in steepland sites has an average exceedance probability between 0.09% and 0.30%, and a recurrence interval between 40 and 90 days. We conclude that flows with similar frequencies influence the establishment of riparian vegetation, soil development, and substrate characteristics along channel margins in similar ways. Thus, these riparian features can be used as an indicator of hydrogeomorphic site conditions to identify active channel boundaries that occur at a constant flow frequency throughout the study stream network.

The long-term response of total soil organic carbon pools (‘total SOC’, i.e. soil and dead wood) to different harvesting scenarios in even-aged northern hardwood forest stands was evaluated using two soil carbon models, CENTURY and YASSO, that were calibrated with forest plot empirical data in the Green Mountains of Vermont. Overall, 13 different harvesting scenarios that included four levels of aboveground biomass removal (20%, 40%, 60% and 90%) and four different rotation lengths (60 year, 90 year, 120 year, and No Rotation (NR)) were simulated for a 360 year period. Simulations indicate that following an initial post-harvest increase, total SOC decreases for several decades until carbon inputs into the soil pool from the re-growth are greater than losses due to decomposition. At this point total SOC begins to gradually increase until the next harvest. One consequence of this recovery pattern is that between harvests, the size of the SOC pool in a stand may change from −7 to 18% of the pre-harvest pool, depending on the soil pool considered. Over 360 years, the average annual decrease in total SOC depends on the amount of biomass removed, the rotation length, and the soil pool considered. After 360 years a stand undergoing the 90yr-40% scenario will have 15% less total SOC than a non-harvested stand. Long-term declines in total SOC greater than 10% were observed in the 60yr-60%, 60yr-90%, and 90yr-90% scenarios. Long-term declines less than 5% were observed in scenarios with 120 year rotations that remove 60% or less of the aboveground biomass. The long-term decreases simulated here for common management scenarios in this region would require intensive sampling procedures to be detectable.

Soil turnover by tree uprooting in primary and secondary forests on the island of Puerto Rico was measured in 42 study plots in the months immediately after the passage of a Category 3 hurricane. Trunk basal area explained 61% of the variability of mound volume and 53% of the variability of mound area. The proportion of uprooted trees, the number of uprooted trees, or the proportion of uprooted basal area explained 84–85% of the variation in hurricane-created mound area. These same variables explain 79–85% of the variation in mound volume. The study indicates that the soil turnover period from tree uprooting by Puerto Rican hurricanes is between 1600 and 4800 years. These rates are faster than soil turnover by landslides and background treefall in the same area and provide a useful age constraint on soil profile development and soil carbon sequestration in these dynamic landscapes.

An extensive survey and topographic analysis of five watersheds draining the Luquillo Mountains in north-eastern Puerto Rico was conducted to decouple the relative influences of lithologic and hydraulic forces in shaping the morphology of tropical montane stream channels. The Luquillo Mountains are a steep landscape composed of volcaniclastic and igneous rocks that exert a localized lithologic influence on the stream channels. However, the stream channels also experience strong hydraulic forcing due to high unit discharge in the humid rainforest environment. GIS-based topographic analysis was used to examine channel profiles, and survey data were used to analyze downstream changes in channel geometry, grain sizes, stream power, and shear stresses. Results indicate that the longitudinal profiles are generally well graded but have concavities that reflect the influence of multiple rock types and colluvial-alluvial transitions. Non-fluvial processes, such as landslides, deliver coarse boulder-sized sediment to the channels and may locally determine channel gradient and geometry. Median grain size is strongly related to drainage area and slope, and coarsens in the headwaters before fining in the downstream reaches; a pattern associated with a mid-basin transition between colluvial and fluvial processes. Downstream hydraulic geometry relationships between discharge, width and velocity (although not depth) are well developed for all watersheds. Stream power displays a mid-basin maximum in all basins, although the ratio of stream power to coarse grain size (indicative of hydraulic forcing) increases downstream. Excess dimensionless shear stress at bankfull flow wavers around the threshold for sediment mobility of the median grain size, and does not vary systematically with bankfull discharge; a common characteristic in self-forming ‘threshold’ alluvial channels. The results suggest that although there is apparent bedrock and lithologic control on local reach-scale channel morphology, strong fluvial forces acting over time have been sufficient to override boundary resistance and give rise to systematic basin-scale patterns. Copyright © 2010 John Wiley and Sons, Ltd.

