Data sets by Time Period (Long-Term or Short-Term)

With the exception of the LFDP Census data (#119), which is collected every five years and the CTE Plot Treatment, collected only after trimming events (#165), we update in this website and in the LTER Network Depository, PASTA, all of our non-spatial and non-meteorological ongoing data sets EVERY 2 Years. Met data from El Verde Field Station are updated each month or after data has gone through our quality control processes (not after a month).

The total number of LUQ's token data sets is: 151 (Added to 1 that has been deprecated and 2 non-LTER hosted data set for a total of 154)
The last Record Number Used for a Data Set is 195
The total number of Long-Term LTER data sets is 69 (Ongoing: 33, Completed: 36)
The total number of Short-Term LTER data sets is 83 (Completed: 83) Note: Long-Term data sets are defined as Duration of more than 5 years

Total of your current selection is: 28

Notes: Highlight both options to obtain a complete list of both Long and Short Term Data Sets
Duration (Period of Time): Long-Term
Numeric Data Set ID Title (Data Set Identifier) Abstract Owner/Creator Duration (Period of Time) Status Date Range
16 Maximum temperature at El Verde Field Station, Rio Grande, Puerto Rico since October 1992

Daily emperature has been measured at the El Verde Field Station since 1975 (see methods). These data was divided into two data set: #16, having data from 1992 to current and #181 which has data from1975 till August 1992. Data-missing gaps where filled in by extrapolated data from other sources, making the subsequent manipulations less valuable for interpreting long term trends.

Monthly averages have been calculated. Maximum values for maximum temperature were recorded from May to October with a range from 29 to 30 and peaks of 29.7 Centigrade in October. The months of October through December show the most dramatic increase, specially December (see chart). Highest average maximum temperatures during these years were recorded in 1998 and 1999 (See chart). Max monthly temperatures appear to be increasing from this years on.

Alonso Ramirez Long-Term Ongoing 1992 to 2019
20 Chemistry of stream water from the Luquillo Mountains William H. McDowell Long-Term Ongoing
54 Shrimp populations in Quebrada Prieta (Pools 0, 8, 9, 15) (El Verde)

Freshwater shrimp from the Quebrada Prieta (a tributary to the Sonadora in the Espiritu Santu drainage, have been censused 6 times yearly since 1988. Atya lanipes, Xiphocaris elongata and Macrobrachium spp. are regularly captured and comprise the species in this data base.

Todd Crowl Long-Term Ongoing 1988 to 2016
88 Phenologies of the Tabonuco Forest trees and shrubs

These data are being used to, among other things, (1) determine the seasonality of flowering and fruiting in Tabonuco forest and test hypotheses concerning the causation of seasonality (or lack thereof), (2) test the effect of annual variation in rainfall and other climatic variables on seed and fruit production of individual species, and (3) compare the relative dispersal of species on the Luquillo Forest Dynamics Plot by applying information on the spatial distribution of canopy trees to the data on seed and fruit fall. These data also serve as background information on the flowering and fruiting of individual species.

Jess Zimmerman Long-Term Ongoing 1992 to 2016
90 Bisley Tower I Meteorological data (Bisley Tower)

Several meteorological parameters are being measured at Bisley since 1993. Correlations between elevation and stream-runoff and rainfall, elevation and air and soil temperature, and between trhoughfall and vegetation types have been found. These relationships are used inhydrologic and nutrient budgets as well as in environmental models .

Electronic sensors are placed at the field location to support other activities.

Grizelle Gonzalez Long-Term Ongoing 1993 to 2016
96 Canopy invertebrate responses to Hurricane Hugo

We study long-term responses of insects and other arthropods to various disturbances and environmental changes, as well as important roles in ecosystem processes. Five tree species in 1991, six during 1992-1995, and seven since 1997 were selected to represent early and late successional forests at El Verde. These trees were sampled in both gap and non-gap plots resulting from Hurricane Hugo.  During the past 27 years, these plots have been subjected to multiple disturbances, including Hurricane Georges in 1998 and major droughts in 1994, 1997 and 2002, in addition to several minor hurricanes and droughts.  Each of these disturbances has influenced the arthropod community.  This project is ongoing.

