LUQ-LTER 3:  DETAILED METHODOLOGY FOR
PROPOSED EXPERIMENTS
FEBRUARY 2002

Note: Sections and Hypotheses are the same as those in the proposal. 

Section 2.3  Measurements of Long-Term Changes in Climatic, Biotic, and Biogeochemical Characteristics Resulting from Disturbance in Tabonuco Forest

2.3.2  Disturbance experiments

Hypothesis 1 

Section 2.4  Experiments to Test how Disturbance Alters Detrital Dynamics in Tabonuco Forest

2.4.2  Experiment 1 - Canopy Trimming Experiment  

Hypothesis 2

Hypothesis 3

Hypothesis 4

2.4.3  Experiment 2 - Detritivore Functional Group Experiment 

Hypothesis 5

Hypothesis 6

Section 2.5 Experiments and Measurements Along Gradients of Climate and Species Richness

2.5.2  Variation in biotic communities and soils with elevation

Hypothesis 7

Hypothesis 8

Hypothesis 9

2.5.3  Litter and detrital dynamics along the elevation gradient of the LEF

Hypothesis 10

2.5.4  Linking litter, soils and watersheds

Hypothesis 11

References

Section 2.3  Measurements of Long-Term Changes in Climatic, Biotic, and Biogeochemical Characteristics Resulting from Disturbance in Tabonuco Forest

2.3.2  Disturbance experiments

Hypothesis 1:  The response to disturbance and the subsequent trajectories of recovery are determined by 1) location along abiotic gradients, 2) the abiotic and biotic conditions resulting from disturbance, and 3) biotic processes subsequent to disturbance.  The relative importance of these three factors will vary with the severity of the disturbance.

Design – We will continue our long-term monitoring of meteorological and hydrological variables, nutrient cycling, and vegetation and faunal surveys.  Specific methods were established in earlier proposals (see http://luq.lternet.edu/publications/) and details are provided below.  Monitoring of herbivory and inputs of green leaf material and insect frass are new to this proposal. 

Meteorology - Meteorological data will continue to be collected at multiple sites with LTER Level 3 weather stations.  Measurement programs use state-of-the-art wireless technology being implemented under a separate grant from NSF.

- Rainfall is currently monitored at El Verde, Bisley, Pico del Este, and other sites in the LEF.  Measurements are taken continuously and hourly and daily totals are stored in data loggers. 
- Air and soil temperature are monitored at El Verde, Bisley, Pico del Este, and 12 other sites in the LEF, readings are taken hourly and daily totals with Level 3 Stations. 
- Humidity, wind speed and direction are monitored at Bisley, El Verde, Pico del Este, and 6 other sites at the LEF. Continuous measurements are taken with Level 3 Stations and vertical profiles at El Verde.
- Light (PAR, total radiation, and albedo) is continuously measured at El Verde, Bisley, Pico del Este, and other sites within the LEF.

Hydrology and nutrient cycling – Our primary measurements include whole-watershed mass balances for nutrients and other ions.  Measurements will include precipitation inputs, biomass accumulation of trees, and stream exports, following methods in McDowell & Asbury (1994). 

- Stream discharge is recorded daily in 12 streams.  Stream water ions are determined weekly in Bisley, El Verde and six other streams in the LEF.  Periodic sampling is also conducted in 13 other streams. 
- Rain chemistry is determined weekly for major ions in samples from El Verde, Bisley, and Pico del Este. 
- Groundwater, soil water, soil oxygen are monitored weekly or periodically in other areas in the LEF.
- In addition, we will measure internal nutrient fluxes (throughfall, litterfall, and litter decomposition), and changes in nutrient stocks in soils and vegetation in forested plots as well as during primary succession on landslides. 
- Throughfall is measured daily and weekly at Bisley and other areas in the LEF.  Throughfall chemistry is determined weekly at Bisley.
- Litterfall is sampled every other week at Bisley, El Verde and Pico del Este.
- Coarse woody debris (CWD) distribution will be determined using methods in Harmon & Sexton (1986). 

Vegetation - In addition to continuing long-term vegetation measurements, we will initiate measurements to determine long-term patterns of herbivory. 
            -   Forest structure and composition (i.e., density, composition, biomass) is determined at Bisley, El Verde and other areas in the LEF in 1-5 year intervals. 
            -   Belowground biomass is measured in Bisley and El Verde yearly.
            -   Canopy structure and leaf area index are measured at Bisley and El Verde every 3 years. 
            -   Distribution of coarse woody debris is measured in Bisley and El Verde 3 years.
            -   Seedling dynamics is determined in Bisley and El Verde monthly to yearly.
            -   Flowering phenology is measured in Bisley and El Verde weekly to monthly. 
            -   Litterfall and litter decomposition are measured in Bisley, El Verde, Pico del Este, in several landslides, and in other areas in the LEF weekly to periodic.
            -   Landslide re-vegetation is measured in several areas in the LEF every 6 months to yearly. 
            -   Abandoned pasture re-vegetation is measured in several areas in the LEF yearly
            -   Herbivory will be monitored on an annual or quarterly basis beginning this year.  Percent area lost to herbivores in the understory will be monitored in 8 common species studied by Angulo-Sandoval and Aide (2000).  Flushing of young leaves in the understory occurs predominately in May-June at our site and almost all herbivory is suffered by young, expanding leaves.  Therefore, we will provide annual measures of herbivory by marking a number of buds in May and then measuring percent area lost one month later.
            -   Inputs of green leaf material and insect frass to the forest floor are currently being measured as part of a cross-site study with Coweeta (investigators are Schowalter, Hunter, and Lowman; see Table 2.6.1).  Based on these results, we will periodically (quarterly or biannually) place collectors (number to be determined by power analysis of data from ongoing study) in the forest understory to collect and record green leaf and frass inputs.

Fauna - Surveys of stream decapods, mollusks, birds, frogs, lizards, and arthropods are coordinated with plant surveys in the Luquillo Forest Dynamics Plot (LFDP) and Bisley.  In aquatic habitats, shrimps have been monitored since 1988 (Covich et al. 1991, 1996) and we will start monitoring algal standing crop biomass and benthic insect abundance and biomass. 

