NPP Tropical Forest: San Carlos De Rio Negro, Venezuela, 1975-1984, R1

Summary:

This data set includes three ASCII text files containing field-measured biomass and net primary productivity (NPP) data for tropical forests at San Carlos de Rio Negro, Venezuella located along an ecosystem gradient from riverine to lateritic hill. There is one file for each sub-site: Tall Amazon Caatinga forest on coarse sandy spodosols close to river level; Bana vegetation on sandy soils less prone to flooding; and Tierra Firme mixed forest on clay oxisols of higher ground. Bioelement concentrations are also provided. Climate data are provided in two ASCII text files.

ANPP for the Tierra Firme forest is estimated at 1,590 g/m²/year, including woody biomass increment of 600 g/m²/year, while partial BNPP values range from 201 g/m² to 1,117 g/m²/year, suggesting a minimum estimate of total NPP of 1,800 to 2,700 g/m²/year. ANPP of a nearby cut-and-burned oxisol plot attained 1,940 g/m² in the fifth year following clearing. In contrast, ANPP of Tall Amazon Caatinga vegetation on the nutrient-poor spodosols is estimated at 1,150 g/m²/year, with root turnover of 120 g/m²/year, giving a minimum estimate of total NPP of 1,270 g/m²/year. NPP estimate for low Bana vegetation, based on annual litterfall accumulation plus root production, is 478 g/m²/year.

The data files for San Carlos have been revised to replace missing values and add additional primary references. Leaflitter nutrient values for the Tall Amazon Caatinga forest, and LAI, litter accumulation, and dead wood biomass in Bana forests were added. Please see the ORNL DAAC Data Set Change Information file for more information.

The Net Primary Production (NPP) data collection contains field measurements of biomass, estimated NPP, and climate data for terrestrial grassland, tropical forest, boreal forest, and tundra sites worldwide. Data were compiled from the published literature for intensively studied and well-documented individual field sites and from a number of previously compiled multi-site, multi-biome data sets of georeferenced NPP estimates. The principal compilation effort (Olson et al., 2001) was sponsored by NASA’s Terrestrial Ecology Program. For more information, please visit the NPP web site at http://daac.ornl.gov/NPP/npp_home.html.

Data and Documentation Access:  

Description and Links to Companion Files and Supplemental Information:

NPP Tropical Forest: San Carlos De Rio Negro, Venezuela, 1975-1984, R1 Data Set Change Information:
ftp://daac.ornl.gov/data/npp/tropical_forest/NPP_SCR_R1.pdf

NPP Database: Site Summary Tables for [Tropical Forests] [Link to Project page -- TBD]

Get Data:  http://daac.ornl.cgi-bin/dsviewer.pl?ds_id=479

Other Data Access Links:

Additional access to the CF compliant netCDF and GeoTIFF formatted data files is also provided through the ORNL DAAC Spatial Data Access Tool (SDAT), and through the ORNL DAAC TDS.

SDAT is an Open Geospatial Consortium (OGC) compliant tool that provides a web interface to visualize and download user selected projection, resolution, format, band, interpolation method, time period, and spatial extent. The ORNL DAAC TDS is a Thematic Real-time Environmental Distributed Data Services (THREDDS) Data Catalogue Service (TDS) that provides data through Open-source Project for a Network Data Access Protocol (OPeNDAP), OGC Web Coverage Service, and bulk Hypertext Transfer Protocol (HTTP).

Additional site ancillary data are available on the FLUXNET project web site: http://daac.ornl.gov/FLUXNET/fluxnet.html. Investigators collected data on site vegetation, soil, hydrologic, and meteorological characteristics at the flux tower sites.