Timothy D. Schowalter Long-Term Ongoing 1991 to 2017
107 El Verde Grid long-term invertebrate data

The data set consists of 41 files: (1) abundance data for the walking stick, Lamponius portoricensis, (2) estimates of abundance (Minimum Number Known Alive) for 17 species of terrestrial snails, and (3) 39 files containing data on size and mark-recapture estimates of population size for the snails Caracolus caracolla and Nenia tridens from the Wet Season of 1995 to the Wet Season of 2016 (no dry season data for 1995 or after 2012). Note: MNKA estimates are not always identical to the number of individuals indicated in the Mark-Recapture data, because some individuals (e.g., those that were lost before being marked) had to be excluded from calculations for Mark-Recapture estimates of abundance.

Michael R. Willig Long-Term Ongoing 1991 to 2017
111 Litterfall in tabonuco (subtropical wet) forest in the Luquillo Experimental Forest, Puerto Rico (MRCE Litterfall data) orig

Treatments common to both sites are quarterly fertilization (macro- and micronutrients) and unmanipulated. At El Verde, a third set of plots was subject to a one time removal of litter and woody debris generated by Hurricane Hugo (September 1989).

Alonso Ramirez Long-Term Ongoing 1990 to 2017
119 Census of species, diameter and location at the Luquillo Forest Dynamics Plot (LFDP), Puerto Rico

Censuses are performed every 5 years, the last one was finished in 2012 and the quality control process in 2013.

File LFDP_CENSUS1 contains data for the Census 1 (Survey 1,2,and 3) of pre LFDP including the Tag number Species code, quadrat location and date and stem diameter D130 (diameter measured at 130 cm from the ground (DBH). It also contains the diameter as recorded for all stems in survey 1, 2 and 3. LFDP_CENSUS1a has the same structure as LFDP_census1. In LFDP_census1a file, however, the stem diameters have been calculated to allocate "missed" stems that were found in survey 2, 3 or Census 2 to either Census 1 survey 1 (stems >=10 cm D130) or Census 1 survey 3 (stems >=1, <10 cm D130). We calculated the diameter the stem would have had, if it had been recorded at the same time the quadrat it was located in was assessed, in the appropriate survey for that stem size.

To extrapolate the stem size back in time, we used the actual growth rate of that individual stem if more than one measurement was available. If only one diameter measurement was available we used the median growth rate for that species in the appropriate size class (median growthrate of stems <10 cm, or median for stems >=10, <30 cm D130). In our publications we will combine data sets LFDP_census1 and LFDP_census1a to make Census 1 and to reconstruct the forest for stems >= 10 cm D130 at the time of Hurricane Hugo. We have divided the data into two separate files to ensure that when stem diameters are compared to future censuses the diameter data in LFDP_census1a is not used to calculate growth rates. The dates in LFDP_census1a show the date at which the real diameter was measured in survey 2 or 3 and not the time that the calculated diameter (Fdiam sur1/s2/s3) represents for the quadrat in which the stem was located. Blank in the date field in LFDP_census1a means that the tree was first measured in Census 2 and the diameter given (Fdiam sur1/s2/s3) was extrapolated back in time to Census 1. The last corrections to the Census 1 data were made in May 2001. The National Science Foundation requires that data from projects it funds are posted on the web two years after any data set has been organized and "cleaned". The data from each census of the LFDP will be updated at intervals, as each survey of the LFDP shows errors in the previous data collection. After posting on the web, researchers who are not part of the project are then welcome to use the data. Given the enormous amount of time, effort and resources required to manage the LFDP, obtain these data, and ensure data accuracy, LFDP Principal Investigators request that researchers intending to use this data comply with the requests below.

Through complying with these requests we can ensure that the data are interpreted correctly, analyses are not repeated unnecessarily, beneficial collaboration between users is promoted and the Principal Investigators' investment in this project is protected. : · Submit to the LFDP PIs a short (1 page) description of how you intend to use the data; · Invite LFDP PIs to be co-authors on any publication that uses the data in a substantial way (some PIs may decline and other LFDP scientists may need to be included); · If the LFDP PIs are not co-authors, send the PIs a draft of any paper using LFDP data, so that the PIs may comment upon it; · In the methods section of any publication using LFDP data, describe that data as coming from the "Luquillo Forest Dynamics Plot, part of the Luquillo Experimental Forest Long-Term Ecological Research Program"; · Acknowledge in any publication using LFDP data the "The Luquillo Experimental Forest Long-Term Ecological Research Program, supported by the U.S. National Science Foundation, the University of Puerto Rico, and the International Institute of Tropical Forestry"; · Supply the LFDP PIs with 10 reprints of any publication using LFDP data; · Accept that the LFDP PIs can not guarantee that the LFDP data you intend to use has not already been submitted for publication or published.