The trajectory of hurricane-induced succession was evaluated in a network of forest plots measured immediately before and 3mo, 5, 10, and 15yr after the direct impact of a Category 4 hurricane. Comparisons of forest structure, composition, and aboveground nutrients pools were made through time, and between species, lifehistory groups and geomorphic settings. The hurricane reduced aboveground biomass by 50 percent, causing an immediate decrease in stem density and diversity indices among all geomorphic settings. After 15yr, basal area and aboveground biomass returned to pre-hurricane levels, while species richness, diversity indices, and stem densities exceeded pre-hurricane levels. Differences in species composition among geomorphic settings had not returned after 15yr but differences in stem densities and structure were beginning to emerge. Significant differences were observed in the nutrient concentration of the three species that comprised the most aboveground biomass, and between species categorized as secondary high-light species and primary, low-light species. Species whose abundance was negatively correlated with the mature forest dominant also had distinct nutrient concentrations. When total aboveground nutrient pools were compared over time, differences in leaf nutrients among species were hidden by similarities in wood nutrient concentrations and the biomass dominance of a few species. The observed successional trajectory indicates that changes in species composition contributed to fast recovery of aboveground biomass and nutrient pools, while the influence of geomorphic setting on species composition occurs at time scales 415yr of succession.

Anthropogenic activities have accelerated the rate of global loss of biodiversity, making it more important than ever to understand the structure of biodiversity hotspots. One current focus is the relationship between species richness and aboveground biomass (AGB) in a variety of ecosystems. Nonetheless, species diversity, evenness, rarity, or dominance represent other critical attributes of biodiversity and may have associations with AGB that are markedly different than that of species richness. Using data from large trees in four environmentally similar sites in the Luquillo Experimental Forest of Puerto Rico, we determined the shape and strength of relationships between each of five measures of biodiversity (i.e., species richness, Simpson’s diversity, Simpson’s evenness, rarity, and dominance) and AGB. We quantified these measures of biodiversity using either proportional biomass or proportional abundance as weighting factors. Three of the four sites had a unimodal relationship between species richness and AGB, with only the most mature site evincing a positive, linear relationship. The differences between the mature site and the other sites, as well as the differences between our richness‐AGB relationships and those found at other forest sites, highlight the crucial role that prior land use and severe storms have on this forest community. Although the shape and strength of relationships differed greatly among measures of biodiversity and among sites, the strongest relationships within each site were always those involving richness or evenness.

Forest soils can act as both sinks and sources for atmospheric CO 2 and therefore have an important role in the global carbon cycle. Yet the controls on forest soil organic matter content (SOM) distribution at the scale of operational land management scales within forest types are rarely quantified in detail. To identify factors that influence the spatial distribution and accumulation of SOM in forests, soils and stand composition data from 42 even-aged northern hardwood forest plots were analyzed using multiple linear regression and non-parametric statistical approaches. The analysis included three layers of SOM pools (forest floor, 0–20 cm mineral soil, and 20+ cm mineral soil) over three spatial scales (point, plot and regional). The largest amounts of total SOM (mean = 289, std dev = 70 Mg ha − 1 ) occurred in deep and well drained soils located on gently grading slopes. When soil layers were analyzed separately, the following relationships were observed: 1) highest forest floor SOM occurred under mixed species composition as opposed to stands dominated by sugar maple, 2) highest 0–20 cm mineral SOM occurred at high elevations (greater than 450 m) in moderately well drained soils, and 3) highest 20+ cm mineral SOM also occurred at high elevations and when soils were deeper. Further analysis of 0–20 cm mineral layer revealed that lower rock volume and finer soil texture resulted in higher SOM at a single point. When SOM that was predicted from models based on plot-specific attributes (soils series, slope and aspect) were compared to soil survey SOM estimates, the mean SOM values for both approaches were similar (253 and 269 Mg ha − 1 respectively). Easily identifiable characteristics such as mixed stand composition, the presence of forest floor and E horizon thickness may be used as field indicators of SOM storage. The variety of controls identified in this study should be considered when assessing soil carbon response to management options and future changes in climate.

Summary Nutrient fluxes in rainfall and throughfall were measured weekly in a mature subtropical wet forest in NE Puerto Rico over a 15-year period that included the effects of 10 named tropical storms, several prolonged dry periods, and volcanic activity in the region. Mean annual rainfall and throughfall were 3482 and 2131 mm yr −1 , respectively. Average annual rainfall and throughfall fluxes of K, Ca, Mg, Cl, Na, and SO 4 –S were similar but somewhat larger than those reported for most tropical forests. Rainfall inputs of nitrogen were comparatively low and reflect the relative isolation of the airshed. More constituents had seasonal differences in rainfall fluxes (6 out of 12) than throughfall fluxes (4 out of 12) and all volume weighted throughfall enrichment ratios calculated for the 15-year period were greater than one. However, median weekly enrichment ratios were less than 1 for sea salts and dissolved organic carbon, between 1 and 2 for Mg, Ca, SiO 2 and SO 4 –S, and greater than 10 for NH 4 –N, PO 4 –P, and K. Droughts tended to reduce enrichment ratios of cations and sea-salts, but increased enrichment ratios of NH 4 –N, PO 4 –P, and K. In the weeks following hurricanes and tropical storms, relative throughfall tended to be higher and enrichment ratios tended to be lower. Saharan dust and the activity of Caribbean volcanoes can also be detected in the time series. Nevertheless, the impacts of particular events are variable and modified by the magnitude of the event, the pre- and post-event rainfall, and the time since the previous event. Rainfall, throughfall, rainfall pH, and rainfall fluxes of seven constituents had decreasing trends over the 15-year period. However, these decreases were small, less than inter-annual and annual variations, and not considered to be ecologically significant. These long-term observations indicate that physical and biological processes associated with water passing through the canopy act to buffer internal nutrient cycles from inter-annual and seasonal variations in rainfall inputs.