Section 2.4  Experiments to Test how Disturbance Alters Detrital Dynamics in Tabonuco Forest

2.4.2  Experiment 1 - Canopy Trimming Experiment: 

Hypothesis 2:  Short-term dynamics of key response variables after disturbance will be a function of the interaction between microclimate and detrital inputs, whereas long-term dynamics (particularly of SOM and NPP) will be a function of detrital inputs. 

Experimental Design - The Canopy Trimming Experiment will consist of four blocks of four plots (30 x 30 m).  Blocks will be identified on the basis of similar slope, soil characteristics, and forest canopy species composition; aspect will be the blocking variable.  This plot size was chosen to provide sufficient space for long-term monitoring of plot responses as well as the biotic manipulations discussed below.  This level of replication has proven sufficient to measure ecosystem responses to manipulations of detritus in previous experiments in tabonuco forest (Zimmerman et al. 1995b, Walker et al. 1996b).  Two plots within each block will have the branches of the canopy trees trimmed (from 3 m height to the tree tops) by a professional arborist to open the canopy.  The other two plots will not experience canopy manipulation and will be subject to normal hurricane frequency.  One year of pre-manipulation measurements will be conducted before initiating the experiment in year 2 of funding.  Canopy manipulations will be repeated every six years. 
            The experimental manipulations will create four treatments in a 2 x 2 factorial design in each block:

(1)      Canopy trimmed and removed biomass distributed on forest floor. 
(2)      Canopy trimmed and removed biomass eliminated. 
(3)      Canopy untrimmed with canopy biomass from a trimmed plot distributed on the forest floor. 
(4)      Canopy untrimmed and no canopy biomass added to forest floor. 

            Within each plot measurements will be made within a core (20 x 20 m) area to minimize edge effects.  The inner plots will be divided into 1 x 1 m quadrats and randomly allocated to be measured for either plant or invertebrate densities, or for soil characteristics (10 quadrats each; Guzmán-Grajales & Walker 1991, Walker et al. 1996b).  Vertebrates will be recorded using the entire core area.  Methods for measurements are detailed below.  Statistical analyses will be made using repeated measure MANOVA by block. 

Abiotic and structural measurements – We will deploy a variety of instruments within the core area of each plot to measure the air temperature and relative humidity, light, and soil temperature and moisture to determine the effect of the experimental manipulations and to record how the abiotic environment varies with time after manipulation.  The sensors will be removed from the plots during the canopy trimming to prevent damage, and replaced as soon as the trimming and/or debris removal is finished.  This will allow us to record the environmental changes as a result of the trimming and subsequent changes in the canopy over time.
          Air temperature and relative humidity sensors will be mounted 0.5 m and 3 m above the ground.  The sensors will be shielded from rain and direct sun.  Similar sensors will also be mounted 3 m above the ground, at the lower limit of canopy trimming.  Temperature probes and tensiometers, attached to dataloggers will record soil temperature and soil moisture at 5 cm and 20 cm below the soil surface. The insulation properties of soil will cause little difference in soil temperature below 5 cm and soil moisture is not likely to be affected below 20 cm (Veenedaal et. al. 1996).
           Hemispherical photographs will be taken at 0.5 m and 3 m above the ground in at least nine locations within the core area to obtain the degree of "canopy" openness at two heights.  Photographs will be taken before canopy trimming, and then repeated at 6 monthly intervals throughout the experiment.  Quantum sensors to measure the amount of photosynthetically active radiation (PAR), and pyranometers to measure total light, will be attached to data loggers to record daily totals and daily and seasonal variation. Changes in the amount of light penetration into the experimental plot as a result of canopy trimming and as the canopy re-grows over time will be measured.  These quantum sensors and pyranometers will be mounted at 0.5 m and 3 m above ground.  Data from the quantum sensors and pyranometers will be compared with sensors outside the forest in an unshaded environment.