Jess Zimmerman Long-Term Ongoing 1990 to 2016
127 Meteorological data from El Verde Field Station: NADP Tower

Meteorological sensors are located at the top of a 20 m tower, the NADP Tower, behind the main buildings of El Verde Field Station, 350 masl. No large trees are present near the tower. Sensors are connected to a Campbell 10X data logger, with a storage module, and downloaded every two weeks using a wireless radio connection from the laboratory to the tower. The data is compiled at the station and send to ITES once a month. The station was initiated in 1999. In a separate physical location, rainfall and maximum and minimum air temperatures are measured manually on a daily basis.

Jess Zimmerman Long-Term Ongoing 2000 to 2018
136 Bisley Watershed 3 and Quebrada Prieta Algae Monitoring

The LTER is conducting annual monitoring of Algae, Chlorophyll a, benthic organic matter, and benthic inorganic matter in order to document baseline stream characteristics. These data may be used to examine effects of disturbances such as hurricanes on stream ecosystems.

Our long-term data shows how algal primary producers respond to disturbance.  The resistance and resilience of primary producer biomass to different types of disturbance (drought and high discharge) is affected by the relative degree of top-down control by different stream macroconsumer assemblages representative of low-order streams draining LUQ.   The LUQ  stream algae dataset is unique among LTER sites because of its long-term nature and because it provides integrated quantitative measures of algal standing crop (AFDM, chlorophyll a) at a relatively large scale (1 km stream reaches) and is not just ‘snapshot’ sampling at one location.  We also sample 2 x per year with more intensive sampling  during droughts (5-6 x per year).

We found dramatically lower levels of algal standing crop during peak drought conditions in 2015.  In Prieta, algal standing crop was 16-22 fold lower in pools and 11 -18 fold lower in riffles, compared to the long-term summer average - - - assumedly due to concentrated grazing of large numbers of omnivorous shrimps in a small volume of water. In Bisley-3, algal standing crop was 5-9 fold lower in pools and 2-6 fold lower in riffles compared to the long-term (2003-14) summer average.  Inorganic sediments in pools were highly variable and ~2 fold greater in riffles compared to the long-term summer average.

We also measured extremely high spatial variability in conductivity and solute chemistry between pools in Prieta and Bisley-3 in July- August 2015 due to the lack of flow in riffles connecting pools.  Conductivity in both streams was 1.5 - 1.6 fold greater than mean conductivity over the last decade (2002-12).

Catherine Pringle Long-Term Ongoing 2003 to 2016
143 Canopy Trimming Experiment (CTE) plant seedling measurements

Seedlings were measured before and after treatments to determine how alterations in canopy openness and detritus affect seedling densities and life histories (growth, mortality, recruitment). It was predicted that plots experiencing canopy openness with detritus removed would have the quickest increase in seedlings (recruitment and growth), especially of early successional (light demanding) species, such as Cecropia. Seedlings in plots that received no canopy manipulation but had detritus deposited is predicted to have near-total mortality initially (low light and high physical inhibition due to detritus). Treatment plots experiencing both canopy manipulation and addition of detritus would initially experience near-total mortality, with some light demanding seedlings establishing and growing through detritus with time.

Jess Zimmerman Long-Term Ongoing 2003 to 2016
144 Canopy Trimming Experiment (CTE) plants greater than 1 centimeter diameter at breast height (DBH)

Plant diameters and growth measurements were taken both before and after treatments to determine how alterations in canopy openness and detritus affect measures of growth, mortality, and production. Both types of trimmed plots are predicted to have relatively quick evidence of re-sprouting (1-3 months; based on Hurricane Hugo studies), especially palms. Because essentially complete defoliation to the canopy took place in the trim plots, the large stems that were trimmed should experience delayed growth and higher mortality than those in non-trimmed plots. The understory vegetation, or otherwise those that were not trimmed (i.e., stems <10cm at dbh) should experience increased growth in both trim plots as a result of increased light. Some nutrient immobilization may occur in the detritus addition plots, but this is predicted to affect belowground processes most. Little influence on aboveground processes from detritus addition might occur and cause some stunted growth rates of stems.