This paper relates differences in flow hydraulics between a main channel (MC) and a side channel (SC) of a river to patterns of upstream migration by Neritina virginea (Neritidae: Gastropoda), a dominant diadromous snail in streams of Puerto Rico (Greater Antilles). Near-bed water velocity, snail density and shell size were measured on a weekly basis between August and December 2000 along cross-sections in a main channel (MC) and an adjacent channel (SC) under a bridge crossing of the Rio Mameyes of Northeastern Puerto Rico. Near-bed velocity and water depth were used to compute Reynolds (Re) and Froude (Fr) numbers, and to classify flows within each channel. During base flow conditions ( 1) in the MC, and non-chaotic and subcritical (Fr 100 ind m � 2 ) of relatively small snails (means.d., 6.3 � 2.8 mm) were consistently recorded in the MC. Conversely, the SC had lower mean densities ( 0.8 m s � 1 . Within the SC, they preferred the channel thalweg and depths > 30 cm. The spatial arrangement that was observed between and within the channels may be related to food resources, predation pressure or biomechanics. Characteristics of preferred upstream migration pathways of N. virginea must be accounted when building road crossings in coastal streams with diadromous fauna. Published in 2007 by John Wiley & Sons, Ltd.

We investigated the influence of recent and future land-cover changes on stream flow of a watershed northeastern Puerto Rico using hydrological modeling and simulation analysis. Monthly and average annual stream flows were compared between an agricultural period (1973–1980) and an urbanized/reforested period (1988–1995) using the revised Generalized Watershed Loading Function model. Our validated results show that a smaller proportion of rainfall became stream flows in the urbanized/forested period compared with the agricultural period, apparently because of reforestation. Sensitivity analysis of the model showed that evapotranspiration, precipitation, and curve number were the most significant factors influencing stream flow. Simulations of projected land-cover scenarios indicate that annual stream flows would increase by 9·6% in a total urbanization scenario, decrease by 3·6% in a total reforestation scenario, and decrease by 1·1% if both reforestation and urbanization continue at their current rates to 2020. An imposed hurricane event that was similar in scale to the largest recent event on the three land-cover scenarios would increase the daily stream flow by 62·1%, 68·4% and 67·1% respectively. Owing to the environmental setting of eastern Puerto Rico, where sea breezes caused by temperature differences between land surface and the ocean dominate the local climate, we suggest that managing local land-cover changes can have important consequences for water management. Copyright © 2007 John Wiley & Sons, Ltd.

Author(s): McDowell, WH; Scatena, FN; Waide, RB; Brokaw, N; Camilo, GR; Covich, AP; Crowl, TA; Gonzalez, G; Greathouse, EA; Klawinski, P; Jean Lodge, D; Lugo, AE; Pringle, CM; Richardson, BA; Richardson, MJ; Schaefer, DA; Silver, WL; Thompson, J; Vogt, DJ; Vogt, KA; Willig, MR; Woolbright, LL; Zou, X; Zimmerman, JK | Abstract: This chapter describes the geologic, geographic, and ecological context of the location of Luquillo Mountains, particularly the factors affecting the response mechanisms of terrestrial and aquatic ecosystems to disturbance. It describes the existing conditions of the physical environment, chemical environment, and the biota of the Luquillo Mountains as they respond to disturbances. It then merges the decade-long research about the Mountains with the other tropical ecosystems around the globe.

Actual evapotranspiration (aET) and related processes in tropical forests can explain 70% of the lateral global energy transport through latent heat, and therefore are very important in the redistribution of water on the Earth's surface [Mauser, M., Schadlich, S., 1998. Modelling the spatial distribution of evapotranspiration on different scales using remote sensing data. J. Hydrol. 212-213, 250-267]. Unfortunately, there are few spatial studies of these processes in tropical forests. This research integrates one Landsat Thematic Mapper (TM) image and three Moderate Resolution Imaging Spectroradiometer (MODIS) images with a hydrological model [Granger, R.J., Gray, D.M., 1989. Evaporation from natural nonsaturated surfaces. J. Hydrol. 111, 21-29] to estimate the spatial pattern of aET over the Luquillo Experimental Forest (LEF) - a tropical forest in northeastern Puerto Rico - for the month of January, the only month that these remotely sensed images were acquired. The derived aETs ranged from 0 to 7.22 mm/day with a mean of 3.08 ± 1.35 mm/day which were comparable to other estimates. Simulated aET was highest in the low elevation forest and decreased progressively toward higher elevations. Because of differences in solar radiation at different elevations, aspects and topographic positions, aET tended to be higher on south slopes and along ridges than on north slopes and in valleys. In addition, the Bowen ratio (the ratio of sensible heat to latent heat) varied across different vegetation types and increased with elevation, thus reflecting differences in the distribution of net solar radiation incident on the earth's surface. Over a day, the highest simulated aET occurred at around noon. We also applied this model to simulate the average monthly aET over an entire year based on the cloud patterns derived from at least two MODIS images for each month. The highest simulated aET occurred in February and March and the lowest in May. These observations are consistent with long term data. The simulated values were compared with field measurements of the sap flow velocity of trees at different elevations and in different forest types. These comparisons had good agreement in the low elevation forest but only moderate agreement in the elfin forest at high elevations.