Biotic characteristics – We will measure a variety of variables within the core area of each plot to monitor plant, soil, invertebrate and vertebrate responses to manipulations.  The inner plots will be divided into 1 x 1 m quadrats and randomly allocated to be measured for different variables.  Vertebrates will be recorded using the entire core area.  Sampling for some variables will be quarterly for one year prior to and after the canopy manipulations and annually thereafter until the next scheduled manipulations.
            Leaf area index (LAI) is measured using “in-canopy” light measurements (Welles 1990, Martens et al. 1993).
            Canopy height profiles. Using a rangefinder and a long pole for sighting, we will determine the height interval of the uppermost canopy surface above each point in 5 x 5 m grid systems within the study plots (intervals:  0-0.5, 0.5- 1.0, 1.0-1.5, 1.5-2.0, 2.0-2.5, 2.5-3.0. 3.0-4.0, 4.0-6.0, 6.0-8.0, 8.0-10.0, 10.0-12.0, 12.0-15.0, 15.0-20.0, 20.0-25.0, 25.0-30.0, and > 30.0 m above ground.  Vegetation intercepts on the vertical line are recorded with a pole held vertically and marked at 0.5 m intervals to 3.0 m.  The height interval of intercepts above 3.0 m is measured with a rangefinder.   We also measure with the rangefinder the maximum height of vegetation on the imaginary vertical line above each grid point.   More detailed description of the method is given in Brokaw and Grear (1991).
            Density of understory plants, defined as all self-supporting, woody plants ³ 1.0 m tall, as well as Heliconia spp. (the major understory herb), will be identified, marked with uniquely number tags, measured for DBH (when reaching 130 cm height), and mapped to the nearest 1.0 m using the same methods as in the Luquillo Forest Dynamics Plot (Condit 1998, Thompson et al. in press).
            Soil microbial biomass C will be determined using a modified chloroform-fumigation- incubation technique (Jenkinson and Powlson 1976b, Liu and Zou in press).  A soil sample from six cores (18.9 mm in diameter) in each quadrat will be collected to a depth of 0-0.10 m, well mixed and weighed for soil bulk density determination.  A subsample of 10 g of soil will be oven-dried at 105oC for 48 hours to determine soil moisture content.  Two sets of soil samples from each enclosure will be prepared: one as a control and the other to be fumigated.  Thirty grams of each soil sample will be weighed into a 100 mL glass beaker.  For the fumigation treatment, a beaker containing 50 mL of ethanol-free chloroform and some solid glass beads will be placed together with the soil samples into a clean glass vacuum desiccator lined with moist paper towels.  The desiccator will be evacuated until the chloroform boiled vigorously for 2 minutes.  The samples will be fumigated in the desiccator for 24 hours.  Unfumigated soil samples will be kept in another moist paper-lined desiccator.  After 24 hours, the chloroform and paper towels will be removed and the desiccator will be evacuated 10 times for 3 minutes each time to extract residual chloroform from the soil.  Fumigated and unfumigated soil samples will be inoculated with 1.0 g of unfumigated soil.  Each inoculated soil sample will be placed in a mason jar of 1 L volume in which approximately 1.0 mL of water will be added to the bottom of the jar.  A bottle containing 20 mL of 1 M NaOH will be placed into each mason jar and the soil will be then incubated in the closed mason jar at 25oC for 10 days.  Three blanks containing no soil will be also prepared.  After the incubation, 5 mL of alkali solution will be titrated to pH 7 using 1 M HCl and 2.5 mL 1 M BaCl2 with phenolphthalein as an indicator.  The same soil sample will be incubated in the mason jar with another 20 mL of 1 M NaOH for another 10 days and the flush of CO2 will be again determined by titration.
            Density of snails, herbivorous insects, and predaceous invertebrates will be estimated by conducting censuses following methods in Willig et al. (1993). We will randomly select five quadrats within each plot and conduct nighttime censuses to identify and count all specimens found within each quadrat. Care will be taken to scan the ground and all vegetation, up to a height of 2 m.
            Density of coquies and anoles will be estimated using the entire inner plot (20 x 20 m). During each census we will search visually for coquies and anoles on tree trunks, foliage and other surfaces. Censuses for coquies will be conducted during nighttime following methods in Woolbright (1991). Censuses for anoles will follow methods for ground surveys in Regan (1991).

Ecosystem processes – A variety of ecosystem processes will be measured within the core area of each plot.  Sampling for some variables will be quarterly for one year prior to and after the canopy manipulations and annually thereafter until the next scheduled manipulations. 
            Litterfall will be measured using baskets placed 1 m above the ground and distributed randomly within each treatment.  Litter samples will be collected biweekly. 
            Tree diameter increment will be determined for each plot quarterly for one year prior to and after the canopy manipulations and annually thereafter until the next scheduled manipulations. 
            Root production/turnover will be measured using sequential ingrowth cores within the core area.  Measurements will be taken quarterly for one year prior to and after the canopy manipulations and annually thereafter until the next scheduled manipulations. 
            Litter decomposition rates will be determined annually using litter from a dominant tree species common to all blocks.
            Herbivory rates will be monitored on quarterly and then annually.  Percent area lost to herbivores in the understory will be monitored in 8 common species studied by Angulo-Sandoval and Aide (2000).  Flushing of young leaves in the understory occurs predominately in May-June at our site and almost all herbivory is suffered by young, expanding leaves.  Therefore, we will provide annual measures of herbivory by marking a number of buds in May and then measuring percent area lost one month later.
            Soil C, N, P pools
will be measured quarterly and then annually. Soil nutrient pools will be measured using standard methods (Robertson et al. 1999). Soils will be sampled quantitatively to allow for calculation of bulk density, and nutrient and C content will be measured as total and exchangeable pools as appropriate. Total C, N and P will be measured on air-dried and ground samples. Total C and N will be measured using an elemental (CHN) analyzer. Total P will be analyzed colorimetrically following lithium metaborate fusion (Lajtha et al. 1999). KCl-extractable ammonium and nitrate will be measured to determine the readily available pools of N. Nutrient losses from the plots in soil solution will be measured using tension lysimeters (standard porous cup) at 40 cm depth, which have been shown to work well at this site in previous work (McDowell 1998). Nutrient flux from the plots will be estimated as the product of average soil solution chemistry and annual hydrologic flux derived from watershed-scale measurements. This method is particularly appropriate at the site as there is little seasonal variation in soil solution chemistry (McDowell 1998).
            Soil N, P mineralization rates will be measured monthly using standard methods. 
            Soil respiration will be measured monthly as carbon dioxide evolution from soil using the alkali absorption technique (Witkamp 1966, Carter 1993, Liu and Zou in press) in polyethylene chambers.  A chamber will be randomly placed on soil surface within each quadrat and inserted 10 mm deep into the soil.  A plastic bottle with a wide-mouth (21.5 mm in diameter) containing 15 mL alkali solution (1 M NaOH) will be placed inside each chamber.  After 24 hours, the bottle will be removed from chamber, closed with a lid, and then brought to laboratory. The total amount of trapped CO2 will be calculated following titrating 5 mL of the alkali solution with 1 M HCl and 2.5 mL saturated solution of 1 M BaCl2 to a phenolphthalein endpoint.  In addition, three blank bottles at each site will be kept closed during the field incubation except open them twice quickly to be exposed to ambient air at the beginning and the end of the incubation.  Soil respiration will be measured once every month at both study sites during the 13 months of the study.  These measurements will also be calibrated using LI-COR instruments.
            Soil trace gas fluxes will be measured monthly using standard chamber techniques (Holland et al. 1999). Sampling collars will be permanently installed in the plots and chambers will be placed over the collars to collect trace gases (CO2, N2O and CH4). A time series of concentrations will be sampled and flux determined from the change in concentration over time. This method has been used in a variety of previous studies at our site (e.g. Keller et al.1986; Bowden et al. 1992; McSwiney et al. 2001).
            Soil solution N, P will be measured monthly and quarterly using standard methods. 

Hypothesis 3:  The absence of invertebrate detritivores will have strong effects on detrital dynamics, retarding decomposition rates and related processes.  Microclimatic changes associated with canopy opening will reinforce these effects, but the addition of detritus will buffer the effects of canopy opening.