Jess Zimmerman Long-Term Ongoing 2003 to 2016
146 Ferns surveys of individuals of terrestrial ferns in Canopy Trimming Experiment (CTE) plots document changes in species richness and abundance over time in response to canopy opening and/or debris deposition

Whole plot surveys of Canopy Triming Experiment (CTE) plots were done to detect changes in the number of terrestrial fern species and individuals in response to canopy opening and debris deposition.  Surveys were conducted annually prior to and after treatments.  A count of all terrestrial ferns, identified to species on the CTE plots was recorded for each subplot in January during CTE1 (2002-2010) and in the fall during CTE2 (2014-present).  The surveys document losses of individuals of shade tolerant fern species and the appearance of open canopy ferns such as the tree fern Cyathea arborea.

Joanne M. Sharpe Long-Term Ongoing 2003 to 2016
147 Meteorological data from several climate stations in the northeast section of the Luquillo Experimental Forest

Electronic sensors are placed at the field location to support other activities. Data set includes measurementes from several climate stations in the northeast section of the Luquillo Experimental Forest. These stations are surround the Bisley Experimental watersheds and the Sabana Field Station that are operated by the USFS and the USGS. The collection of data in the Sabana Field Station ended in 9/10/2009.

Grizelle Gonzalez Long-Term Ongoing 2002 to 2015
152 Canopy Trimming Experiment (CTE) Snail data

The objective of these data is to determine how green litter deposition and canopy opening associated with a hurricane independently and jointly affect population dynamics and community composition of terrestrial gastropods. Because canopy openness can be expected to increase abiotic stress on gastropods, whereas litter deposition should provide increased resources and refugia, tradeoffs can be expected.

Michael R. Willig Long-Term Ongoing 2003 to 2017
155 Canopy Trimming Experiment (CTE) Canopy invertebrate responses to disturbance

Seven tree species were selected to represent early (Cecropia, Prestoea) and late (Dacryodes, Manilkara, Sloanea) successional, and overstory (Cecropia, Dacryodes, Manilkara, Sloanea) and understory (Prestoea, Miconia, Psychotria), species in forests at El Verde. These trees were sampled in all CTE plots.

Timothy D. Schowalter Long-Term Ongoing 2004 to 2016
162 Canopy Trimming Experiment (CTE) Litterfall

This data set shows forest litter production and its response to the canopy trimming and debris addition or removal treatments. In addition, it shows patterns of interannual variation that can be related to environmental variables. In combination with data set 111 (MRCE Litterfall data), this is the long-term monitoring of leaf litter production at El Verde Research Area.  The data reported here are for mass of litterfall pooled among collection baskets placed in the inner 20x20 m measurement plots collected at two-week intervals, beginning two years before the first treatments were applied. The litter mass data are reported for litter separated by plant parts. Litterfall contributes to mobilization of organic matter and nutrients from primary producers to soil.

Alonso Ramirez Long-Term Ongoing 2002 to 2017
165 Canopy Trimming Experiment (CTE) Plot Treatments

This data set is updated every time a trim is performed on the plots. The last trim was in 2014.

The Luquillo Experimental Forest LTER program in Puerto Rico was initially focused on the understanding of the effects that two major hurricanes, Hugo in 1989 and George in 1998, had on the structure and function of a tropical mountain forest and how the biota responded to these disturbances. This focus provided insights into the key characteristics of disturbance that alter forest function over long time scales.  After several years of research through the LEF-LTER program, it became clear that one primary effect of disturbances associated with the impact of a hurricane is to redistribute organic matter from live biomass compartments to the detrital pool.  Then a combination of biotic and abiotic processes, all modified by the disturbance, contributes to the decomposition of detritus and to the subsequent fate of associated C and nutrients. These critical regulating processes define detrital dynamics and play a central role in the recovery of forest structure and function by regulating decomposition and therefore carbon and nutrient storage and flow.