We report the effects of an invasive tree (Syzygium jambos, Myrtaceace) on species composition, plant diversity patterns, and forest regeneration in primary and secondary forest in the Luquillo Mountains of northeastern Puerto Rico, including the area in and around the Caribbean National Forest (CNF) and the Luquillo Long Term Ecological Research site (Luquillo LTER). Land use history was reconstructed using aerial photographs from 1936 to 1989 and study sites were categorized into four groups that corresponded to their status in 1936: unforested, young secondary, mature secondary, and primary forests. In randomly selected forest stands in each forest type, we measured the abundance of invasive and native tree species, seedling recruitment for S. jambos as well as soil nutrient pools and tested for the effects of land use history on S. jambos density and diversity. A partial Mantel test was used to control for historical and elevational differences across study sites. The results indicate that S. jambos density was highest in habitats classified in 1936 as unforested, young, or mature secondary forests. Compared to all other forest classes, species diversity was significantly higher in primary forests. However, there was no statistically significant difference between observed and estimated species richness across the four forest types. S. jambos density and species diversity were strongly negatively correlated, even after controlling for land use history and elevation. There was significantly higher S. jambos seedling recruitment in areas that were either unforested or had young secondary forests in 1936. The results also indicate that S. jambos is able to establish viable populations in habitats with different soil nutrient status. S. jambos has also altered vegetation composition and diversity patterns in habitats where it is the dominant tree species. After nearly 185 years since its introduction to the island, S. jambos is not only well established within 30 m of stream channels, its presence does not appear to be limited by topographic, soil nutrient, or elevational conditions. This study suggests that land use change and subsequent plant invasions have produced a new vegetation assemblage that has led to potentially long-term changes in community structure, species composition, and successional trajectory in regenerating secondary forests in the Luquillo Mountains.

Diadromous faunas dominate most tropical coastal streams and rivers, but the factors controlling their distribution are not well understood. Our study documents abiotic variables controlling the distribution and abundance of the diadromous snail Neritina virginea (Gastropoda:Neritidae) in the Caribbean island of Puerto Rico. An intensive survey of N. virginea density and shell size, and channel substrate, velocity, and depth was conducted at microhabitat, habitat, and reach scales of a coastal plain reach of the Ro ´o Mameyes between August and December 2000. In addition, the inland extent of distribution (stream-network scale) and presence (regional scale) of N. virginea were surveyed in 32 coastal rivers around the island during summer 2001 and 2003. At the microhabitat scale, snail density and microhabitat electivity were greater in patches consisting of a mix of boulders and cobbles than in other types of substrate. At the habitat scale, snail density increased with depth. At the reach scale, snail density increased with fast and turbulent flows (riffle . pools . pond), whereas snail size showed the opposite pattern. At the regional scale, populations were present in 13 of 32 streams. Populations of N. virginea were not found in rivers that were disconnected from the ocean for most of the year because of channel dewatering, formation of sediment bars at their mouths, and low mean monthly discharge (Q ¼ 0.69 m 3 /s). In contrast, rivers with N. virginea populations had a permanent (Q ¼ 4.04 m 3 /s) or seasonal (Q ¼ 2.88 m 3 /s) connection to the ocean over the year. At the regional scale, the inland distribution of populations was not correlated with stream gradient, but was negatively correlated with concentrations of SiO2, P, and acid neutralizing capacity of the water. Populations colonized montane reaches in only 5 rivers, all of which were forested and protected. Our study highlights the importance of taking a hierarchical approach in managing tropical coastal rivers, and the usefulness of neritid snails as biological indicators of the physical and chemical integrity of rivers.

Programmes to promote water-related research are expanding rapidly around the globe. Whereas many have unique mission and objectives, several share common and overlapping goals and have areas of potential intersection. Identifying overlap between initiatives allows recognition of potential linkages and areas where the programmes complement each other. This allows each programme to retain its focus while ensuring that the combined programmes cover all the important issues concerning sustainable drainage-basin management. The objective of this commentary is to describe the areas of intersection between three international water research efforts and to explore synergies between them. We start by describing the evolution of the two UNESCO activities (FRIEND and HELP) and the new initiative of the International Association of Hydrological Sciences (IAHS) called PUB. We then explore possible areas of intersection among FRIEND, HELP and PUB and list the hitherto unrealized opportunities for research that would add value to each programme and, consequently, hydrological sciences and integrated drainage-basin management. We end by defining tangible activities that can be accomplished in the next 1–5 years.