Experimental Design - In the context of the Canopy Trimming Experiment, we will experimentally eliminate all invertebrate detritivores (and understory herbivores) to examine microbial decomposition of litter.  Manipulations will take place within the core area of each plot, using two subplots that will be isolated in corners at a distance of 5 m from monitoring areas and other experiments. 

Manipulations within each subplot: 

(1) Invertebrate detritivores excluded - microbial decomposition only.  Subplots (2 x 2 m) will be trenched and barriers placed to prevent earthworm immigration.  Earthworms will be eliminated from subplots by electroshocking (Liu & Zou, in press).  Naphthalene will be placed at ground level in each subplot to exclude arthropod detritivores (González & Seastedt 2001). 
(2) Invertebrate and microbial decomposition.  These plots are essentially control plots within each of the experimental units of the larger canopy trimming experiment. Subplots will be trenched and barriers placed but earthworms will not be removed nor will naphthalene be applied.  These subplots will be placed in the remaining inner plot corners

           Twenty-four litterbags (12 each of two species of plant leaves) will be placed in each enclosure. The bottom layer of these bags will be made from fiberglass window screening (1 mm mesh), and the top layer of these bags will be 0.5 cm plastic mesh in order to allow the entry of larger detritivores. These bags will contain either 5 g of air-dried, fresh leaves of Cecropia schreberiana or Dacryodes excelsa (tabonuco), common fast- and slow-growing species, respectively, that have been used previously (Crowl et al. 2000). These litter bags will be replaced with 24 more bags at the conclusion of the first year of the experiment. One bag per month will be removed at random from each enclosure. Invertebrates will be extracted from the bags using a modified Macfayden extractor (Macfayden 1962) and the litter will then be oven-dried. The litter will be weighed, ground, and have its ash-free dry weight determined. Subsamples will be analyzed for nutrient content using dry combustion techniques (Jarrel et al. 1999).
           Monthly collections of invertebrates will be counted and identified to species. Attempts will also be made to assign all invertebrates to functional feeding groups. These collections will enable us to determine whether the experimental manipulation was effective and will also allow us to determine (from control plots) the degree to which litter arthropods respond to the larger canopy trimming manipulations.

Statistical Analysis – The experiments will be run for two years and will be analyzed using repeated measures MANOVA by block.  Estimates from post-hoc comparisons for particular effects will be used to develop modeling parameters for TIM (see below).

Hypothesis 4:  Presence of herbivores will significantly alter patterns of detrital processing by differentially reducing the abundances of early successional plant species.  This effect will be most pronounced under open canopy conditions.

Experimental Design - In the context of the Canopy Trimming Experiment, we will manipulate herbivores and the plant community in screened cages (2 x 2 x 3 m) to assess their effect on detrital processing.  Each plot will contain four cages, randomly located but at least 5 m from the detrital-based food web experiments to prevent interference.  The litter and soil down to 2 cm will first be cleared and homogenized between plots within the block and then redistributed in each cage to minimize variability.  The caged area also will have understory vegetation removed and will not contain tree trunks.  Crossed wooden walkways will be placed within each cage, for access with minimal disturbance to the soil.  Cage roofs will be constructed to deflect litter input from the canopy.  Thus, subsequent litter inputs will come only from and downslope of the monitoring areas and other experiments. the artificial community and any consumers that might be present.
           These plots will be placed down slope of other monitoring activities and experiments because naphthalene is typically more dense than the surrounding air and tends to move downslope, thus potentially affecting other areas beyond the experimentally manipulated areas.
            Within each cage a simplified understory community of two species will be transplanted from nearby forest.  Piper glabrescens is a common, fast-growing shrub in closed canopy forest and openings.  Manilkara bidentata is a common, slow-growing canopy tree.  Individuals of each species approximately 1 m tall will be planted within each subplot and measured for size (total branch length and leaf number) as an estimate of aboveground biomass.  We will manipulate numbers of an herbivore, the walking stick Lamponius portoricensis.  These are common herbivores in the understory of tabonuco forest and are generalists feeders that nonetheless exhibit a preference for P. glabrescens and are present throughout the year (Willig et al. 1993).  Cages (two in each plot) will be stocked with adult individuals of an average size to achieve the average forest density and sex ratio (Willig et al. 1993).  Initial stocking densities are at the forest average, because this would be the number at the time that a hurricane strikes.  Walking sticks will be surveyed in the cages every month and the numbers will be reduced or augmented to maintain experimental densities.  Two cages will contain no herbivores, to measure ecosystem processes in the absence of any consumers.  Using the methods described above (see hypothesis 2 and Table 2.4.1 in the proposal) we will monitor within cage density of snails and of herbivorous insects, litter decomposition rates, herbivory rates, and density of understory plants.  In addition we will measure soil microbial biomass nutrient (C, N, and P) pools, mineralization rates, and respiration and trace gas fluxes. 

Statistical Analysis – The experiments will be run for two years and will be analyzed using repeated measures MANOVA by block.  Estimates from post-hoc comparisons for particular effects will be used to develop modeling parameters for TIM (see below).

2.4.3        Experiment 2 - Detritivore Functional Group Experiment:  

Hypothesis 5:  Within ecosystems:  decomposition rates will be most rapid in the presence of all detritivore functional types.  The effect of excluding functional groups will vary depending on the group excluded.  

Hypothesis 6:  Between ecosystems: exclusion of analogous functional groups will have parallel effects on decomposition rates in the two ecosystems. 