Figure 1. Aerial view of the terrain near the San Carlos de Rio Negro tropical forest site, Venezuela. (Patches of dark green are caused by distinctive vegetation on Oxisol hills which contrasts with the lighter caatinga vegetation between the hills. Photograph reproduced by kind permission of Prof. C.F. Jordan, University of Georgia, U.S.A.). (SCR2-1.jpg)

Data Citation:

Cite this data set as follows:

Jordan, C.F., E. Cuevas, and E. Medina. 2009. NPP Tropical Forest: San Carlos de Rio Negro, Venezuela, 1975-1984, R[evision]1. Data set. Available on-line [http://www.daac.ornl.gov] from Oak Ridge National Laboratory Distributed Active Archive Center, Oak Ridge, Tennessee, U.S.A. doi:10.3334/ORNLDAAC/479

Table of Contents:

• 1 Data Set Overview
• 2 Data Description
• 3 Applications and Derivation
• 4 Quality Assessment
• 5 Acquisition Materials and Methods
• 6 Data Access
• 7 References

1. Data Set Overview:

Project: Net Primary Production (NPP)

The study area is a moist tropical forest near the confluence of the Casiquiare River and the Rio Negro in the Amazon Territory of Venezuela where well-differentiated vegetation-soil associations occur along a topological gradient from gently rolling hills up (< 40 m) to the surrounding lowlands and floodplains. The data set presented here contains biomass dynamics, nutrient flux, and NPP component data for three San Carlos study sites located along this forest sequence.

For the non-flooded mixed forest on oxisols (Tierra Firme forest), a minimum total NPP estimate is based on field measurements of total annual litterfall accumulation + increase in woody biomass increment + a partial determinations of below-ground NPP (root biomass increment and root production). For the nutrient-poor, less flood-prone forest on spodosols (Tall Amazon Caatinga forest), a minimum NPP estimate is based on field measurements of annual leaf and branch litterfall accumulation + increase in branch and trunk biomass increment + root production. For the low Bana forest, where there is frequent alternation of flooded and dry periods, a minimum NPP estimate is based on field measurements of annual litterfall accumulation (leaf + wood + fruit/flower) plus root production. Other field data presented in the data files, where measured, include stand height, average basal area, above- and below-ground biomass, LAI, litter decomposition rates, and nutrient concentrations in live vegetation (above- and below-ground) and in litterfall components. These studies were conducted at the Tierra Firme and Bana sites from 1975 to 1984, and at the Tall Caatinga site from 1975 to 1983.

Two ASCII files containing climate data are also provided in this data set. Monthly and annual precipitaiton amount, average temperature, and maximum/minimum temperature (and their mean values) are derived from measurements at a weather station in San Carlos village over two time periods (1975-1981 and 1950-1991).

2. Data Description:

Spatial Coverage

Site: San Carlos De Rio Negro, Venezuela

Site Boundaries: (All latitude and longitude given in degrees and fractions)

Site Information

The San Carlos study area (1.93 N 67.05 W) is situated near the confluence of the Casiquiare River and the Rio Negro in the Amazon Territory of Venezuela. It is about 4 km east of the village of San Carlos, 1,000 km south of Caracas, and close to the common border between Venezuela, Colombia, and Brazil. Annual rainfall is relatively high at 3,565 mm (based on the 1950-1981 time series in this data set). In spite of the high monthly rainfall, a clear seasonality in phenological behavior of the dominant tree species has been observed (Cuevas and Medina, 1986). Although the area is very remote and apparently free from anthropogenic disturbance, there is some evidence of burning around the year 1750.

The study area is classified as humid tropical forest (Bailey ecoregion #423). As is typical for much of the central part of the Amazon Basin, the soils near San Carlos are very low in nutrients as a result of both intensive leaching under humid tropical conditions for millions of years, and the lack of unweathered parent material as a source of nutrients. The terrain is gently rolling, with hills up to 40 m higher than the surrounding lowland. The hills and valleys are a reflection of the surface of underlying granitic bedrock. On the top of the hills, where the clay formed from the granite is exposed at or near the soil surface, the soils are classified as oxisol. The soils in the areas between the hills is comprised of coarse sands, or spodosols.