The deposition of fog to a wind-exposed 3 m tall Puerto Rican cloud forest at 1010 m elevation was studied using the water budget and eddy covariance methods. Fog deposition was calculated from the water budget as throughfall plus stemflow plus interception loss minus rainfall corrected for wind-induced loss and effect of slope. The eddy covariance method was used to calculate the turbulent liquid cloud water flux from instantaneous turbulent deviations of the surface-normal wind component and cloud liquid water content as measured at 4 m above the forest canopy. Fog deposition rates according to the water budget under rain-free conditions (0! 11 " 0! 05 mm h # 1 ) and rainy conditions (0! 24 " 0! 13 mm h # 1 ) were about three to six times the eddy-covariance-based estimate (0! 04 " 0! 002 mm h # 1 ). Under rain-free conditions, water-budget-based fog deposition rates were positively correlated with horizontal fluxes of liquid cloud water (as calculated from wind speed and liquid water content data). Under rainy conditions, the correlation became very poor, presumably because of errors in the corrected rainfall amounts and very high spatial variability in throughfall. It was demonstrated that the turbulent liquid cloud water fluxes as measured at 4 m above the forest could be only $ 40% of the fluxes at the canopy level itself due to condensation of moisture in air moving upslope. Other factors, which may have contributed to the discrepancy in results obtained with the two methods, were related to effects of footprint mismatch and methodological problems with rainfall measurements under the prevailing windy conditions. Best estimates of annual fog deposition amounted to $ 770 mm year # 1 for the summit cloud forest just below the ridge top (according to the water budget method) and $ 785 mm year # 1 for the cloud forest on the lower windward slope (using the eddy-covariance-based deposition rate corrected for estimated vertical flux divergence). Copyright ! 2006 John Wiley & Sons, Ltd.

Gradients in photosynthesis (P) and respiration (R) were measured on an unperturbed portion of the Rio Mameyes, a tropical stream in the Luquillo Experimental Forest, northeastern Puerto Rico. Rates of P, which were similar to those of streams in temperate-deciduous forests, were low in the heavily canopied headwaters (

The ecologically unique forest ecosystems of the Luquillo Mountains in Eastern Puerto Rico and the scenic value of its forests, rivers and surrounding beaches have promoted population growth, tourism and light industry, thus increasing regional water demands. In light of further increases in future water demand, integrated water resources management (IWRM) initiatives are rapidly evolving in this area. In an effort to seek international collaboration and information exchange on IWRM, the Luquillo Mountains joined the Hydrology for the Environment, Life, and Policy (HELP) Programme of the United Nations Educational, Scientific, and Cultural Organization (UNESCO)/International Hydrological Programme (IHP) in 1999. This paper was prepared to document existing IWRM efforts and to promote internal discussion for further IWRM development in the region.

Carbon availability often controls soil microbial growth and there is evidence that at regional scales soil microbial biomass is positively correlated with aboveground forest litter input. We examined the influence of plant litterfall on annual variation of soil microbial biomass in control and litter-excluded plots in a tropical wet forest of Puerto Rico. We also measured soil moisture, soil temperature, and plant litterfall in these treatment plots. Aboveground plant litter input had no effect on soil microbial biomass or on its pattern of fluctuation. Monthly changes in soil microbial biomass were not synchronized with aboveground litter inputs, but rather preceeded litterfall by one month. Soil microbial biomass did not correlate with soil temperature, moisture, or rainfall. Our results suggest that changes in soil microbial biomass are not directly regulated by soil temperature, moisture, or aboveground litter input at local scales within a tropical wet forest, and there were asynchronous fluctuation between soil microbial biomass and plant litterfall. Potential mechanisms for this asynchronous fluctuation include soil microbial biomass regulation by competition for soil nutrients between microorganisms and plants, and regulation by below-ground carbon inputs associated with the annual solar and drying-rewetting cycles in tropical wet forests.

To evaluate the current status of instream flow practices in streams that drain into the Caribbean Basin, a voluntary survey of practising water resource managers was conducted. Responses were received from 70% of the potential continental countries, 100% of the islands in the Greater Antilles, and 56% of all the Caribbean island nations. Respondents identified ‘effluent discharges’, ‘downstream water quality’ and ‘existing extraction permits’ to be the most common sources of instream flow conflicts. In 75% of the countries, some type of ‘formal procedures’ exist for reviewing permit applications for freshwater extraction. In 82% of the countries, effluent discharge permits state the amount of effluent that can be discharged into a water body while only 69% require that surface water extraction permits explicitly state the quantity of water that can be extracted. In setting instream flow requirements, record low flow is used over 83% of the time. Freshwater fish were identified as the most important aquatic organism but no country ‘always’ considers the ecology or habitat requirements of aquatic species in their instream flow determinations and nearly 70% of the respondents indicated that multivariate, ecological-based methods are ‘never’ used in their country. Survey responses also indicate there is a notable lack of public involvement during the issuing of water permits. Moreover, over 80% of the countries do not provide public announcements or hearings during the permit process. In summary, this survey indicates that while there is a widespread recognition of the need for instream flows, there is a general lack of regionally based information and public involvement regarding stream flow determination. Copyright © 2003 John Wiley & Sons, Ltd.