Experimental Design - In terrestrial habitats we will create a series of enclosures (2 x 2 m; the design of Lawrence & Wise 2000 modified by trenching) that will limit the passage of litter invertebrates and earthworms. We will extract invertebrates from each of these plots and then recolonize them with normal densities of combinations of fragmenters and microbial grazers (see Table 2.4.2 for details). Invertebrates will initially be removed by removing all leaf litter and the upper 2 cm of the soil from each of the plots. Arthropods from this material will be extracted using Macfayden extractors (Macfayden 1968) and then subjected to chemical fumigation to ensure that no invertebrates remain. Earthworms wil be driven from the plots using electroshocking techniques. This material will then be homogenized, divided evenly among the different plots and returned after each plot has also been fumigated. Mites and collembolans are grazers, and millipedes, isopods and earthworms are fragmenters. These are the five most numerically dominant groups of detritivores in tabonuco forest litter (Pfeiffer 1996). Representative densities of these groups will be collected from undisturbed forest and placed into the appropriate enclosures at observed field densities. All enclosures will be covered with window screening to prevent passage of invertebrate detritivores. Five replicates of each treatment will be constructed, and 12 litterbags will be placed in each enclosure. The bottom layer of these bags will be made from fiberglass window screening (1 mm mesh), and the top layer of these bags will be 0.5 cm plastic mesh. This will allow the entry of larger detritivores. These bags will contain either 5 g of air-dried, fresh leaves of Cecropia schreberiana or Dacryodes excelsa (tabonuco), common fast- and slow-growing species, respectively, that have been used previously (Crowl et al. 2000). At the end of each year, a new set of litter bags will be added for the duration of the experiment.
            One bag per month will be removed at random from each enclosure. We will extract invertebrates from the bags using a modified Macfayden extractor (Macfayden 1962) and then oven-dry the litter and weigh, grind, and determine its ash-free dry weight. Subsamples will be analyzed for nutrient content using dry combustion techniques (Sollins et al. 1999). The results will allow us to determine the degree to which proposed functional groups affect rates of decomposition individually and in combination.
            In streams, we will use multiple exclusion methods at multiple spatial scales to determine the relative roles of the different functional groups of detritivores (shrimp, insects, crabs). We will use a combination of whole-pool, electrical exclusion patches, and litterbags to exclude insects, shrimps, and crabs in a nested design (Table 2.4.3). To exclude all invertebrate taxa, we will use fine mesh litterbags that prevent insect colonization. The insects-only treatment will consist of coarse mesh litterbags placed within an electrified fence. Decapods will be excluded by a combination of electricity and manual removal from fenced pools. Bags will contain 5 grams of fresh C. schreberiana or D. excelsa litter (Crowl et al. 2001), and will be removed from each replicate on weeks 2, 4, 8, 16, and 32 or until litter has been completely decomposed within any treatment plot. Upon removal, invertebrates will be extracted from the litter, oven-dried, weighed, and ground, and ash-free dry weight will be determined. Subsamples also will be analyzed for nutrient content.
Statistical Analysis - For the terrestrial component, a one-way, randomized block ANOVA with repeated measures analysis will be used with the different combinations of invertebrate taxa being the main effects (see Table 2.4.2). For the aquatic component we will employ a split-plot design to minimize the total number of blocks necessary.

Table 2.4.2.Summary of the Decomposer Functional Group Experiment.  The design for the experiment parallels the aquatic study of Crowl et al. (2001) by excluding grazers, fragmenters and both in a 2 x 2 factorial design. _____________________________________________________________________

Treatment                           Mites      Collem.   Millipedes     Isopods      Earthworms
_____________________________________________________________________

Control                                  +            +                 +                 +                   +

Fragmenter Exclusion             +            +                  -                  -                   -

Total Exclusion                       -             -                  -                  -                   -

____________________________________________________________________

Variable                                                 Time Schedule                 Sampling Unit
____________________________________________________________________

Invertebrate Density                                   Monthly                         Litter bags

Decomposition Rate                                   Monthly                         Litter bags

Soil C, N, P                                               2x per year                     Soil Cores
___________________________________________________________________

Table 2.4.3.  Summary of the Aquatic Decomposition Experiment.  The design builds on  previously reported experiments (Crowl et al. 2001, March et al. 2001) and will use a nested design with insect exclusions (fine/coarse litter bags) and macro-fauna exclusions (electric exclosures) nested within whole pool manipulations.  At the whole-pool level, shredders and filter feeders will be excluded in 2 x 2 factorial design. _______________________________________________________________________

Treatment                          Insects   Shrimp   Crabs   Shrimp & Crabs _______________________________________________________________________

Control                                  +           +           +                   +

Whole pools                          +           +/-         +/-               +/-
(decapod treatments)

Electricity                               +           +/-        +/-                +/-
(excludes decapods only)

Fine/Coarse Bags                   -             -             -                 -
(insect exclusions)

_______________________________________________________________________

Variable                                    Time Schedule             Sampling Unit   

_______________________________________________________________________

Algae                                            Weekly               Pools and electric exclosure

BOM                                            Weekly               Pools and electric exclosure

POC export                                  Weekly               Pools

Decomposition Rates                    Weekly               Pools, exclosures, litter bags

_______________________________________________________________________

Section 2.5  Experiments and Measurements Along Gradients of Climate and Species Richness

2.5.2  Variation in biotic communities and soils with elevation

Hypothesis 7:  As plant diversity declines with elevation, important functional groups will be lost from communities at higher elevations. 

Hypothesis 8:  Community structure of the vegetation will conform to the hierarchical continuum model and not the community unit or community continuum models. 

Hypothesis 9:  Diversity of the detrital community will decline with elevation and biotic communities will be less functionally redundant at higher elevations or will lack important functional groups. 

Design – We are establishing a network of Extensive and Intensive Plots in the LEF.  Extensive Plots are used to determine variation in the plant community and in soil C and nutrient pools.  Information on plant community variation will be used to locate Intensive Plots for detailed, long-term measurements (climate, rate processes), determination of detritivore community variation, and experiments designed to test how climate and litter quality affect detrital dynamics. 