Our understanding of hurricane impacts comes from measurements of the effects of naturally occurring hurricanes on tabonuco forest and comparisons with similar disturbances in other forests (Walker et al. 1991, 1996a).  Base on the evaluation of long-term measurements after the impact of hurricane Hugo and George it becomes clear that the two primary effects of hurricane disturbance are changes in microclimate and redistribution of biomass, and that the interaction of both factors propagate through the system in complex ways.  These measurements are informative but cannot tease apart the effects of various aspects of hurricane disturbance and suffer from the lack of a control or reference condition.  From this assessment the LUQ-LTER principal investigators jointly identified the need for an experimental manipulation to decouple the effect of shifts in resource availability due to redistribution of biomass and altered microclimate conditions due to canopy opening on community and ecosystem processes and forest recovery.

As a long term experiment, the CTE is also designed to help evaluating predictions regarding the effects of an increased intensity and rate of hurricane disturbance on tabonuco forest (Sanford et al. 1991) as predicted by climate change models for Caribbean hurricanes (Emmanuel 1987, Goldenberg et al. 2001).  This long-term experiment will increase the frequency of simulated hurricane effects above background levels to once every six years. The experiment will determine effects of repeated disturbance of the forest canopy and increased detrital inputs to the forest floor on germination, growth, survival, nutrient cycling, soil conditions, and trophic structure.

Jess Zimmerman Long-Term Ongoing 2004 to 2014
174 Chemistry of rainfall and throughfall from El Verde and Bisley

Rain, throughfall, and stream water are collected weekly at the LEF sites listed below. These data sets begin as early as 1983; LTER sampling began in 1988.

Rain and throughfall samples are the total catch for the week, and are exposed to field conditions for that time. No event sampling is conducted on a routine basis. Rain samples from WDEV are wet only from an automatically-closing collector that prevents any dry deposition (Aerochem Metrics NADP collector). RCEV and RCB are bulk or always-open collectors that receive dry deposition by sedimentation.

All samples are measured for pH and conductivity, and then filtered (pre-combusted Whatman GF/F glass fiber filter) prior to further analysis. From 1983-1994 samples were cooled and returned to the San Juan chemistry laboratory for analysis. During those years, samples for NH4 and NO3 analyses were refrigerated continuously until analysis. Subsamples for NH4 analysis were also preserved with 1 molar H2SO4. From 1994 on, samples for NH4 and NO3 were frozen until analysis, were not acidified, and all analyses were conducted at the University of New Hampshire.

Rain and Throughfall Sampling Sites
Descriptions of LTER LUQ rain and throughfall weekly sample chemistry data from 1988 onwards. Chemical concentrations are recorded as mg/L or mg/L as appropriate. Values below detection limits are recorded as 1/2 the detection limit.





Rain collector Bisley .


Bulk collector.


Rain collector El Verde 


Bulk collector.


Wet/dry El Verde .


Wet only collector.


Throughfall Bisley.


TF bulk 10-collector composite


Bisley gap .


TF bulk 10-collector composite

William H. McDowell Long-Term Ongoing 1983 to 2016
175 LFDP phenology plot seedlings-16 ha plot

These data on temporal and spatial patterns of seedling recruitment in conjunction with data on flower and seed rain production allow tests of hypotheses concerning causes of inter-annual variation and roles of different mechanisms in facilitating species coexistence. The data are being used to, among other things 1) quantify seasonal and inter-annual variation in reproductive output in tropical plant populations and communities; 2) analyze relationships between reproduction and a variety of hypothesized local and regional climate drivers; 3) evaluate evidence for long-term directional trends which could reflect responses to anthropogenic global change; 4) quantify spatiotemporal variation in seed arrival and seedling recruitment of individual species; 5) analyze associated evidence of recruitment limitation, life history trade-offs, density dependent recruitment and regeneration niche partitioning; and 6) evaluate how these processes interact with inter-annual variation in climate and plant reproduction.