Various complementary techniques were used to investigate the stormflow generating processes in a small headwater catchment in northeastern Puerto Rico. Over 100 samples were taken of soil matrix water, macropore flow, streamflow and precipitation, mainly during two storms of contrasting magnitude, for the analysis of calcium, magnesium, silicon, potassium, sodium and chloride. These were combined with hydrometric information on streamflow, return flow, precipitation, throughfall and soil moisture to distinguish water following different flow paths. Geo-electric sounding was used to survey the subsurface structure of the catchment, revealing a weathering front that coincided with the elevation of the stream channel instead of running parallel to surface topography. The hydrometric data were used in combination with soil physical data, a one-dimensional soil water model (VAMPS) and a three-component chemical mass-balance mixing model to describe the stormflow response of the catchment. It is inferred that most stormflow travelled through macropores in the top 20 cm of the soil profile. During a large event, saturation overland flow also accounted for a considerable portion of the stormflow, although it was not possible to quantify the associated volume fully. Although the mass-balance mixing model approach gave valuable information about the various flow paths within the catchment, it was not possible to distill the full picture from the model alone; additional hydrometric and soil physical evidence was needed to aid in the interpretation of the model results. Copyright © 2004 John Wiley & Sons, Ltd.

Multivariate analyses were used to develop equations that could predict certain water quality (WQ) conditions for unmonitored watersheds in Puerto Rico based on their physical characteristics. Long term WQ data were used to represent the WQ of 15 watersheds in Puerto Rico. A factor analysis (FA) was performed to reduce the number of chemical constituents. Cluster analysis (CA) was used to group watersheds with similar WQ characteristics. Finally, a discriminant analysis (DA) was performed to relate the WQ clusters to different physical parameters and generate predicting equations. The FA identified six factors (77 percent of variation explained): nutrients, dissolved ions, sodium and chloride, silicacious geology, red ox conditions, and discharge. From the FA, specific conductance, sodium, phosphorous, silica, and dissolved oxygen were selected to represent the WQ characteristics in the CA. The CA determined five groups of watersheds (forested, urban polluted, mixed urban/rural, forested plutonic, and limestone) with similar WQ properties. From the five WQ clusters, two categories can be observed: forested and urban watersheds. The DA found that changes in forest cover, percent of limestone, mean annual rainfall, and watershed shape factor were the most important physical features affecting the WQ of watersheds in Puerto Rico.

There are few studies that have examined the spatial variability of forest productivity over an entire tropical forested landscape. In this study, we used a spatially-explicit forest productivity model, TOPOPROD, which is based on the FOREST-BGC model, to simulate spatial patterns of gross primary productivity (GPP), net primary productivity (NPP), and respiration over the entire Luquillo Experimental Forest (LEF) in the mountains of northeastern Puerto Rico. We modeled climate variables (e.g. solar insolation, temperature, rainfall and transpiration) using a topography-based climate model, TOPOCLIM. The simulated GPP ranged from 8 to 92 t C/ha per year with a mean of 51 t C/ha per year. The simulated NPP ranged from 0.5 to 24 t C/ha per year with a mean of 9.4 t C/ha per year. The simulated plant respiration ranged from 31 to 68 with a mean of 42 t C/ha per year. Simulated GPP and respiration declined with increased elevation whereas simulated NPP increased from low to middle elevation but decreased from middle to high elevations. Statistical analyses indicate that variation in solar insolation, which decreases with increase in elevation, is the most important factor controlling the spatial variation of forest productivity in the LEF. Validation with the limited spatial empirical data indicated that our simulations overestimated GPP by 2% for a middle elevation test site, and by 43% for a mountain peak site. Our simulations also overestimated NPP in the middle elevation Colorado forest and higher elevation Dwarf forest by 32 and 36%, respectively, but underestimated NPP in the Tabonuco and Palm forests at low to middle elevations by 9–15% and 18%, respectively. Simulated GPP and NPP would decrease under CO 2 doubling as projected temperatures increase and precipitation decreases. Different forest types respond differently to potential climate change and CO 2 doubling. Comparison with other tropical forests suggests that the LEF as a whole has higher GPP (51 t C/ha per year versus 40 t C/ha per year) but lower NPP (9.4 t C/ha per year versus 11 t C/ha per year) than other tropical rain forests.