Extensive Plots: plant communities and soils - Forest plots are being placed at 50 m intervals from the lower boundary of the LEF at 250 m to 1000 m in three focal watersheds: the Sonadora, the Mameyes, and the Icacos. These watersheds represent the variation in geologic parent material in the LEF. At every 50 m elevation in each watershed we randomly locate a 0.1-ha belt transect (20 x 50 m) near the main stream (outside the riparian zone) and on moderate slopes. Plots are permanently but discreetly marked with PVC pipe and metal tags a few centimeters above the ground at each corner. Within each plot we identify, map, and measure the diameter at 130 cm above ground of all self-supporting woody plants ³ 1 cm. Tags are loosely tied around the stem of each plant and allowed to rest on the ground . The field work began in September 2001 is taking about eight months.
            We have purposely decided to locate all plots on moderate slopes at all elevations.  Topography (i.e., slope, valley, ridge) greatly affects plant species composition in the Luquillo Mountains (Weaver 1991).  Our question, however, is how elevation affects composition.  Therefore we are attempting to factor out the effects of topography by sampling plants in more or less the same topographic situation at different elevations.  Moderate slopes constitute the majority of the topography in the LEF; thus sampling these slopes will give us pictures of elevational variation in forest on the major topographic class in the LEF. 
            Soil samples will be taken from these same plots to determine SOM and other soil characteristics.  In the future, forest composition will be surveyed at 5-year intervals and after hurricane disturbances in order to determine changes in community composition resulting from disturbance and long-term climate change.
            We are analyzing three characteristics of species distributions: 1) patterns of species boundaries; 2) patterns of modes of species response curves; and 3) the hierarchical structure of species distributions (Hoagland & Collins 1997, Hofer et al. 1999).

Intensive Plots: detrital communities – The Intensive Plots will be selected in four study areas from 400 m to 1000 m elevation, located to sample the relevant variation in plant communities as identified in the Extensive Plot study as well as with respect to logistic considerations.  Lowest and highest elevation plots will include existing plots in tabonuco and elfin forest; the significant increase in research effort will occur at intermediate elevations.  Plots will be replicated four times within each study area.  Within the Intensive Plots we will characterized the composition of invertebrate, fungal, and bacterial groups in litter and soil. 
            Soil microarthropods (meso and microfauna) will be extracted using a modified Merchant-Crossley high gradient extractor.  Litter fauna will be sampled using Burlese funnels (BioQuip, CA).  Macrofauna (e.g., earthworms, millipedes, and centipedes) will be hand sorted from the soil within each plot.  Dry mass of macrofauna will be obtained after fauna is rinsed with water, freeze-killed, and oven-dried at 70 oC for 72 hrs.  Fauna will be classified as Cryptostigmata, Mesostigmata, Prostigmata (suborders, Acarina), Collembola, and other groups of fauna to the level of order.  Fauna densities will be calculated per gram of dry litter or per square meter.  The activity of soil fauna will be studied by using wet pitfall traps.  The pitfalls consist of small jars filled partially with a 50% preserving solution of ethanol.  Traps will be dug into the soil (10-15 cm depth) with funnels (ca 8 cm in dia) level to the soil surface (sensu Addington and Seastedt 1999).  Plastic cover plates will be placed over the traps to ensure the funnels will not become plugged with debris. 
            Streams invertebrates will be sampled using a combination of sampling techniques, including benthic samples and emergence traps for insects and trapping for decapods. 
            Fungi, primarily white- and brown-rot fungi, will be assessed to determine diversity and abundance.  Fruiting bodies of white- and brown-rot fungi will be collected and identified, when possible, but some species can be identified based solely on mycelial characteristics.  Plots will be surveyed for fruiting bodies of basidiomycetes to capture the diversity of fungi.
            Bacterial functional groups are currently under study by W. Silver, funded through the A.W. Mellon Foundation.  We will capitalize on these studies to characterize bacterial functional groups, including some groups that can influence C processing in soils.  Copy of the proposals funded by the A.W. Mellon Foundation:  Biogeochemistry of Iron and Microbial Control of Nitrogen.

2.5.3  Litter and detrital dynamics along the elevation gradient of the LEF

Hypothesis 10:  Litter production will vary as a function of light and temperature, while litter decomposition rates will be best predicted by litter quality along the elevation gradient with climate having a smaller but significant influence.  This will result in a decoupling of litter production and decomposition along the elevation gradient 

Design – We will study litter production and litter decomposition rates along the elevation gradient using the intensive plots established in section 2.5.2.  In each plot we will conduct long-term measurements and will conduct reciprocal litter transplants to decouple litter production and decomposition along the elevation gradient. 

Long-term measurements - We will survey the Intensive Plots for plant community composition, structure, and plant chemical properties of common species within each plot.  Litter production will be sampled every two weeks from litter baskets (Vogt et al. 1996).  These will be established along the catena (geomorphic profile from ridge to ridge) in each plot so that terrestrial studies can be integrated with aquatic studies in the future.  Replicate samples of forest floor standing stocks will be collected monthly in a buffer zone outside the plots to minimize side effects.  Data from permanent weather stations located at each site will provide background information that will guide us in understanding interactions between climate and biotic components of the LEF ecosystem. 

Reciprocal litter transplant experiment - We will address the effects of climate and litter quality on terrestrial decomposition rates by factorial experiments with litter types from each of the 16 intensive plots.  Leaf litter from each plot will be decayed in situ, and litterbags will be transplanted to each of the other sites along the gradient.  We will include a common substrate (birch craft sticks, previously tested) as a control on litter quality effects. 
            To describe the potential effects of disturbance on litter decay along the gradient, and as a way of potentially altering litter chemistry within sites and species, we will include a treatment that looks at the potential priming effects of fresh leaves on decomposition rates in each forest type.  We will construct litterbags that: 1) include only green leaves (picked from a random selection of trees in the plot); 2) include only naturally senesced litter; and 3) include 40% green leaves and 60% naturally senesced litter.  This approximates the ratio of green leaves to brown litter recorded immediately after Hurricane Georges (Ostertag et al., submitted). 
            Subsamples of green leaves and senesced litter from each plot will be analyzed separately for nutrient and C content, exchangeable Al, and the concentration of secondary compounds.  Decomposition rates will be determined as k yr-1 using the best fit among exponential, double exponential or linear models (Wieder & Lang 1982).  Soil organic matter fractions, soil nutrient availability, and microbial biomass also will be measured four times per year.  The response of decay rates to climate, litter quality, and soil characteristics will be tested using analysis of variance. 
            We will determine the degree of coupling of production and decomposition along the gradient by comparing the flux of litter and nutrients to the foret floor, the changes in standing stocks of forest floor over time, and the flux of carbon and nutrients via decomposition.  These data will be compared to data on soil C and nutrient pools to arrive at a preliminary C and nutrient budget for the forest floor and surface soil layers.