Jess Zimmerman Long-Term Ongoing 2007 to 2016
182 Prieta stream - Discharge and water level at pool 0

This is a long-term monitoring of water discharge and water level at Prieta Stream. The monitoring point is located at Pool 0: which is the same location where we sample stream chemistry. Measurements are taken automatically with a pressure transducer: every 15 minutes. Measure are originally of pressure: which are translated to water level using the reference gauge. Using an equation provided by Dr. Fred Scatena: water level is converted into discharge. Equation: discharge (m3/sec) = 1.095 - 1.585 * stage + 0.578 * stage * stage

Alonso Ramirez Long-Term Ongoing 2006 to 2016
183 Sonadora elevational plots: long-term monitoring of air temperature

This is a long-term monitoring of air temperature at each elevation plot along the Sonadora gradient. At each plot a HOBO sensor is located close to the middle of the plot: at 1m above ground: and placed inside a radiation shield (a plastic cup). Sensors are programmed to sample and store air temperature every hour. A daily average is computed from hourly readings. Sensors are downloaded twice a year: thus blanks represent sensor malfunction: loss of battery: or memory full. Initial blanks were due to lack of enough sensors to cover the gradient

Alonso Ramirez Long-Term Ongoing 2006 to 2016
184 El Verde Field Station Air temperature from automatic sensor

This is a long-term monitoring of air temperature in the same box where we measure manual min-max temperature at El Verde Field Station. The data set is meant as a back up to the long-term manual data set. Data is measured hourly using a HOBO Pendant data logger placed inside a wooden box: under the shade of trees. The min-max thermometer that is measured manually during work days is located next to the HOBO in the same box.

Alonso Ramirez Long-Term Ongoing 2007 to 2017
185 Leaf litter collected using baskets over the middle of the channel in Prieta arm B and Gatos

This is a long-term monitoring of leaf litter inputs into Prieta Stream, arm B, and into the Gatos stream. The monitoring is conducted using litter traps similar tho those used in other LTER project. Baskets are suspended over the channel and are emptied every other week. All material is dried and identified.

Alonso Ramirez Long-Term Ongoing 2010 to 2017
190 El Yunque Chronosequence Tree Census data

The El Yunque Chronosequence plots consist of four sites, El Verde 1 (EV1), Sabana 1 (SB1), Sabana 2 (SB2), and Sabana 3 (SB3), which are located at the edges of El Yunque National Forest at sites to the south of El Verde and Sabana Field Stations.  The plots represent a range of successional stages representing areas in agriculture or recently abandoned in 1936 but reforested after 1950, and areas in agriculture or recently abandoned in 1977 and reforested since that time.  They range in size from ~0.5 to 1 ha, vary in elevation from ~150m to 550m a.s.l. and span a wide range of ages and land use histories (Table 1).

Plot Name





10,000 m2 (1 ha)

>62 yrs but < 76 yrs

~ 550m


4,625 m2 (~0.5 ha)

>62 yrs but not primary forest



6,400 m2 (~0.6 ha)

>35 yrs but < 62 yrs



4,800 m2 (~0.5 ha)

Primary forest


One of these plots (EV1) is south of El Verde Field Station, on Forest Service land just over the property boundary.  This area was in agriculture in 1936 but appeared forested in a 1950 aerial photograph, and there are differences in forest structure and species composition consistent with the known differences in land use history.  The other three Chronosequence sites are just south of the Sabana Field Station on Forest Service Land on the opposite side of the forest from El Verde.  One plot (SB2) is located in young secondary forest in an area immediately adjacent to an old teak plantation forest.  Another plot (SB1) is located in an area that was sparsely forested in 1936 and which appeared reforested in 1950.  The third plot in Sabana (SB3) is located in a patch of primary “tabonuco” (named for the abundance of this tree species) forest on a steep slope on the west side of the Sabana River.

Maria Uriarte, Jess Zimmerman Long-Term Ongoing 2011 to 2015
193 Canopy Trimming Experiment (CTE) Walking Stick Data

The objective of these data is to determine how green litter deposition and canopy opening associated with a hurricane independently and jointly affect population dynamics of walking sticks (Lamponius portoricensis).

Michael R. Willig Long-Term Ongoing 2003 to 2016
194 Habitat selection/Caracolus caracolla and other snails (2016 Ongoing) Various habitat characteristics are presented, as well as the apparency of common plant taxa at 7 heights (every 0.5 m from ground level to 3 m). The files are divided because variables measured varied by year. This is Version 2016 of these data. Michael R. Willig Long-Term Ongoing 1996 to 2016