1. Long-term data on rainfall suggests that perennial rainforest streams rarely are subject to drying of riffles or pools in the wet, non-seasonal Caribbean climate of Puerto Rico. Unusually low rainfall in 1994 caused some headwater riffles to dry out completely, resulting in isolated pools, reduced pool volumes and loss of access to microhabitats by benthic invertebrates. 2. From 1992 to 1998, shrimp populations were sampled bimonthly using baited traps in six pools along 1200 m (from 305 to 480 m in altitude) of Quebrada Prieta, a second-order headwater stream in the Luquillo Experimental Forest (Caribbean National Forest). 3. Following contraction of the smaller and shallower pools in the most upstream section of the stream, mean densities of the dominant shrimp (Atya lanipes) increased from 22 to 75 shrimp m -2 of pool area during the 1994 drought year. A second common species (Xiphocaris elongata) increased from 5 to 14 shrimp m -2 . A smaller percentage of adults of both species was gravid during the drought. 4. Following the 1994 drought (1995-1998), densities of both shrimp species and reproductive activity of Atya returned to predrought (1990-1993) levels. However, the reproductive activity of Xiphocaris remained lower than in the predrought period. 5. It is suggested that prolonged droughts, even in tropical rainforest biomes, may significantly alter aquatic communities through localised crowding effects resulting from habitat contraction, and lead to prolonged decreases in reproductive output. Consequently, major alterations in aquatic populations and communities would be predicted by current climate change scenarios of decreased total rainfall and increased variability.

Hurricanes affect ecosystem processes by altering resource availability and heterogeneity, but the spatial and temporal signatures of these events on biomass and nutrient cycling processes are not well understood. We examined mass and nutrient inputs of hurricane-derived litter in six tropical forests spanning three life zones in northeastern Puerto Rico after the passage of Hurricane Georges. We then followed the decomposition of forest floor mass and nutrient dynamics over 1 year in the three forests that experienced the greatest litter inputs (moist, tabonuco, and palm forests) to assess the length of time for which litter inputs influence regeneration and nutrient cycling processes. The 36-h disturbance event had litterfall rates that ranged from 0.55 to 0.93 times annual rates among the six forests; forest floor ranged between 1.2 and 2.5 times prehurricane standing stocks. The upperelevation forest sites had the lowest nonhurricane litterfall rates and experienced the lowest hurricane litterfall and the smallest relative increase in forest floor standing stocks. In the three intensively studied forests, the forest floor returned to prehurricane values very quickly, within 2‐10 months. The palm forest had the slowest rate of decay (k 0.74 0.16 y ‐1 ), whereas the tabonuco forest and the moist forest had similar decay rates (1.04 0.12 and 1.09 0.14, respectively). In the moist forest, there were short-term increases in the concentrations of nitrogen (N), phosphorus (P), calcium (Ca), and magnesium (Mg) in litter, but in the other two forests nutrient concentrations generally decreased. The rapid disappearance of the hurricane inputs suggests that such pulses are quickly incorporated into nutrient cycles and may be one reason for the extraordinary resilience of these forests to wind disturbances.

This research used knowledge of the indigenous practice of timing nontimber forest product harvest with the full moon to demonstrate that chemicals controlling the decomposition rate of foliage fluctuate with the lunar cycle and may have developed as a result of plant-herbivore interactions. Indigenous knowledge suggests that leaves harvested during the full moon are more durable. Palm leaves harvested during the full moon had higher total C, hemicellulose, complex C and lower Ca concentrations. These chemical changes should make palm leaves less susceptible to herbivory and more durable when harvested during the full moon. This study proposes a mechanism by which plants in the tropics minimize foliage herbivory and influence the decomposition rates of senesced leaves and their durability, especially during the full moon. This research supports the need to use natural life cycles in managing forests and provides a scientific basis for an indigenous community's harvesting practice.

An analysis of hydrograph recessions and rainfall data was performed to estimate the recession constants for two watersheds in the Luquillo mountains of Puerto Rico. To account for seasonal rainfall patterns, the data were grouped into dry and wet seasons. Sets of three Master Recession Curves (MRC) per season for each watershed were developed: one using the Matching Strip Method (MS) and two using variations of the Correlation Method (CM). These variations were the envelope line (CME) and the least squares regression (CMR). Other regression based analytical expressions that consider the streamflow recession as an autoregressive or an integrated moving average process were also applied. The regression based methods performed consistently better than the graphical ones and they proved to be faster, easier, and less subjective. The recession constants from these methods were then used to estimate the time it would take the streamflow to reach the critical Q 99 flow duration. Based on this study, once the streamflow reaches Q 90 , water managers have 6 to 12 days warning before streamflow reaches critical levels.