2.5.4  Linking litter, soils and watersheds

Hypothesis 11:  Export of inorganic nutrients and dissolved organic matter (DOC and DON) will vary across the Luquillo Mountain landscape as a function of soil C:N.  High soil C:N will be associated with high DOC and DON losses and low inorganic N losses. 

Design - Linking watershed properties to nutrient and C flux in stream water will rely on weekly sampling of 11-gauged streams. Grab samples are taken every Tuesday, and additional high flow samples are taken periodically to provide samples from the full range of flow conditions (McDowell and Asbury 1994). Stream flux is calculated using concentration-discharge relationships and daily stream flow where discharge explains greater than 10% of variance in concentration (most elements). For those solutes that do not change concentration in response to discharge, a simple flow-weighted average is used to calculate flux (McDowell and Asbury 1994). Analysis of stream chemistry will be conducted in the Water Quality Analysis Laboratory at the University of New Hampshire, a facility directed by W. McDowell. Dissolved organic carbon and total dissolved nitrogen will be analyzed using high temperature Pt-catalyzed combustion (Cauwet 1994; Merriam et al. 1996). Nitrate will be analyzed using ion chromatography (Dionex micromembrane suppression and conductivity detection) and ammonium and phosphate will be analyzed using flow injection analysis (phenol hypochlorite and ammonium molybdate methods, respectively). Soils will be characterized for each watershed with a minimum of 20 samples, stratified by elevation. Soil C and N (elemental analyzer) and texture will be measured.
            Funds to characterize soil C:N and analysis of DOC and DON export from multiple watersheds during a single year are available from a recent NSF award to McDowell. The LTER contribution will be to establish three new, gauged, long-term sites, continue sampling of existing sites, and provide data on litter quality and soil characteristics from the Extensive Elevation Plots along the elevation gradient. Two gauged watersheds in elfin forest at 900 to 950 m elevation will be added to the set of LTER study watersheds. A tributary of the Río Icacos, the Guaba, drains a watershed ranging from 750 to 900 m elevation and is currently gauged by the USGS WEBB project.