We assessed the effects of landscape features (vegetation type and topography), season, and spatial hierarchy on the nutrient content of surface soils in the Luquillo Experimental Forest (LEF) of Puerto Rico. Considerable spatial variation characterized the soils of the LEF, and differences between replicate sites within each combination of vegetation type (tabonuco vs. palo colorado vs. dwarf vs. pasture) and topographic position (ridge vs valley) accounted for 11–60% of the total variation in soil properties. Nevertheless, mean soil properties differed significantly among vegetation types, between topographic positions, and between seasons (wet vs dry). Differences among vegetation types reflected soil properties (e.g., bulk density, soil moisture, Na, P, C, N, S) that typically are related to biological processes and inputs of water. In forests, differences between topographic positions reflected elements (e.g., Ca, Mg, K, and Al) that typically are associated with geochemical processes; however, the nutrients and elements responsible for topographic differences in dwarf forest were different from those in other forest types. In pastures, differences between topographic positions were associated with the same soil properties responsible for differences among the other vegetation types. Pastures also had reduced N levels and different soil characteristics compared to undisturbed tabonuco forest. The only soil parameter that differed significantly between seasons was soil moisture. Soils of the LEF do not support the contention that N becomes limiting with an increase in elevation, and suggest that absolute pool sizes of N and P are not responsible for the reduction in productivity with elevation.

Gastropods of the Neritinidae family exhibit an amphidromous life cycle and an impressive variability in shell coloration in Puerto Rican streams and rivers. Various nominal species have been described, but Neritina virginea [Linne 1758], N. punctulata [Lamarck 1816] and N. reclivata [Say 1822] are the only broadly reported. However, recent studies have shown that these three species are sympatric at the river scale and that species determination might be difficult due to the presence of intermediate color morphs. Individuals (8 751) were collected from ten rivers across Puerto Rico, and from various segments and habitats in Mameyes River (the most pristine island-wide) during three years (2000-2003), and they were assigned to one of seven phenotypes corresponding to nominal species and morphs (non-nominal species). The "axial lines and dots" morph corresponding to N. reclivata was the most frequent island-wide, while the patelliform N. punctulata was scant, but the only found in headwater reaches. The "yellowish large tongues" phenotype, typical of N. virginea s.s. was the most frequent in the river mouth. The frequency of secondary phenotypes varied broadly among rivers, along the rivers, and among habitats, seemly influenced by salinity and predation gradients. The occurrence of individuals with coloration shifts after predation injuries, suggests phenotypic plasticity in the three nominal species, and urges for the use of molecular markers to unravel the possible occurrence of a species complex, and to understand the genetic basis of polymorphism. The longitudinal distribution of individual sizes, population density and egg capsules suggested the adaptive value of upstream migration, possibly to avoid marine predators.

Los gasterópodos Neritinidae de Puerto Rico presentan un ciclo de vida anfídromo y exhiben polimorfismo en los patrones de coloración de la concha. Se identificó que la oviposición aumenta río arriba conforme los individuos alcanzan mayores tamaños, a pesar de que la densidad poblacional disminuye. La frecuencia de siete fenotipos previamente establecidos se analizó en diez ríos de la isla durante dos años (n=8 751 individuos). Debido a que el río Mameyes es el más conservado, se tomó como modelo para estudiar la variación a lo largo del río y entre hábitats durante tres años. El fenotipo de líneas axiales y puntos (N. virginea) fue el predominante en toda la isla y el fenotipo pateloide (N. punctulata) fue el menos frecuente. Se observaron diferencias significativas de las frecuencias de los fenotipos (principalmente en los secundarios) entre ríos, a lo largo de ellos y entre hábitats. Esta variabilidad espacial puede responder a gradientes de salinidad y depredación. Se observó plasticidad fenotípica en algunos individuos. Phenotypic variability of the shell in Neritinidae (Gastropoda: Neritimorpha) in Puerto Rican rivers. Gastropods of the Neritinidae family exhibit an amphidromous life cycle and an impressive variability in shell coloration in Puerto Rican streams and rivers. Various nominal species have been described, but Neritina virginea [Linné 1758], N. punctulata [Lamarck 1816] and N. reclivata [Say 1822] are the only broadly reported. However, recent studies have shown that these three species are sympatric at the river scale and that species determination might be difficult due to the presence of intermediate color morphs. Individuals (8 751) were collected from ten rivers across Puerto Rico, and from various segments and habitats in Mameyes River (the most pristine island-wide) during three years (2000-2003), and they were assigned to one of seven phenotypes corresponding to nominal species and morphs (non-nominal species). The “axial lines and dots” morph corresponding to N. reclivata was the most frequent island-wide, while the patelliform N. punctulata was scant, but the only found in headwater reaches. The “yellowish large tongues” phenotype, typical of N. virginea s.s. was the most frequent in the river mouth. The frequency of secondary phenotypes varied broadly among rivers, along the rivers, and among habitats, seemly influenced by salinity and predation gradients. The occurrence of individuals with coloration shifts after predation injuries, suggests phenotypic plasticity in the three nominal species, and urges for the use of molecular markers to unravel the possible occurrence of a species complex, and to understand the genetic basis of polymorphism. The longitudinal distribution of individual sizes, population density and egg capsules suggested the adaptive value of upstream migration, possibly to avoid marine predators. Rev. Biol. Trop. 62 (Suppl. 2): 53-68. Epub 2014 April 01.