Literature cited
Addington, R., and T. R. Seastedt. 1999. Activity of soil microarthropods beneath snowpack in alpine tundra and subalpine             forest. Pedobiologia 43:47-53.
Angulo-Sandoval, P., and T. M. Aide. 2000. Leaf phenology and leaf damage of saplings in the Luquillo Experimental Forest,             Puerto Rico. Biotropica 32:415-422.
Bowden, W. B., W. H. McDowell, C. E. Asbury, and A. M. Finley. 1992. Riparian nitrogen dynamics in two                        geomorphologically distinct tropical rain forest watersheds - nitrous oxide fluxes. Biogeochemistry, 18(2): 77-99.
Brokaw, N. V. L., and J. S. Grear. 1991. Forest structure before and after Hurricane Hugo at three elevations in the Luquillo             Mountains, Puerto Rico. Biotropica 23:386-392.
Carter, M. R. 1993. Soil sampling and methods of analysis. Lewis Publishers, Toronto.
Cauwet, G. 1994. HTCO method for dissolved organic carbon analysis in seawater - influence of catalyst on blank estimation.             Marine Chemistry, 47: 55-64.
Condit, R. 1998. Tropical forest census plots: methods and results from Barro Colorado Island, Panama, and a comparison             with other plots. Springer-Verlag, Berlin, Germany.
Covich, A. P., T. A. Crowl, S. L. Johnson, and D. L. Certain. 1991. Post-Hurricane Hugo increases in atyid shrimp                         abundances in a Puerto Rico montane stream. Biotropica 23: 448-454.
Covich, A. P., T. A. Crowl, S. L. Johnson, and M. Pyron. 1996. Distribution and abundance of tropical freshwater shrimp             along a stream corridor: response to disturbance. Biotropica 28:484-492.
Crowl, T. A., N. Bouwes, M. J. Townsend, A. P. Covich, and F. N. Scatena. 2000. Estimating the potential role of feshwater             shrimp on an aquatic insect assemblage in a tropical headwater stream: A bioenergetics approach. Verhanlungen             Internationale Vereinigung Limnologie 27: 2403-2407.
Crowl, T. A., W. H. McDowell, A. P. Covich, and S. L. Johnson. 2001. Freshwater shrimp effects on detrital processing and             nutrients in a tropical headwater stream. Ecology 82:775-783.
González, G., and T. R. Seastedt. 2001. Soil fauna and plant litter decomposition in tropical and subalpine forests. Ecology             82:955-964.
Guzmán-Grajales, S. M., and L. R. Walker. 1991. Differential seedling responses to litter after Hurricane Hugo in the Luquillo             Experimental Forest, Puerto Rico. Biotropica 23:407-413.
Harmon, M. E., and J. Sexton. 1986. Guidelines for measurements of woody detritus in forest ecosystems. LTER Network             Office web document: http://www.lternet.edu/ documents/Publications/woodydetritus/
Hoagland, B. W., and S. L. Collins. 1997. Gradient models, gradient analysis, and hierarchical structure in plant communities.             Oikos 78:23-30.
Hofer, U., L.-F. Bersier, and D. Borcard. 2000. Ecotones and gradient as determinants of herpetofaunal community structure             in the primary forest of Mount Kupe, Cameroon. Journal of Tropical Ecology 16:517-533.
Holland, E. A., G. P. Robertson, J. Greenberg, P. M. Groffman, R. D. Boone, and J. R. Gosz. 1999. Soil CO2, N2O, and             CH4 exchange. In: G.P. Robertson, D.C. Coleman, C.S. Bledsoe, and P. Sollins (Editors), Standard Soil Methods for             Long-term Ecological Research. Oxford University Press, New York.
Jenkinson, D. S., and D. S. Powlson. 1976. The effects of biocidal treatments on metabolism in soil-V. A method for                         measuring soil biomass. Soil Biology & Biochemistry 8:209-213.
Keller, M., W. A. Kaplan, and S. C. Wofsy. 1986. Emissions of N2O, CH4, and CO2 from tropical forest soils. Journal of             Geophysical Research, 91: 11,791-11,802.
Lajtha, K., C. T. Driscoll, W. M. Jarrell, and E. T. Elliott. 1999. Soil phosphorus characterization and total element analysis.             In: G.P. Robertson, D.C. Coleman, C.S. Bledsoe, and P. Sollins (Editors), Standard Soil Methods for Long-term             Ecological Research. Oxford University Press, New York, pp. 115-142.
Lawrence, K. L., and D. H. Wise. 2000. Spider predation on forest-floor Collembola and evidence for indirect effects on             decomposition. Pedobiologia 44:33-39.
Liu, Z.G., and X. M. Zou. In press. Exotic earthworms accelerate plant litter decomposition in a tropical pasture and a wet             forest of Puerto Rico. Ecological Applications.
Macfayden, A. 1962. Soil arthropod sampling. Pp. 1-24 in J.B. Cragg (ed.), Advances in ecological research. Volume 1.             Academic Press:London, U.K.
March J. G., J. P. Benstead, C. M. Pringle, and M. W. Ruebel. 2001. Linking shrimp assemblages with rates of detrital             processing along an elevational gradient in a tropical stream. Canadian Journal of Aquatic Sciences 58:470-478.
Martens, S. N., S. L. Ustin, and R. A. Rousseau. 1993. Estimation of the canopy leaf area index by gap fraction analysis.             Forest Ecology and Management 61:91-108.
McDowell, W. H., and C. E. Asbury. 1994. Export of carbon, nitrogen, and major ions from three tropical montane                         watersheds. Limnology and Oceanography 39:111-125.
McDowell, W. H., 1998. Internal nutrient fluxes in a tropical rain forest. Journal of Tropical Ecology, 14: 521-536.
McSwiney, C. P., W. H. McDowell, and M. Keller. 2001. Distribution of nitrous oxide and regulators of its production across             a tropical rainforest catena in the Luquillo Experimental Forest, Puerto Rico. Biogeochemistry 56:265-286.
Merriam, J., W. H. McDowell, and W. S. Currie. 1996. A high-temperature catalytic oxidation technique for determining total             dissolved nitrogen. Soil Science Society of America Journal. 60: 1050-1055.
Ostertag, R., F. N. Scatena, and W. L. Silver. In revision. Temporal dynamics of hurricane litter inputs and decomposition in             Puerto Rican Forests. Ecosystems.
Pfeiffer, W. J. 1996. Litter invertebrates. Pages 137-181 in D. P. Reagan and R. B. Waide, editors. The food web of a             tropical rain forest. University of Chicago Press, Chicago, IL, USA.
Reagan, D. P. 1991. The response of Anolis lizards to hurricane-induced habitat changes in a Puerto Rican forest. Biotropica             23: 468-474.
Robertson, G. P., C. S. Bledsoe, D. C. Coleman, and P. Sollins, editors. 1999. Standard soil methods for long-term ecological             research. Oxford University Press, New York, NY, USA.
Sollins, P., C. Glassman, E.A. Paul, C, Swanston, K. Lajtha, J.W. Heil and E.T. Elliott. 1999. Soil carbon and nitrogen: Pools             and fractions. Pp. 89-105 in G.P. Robertson, D.C. Coleman, C.S. Bledsoe and P. Sollins (eds.), Standard soil                         methods for long-term ecological research. Oxford University Press:New York.
Thompson, J., N. Brokaw, J. K. Zimmerman, R. B. Waide, E. M. Everham, III, D. J. Lodge, C. M. Taylor, D. García-                        Montiel, and M. Fluet. Land use history, environment, and tree composition in a tropical forest. Ecological                         Applications. In press.
Veenendaal E. M., M. D. Swaine, V. K. Agyman, D. Blay, I. K Abebrese, and C. E. Mullins 1996. Plant and soil water              relations in and around a forest gap in West Africa during the dry season. Journal of Ecology 84, 83-90.
Vogt, K. A., D. J. Vogt, P. Boon, A. Covich, F. N. Scatena, H. Asbjornsen, J. L. O'Hara, J. Pérez, T. G. Siccama, J.                         Bloomfield, and J. F. Ranciato. 1996. Litter dynamics along stream, riparian and upslope areas following Hurricane              Hugo, Luquillo Experimental Forest, Puerto Rico. Biotropica 28:458-470.
Walker, L. R., D. J. Zarin, N. Fetcher, R.W. Myster, and A.H. Johnson. 1996. Ecosystem development and plant succession             on landslides in the Caribbean. Biotropica 28:566-576.
Welles, J. M. 1990. Some indirect methods for evaluating canopy structure. Remote Sensing and Environment 5:31-43.
Wieder, R. K., and G. E. Lang. 1982. A critique of the analytical methods used in examining decomposition data obtained             from litter bags. Ecology 63:1636-1642.
Willig, M. R., E. A. Sandlin, and M. R. Gannon. 1993. Structural and taxonomic components of habitat selection in the                         Neotropical folivore, Lamponius portoricensis (Phasmatodea: Phasmatidae). Environmental Entomology 22:634-641.
Witkamp, M. 1966. Rate of CO2 evolution from the forest floor. Ecology 47:492-494.
Woolbright, L. L. 1991. The impact of hurricane Hugo on forest frogs in Puerto Rico. Biotropica 23: 462-467.
Zimmerman, J. K., W. M. Pulliam, D. J. Lodge, V. Quiñones?Orfila, N. Fetcher, S. Guzmán?Grajales, J. A. Parrotta, C. E.              Asbury, L. R. Walker, and R. B. Waide. 1995. Nitrogen immobilization by decomposing woody debris and the               recovery of tropical wet forest from hurricane damage. Oikos 72:314?322.