The study area consists of three sub-sites representing three well-differentiated soil types along the topological gradient in San Carlos:

• Tierra Firme mixed forest on the clay oxisols of the higher ground is characterized by species of Apocynaceae (Aspidosperma spp.), Burseracease (Protium spp.), Caryocaraceae (Caryocar spp.), Chrysobalanaceae (Licania spp.), legumes (Swartzia spp.), Moraceae (Eschweilera spp.), Palmae (Jessenia bataua, Iryanthera sp.), and Vochysiaceae (Vochysia spp.);
• Tall Amazon Caatinga forest complex on the coarse sandy spodosols between the rolling hills close to river level is characterized by highly sclerophyllous species which show particular change in tree density and species dominance along subtle changes in relief and drainage. The canopy trees have a medium stature and relatively small maximum diameter. Characteristic species include Eperua leucantha, Glycoxylon inophyllum, Hevea spp., Manilkara sp., and Micrandra spruycei;
• Bana or "low Amazon caatinga" on sandy soils less prone to shallow flooding has a low tree and shrub layer characterized by Mauritia carana (a palm), Aspidosperma album, Clusia spp., and several species of Bombacaeae (Rodognaphalopsis spp., Catostemma spp.), Rubiaceae (Calycophyllum obovatum, Remigia involucrata, Retiniphyllum spp.), and Melastomataceae (Mouriri uncitheca). The open-structured and stunted vegetation is represented by low above-ground biomass, while the composite root mass is, relative to the above-ground biomass, extremely high. During heavy rains, the soil becomes soaked with water. Since the drainage is good due to both the topographical position and the sandy nature of the soil, the soil is rapidly drained after rainfall ceases and may even become exposed to drought.

An idealized transect (Figure 2) shows variation in soils as a function of topography and how vegetation is correlated with topography, soil, and water table. In the forests on the oxisol hills, species diversity is high and there is no strong dominance of any one species. The height diameter of the trees and biomass of the forest is not particularly large. On the sides of the hills on ultisols, the trees are taller and larger in diameter, but the diversity is low, and often just one or a few species are dominant.

The vegetation occurring on the spodosols depends primarily on the depth of the water table. Where the water table is deepest, a very reduced vegetation or 'bana' occurs. With decreasing depth of the water table along the transect, bana grades into intermediate vegetation, sometimes called campina or low caatinga, and then into high caatinga. The caatinga of the Rio Negro region is called 'Amazon caatinga' to distinguish it from the caatinga of the arid region of northeastern Brazil. Amazon caatinga forests resemble the so-called heath forests, Kerangas, of Southeastern Asia.

Figure 2. Idealized transect along a gradient from river to lateritic hill. The igapo forest on the banks of the Rio Negro on the left is flooded for part of the year. Where coarse sand has been deposited, caatinga forest occurs with shallow water table and bana occurs with deeper water table. Where bedrock forms hills, as towards the right of the diagram, granite grades into clay which comprises the subsoil. In places, the clay reaches the surface, but at the study site it was covered with a shallow layer of fine sand. On the shoulders of the hills, there sometimes occurs a mixture of fine sand and clay which supports a forest dominated by one or a few species. Figure reproduced by kind permission of Dr. C.F. Jordan, UNESCO and the Parthenon Publishing Group. (SCRD-1.jpg)

The climate of San Carlos de Rio Negro is humid tropical with mean annual rainfall of 3,565 mm and mean annual temperature of 26 C (based on the 1950-1981 time series in this data set). A less rainy season (200-300 mm per month) lasts from July to March, while the real rainy season is characterized by 300-500 mm rain per month. The climate data in this data set are available for various time periods, 1950 through 1992, from a government weather station in the village of San Carlos.

Figure 3. Aerial view of the terrain near the San Carlos de Rio Negro tropical forest site, Venezuela. (A recently abandoned agricultural site is visible near the river, and a recently cut and burned site near the center of the photograph. Both sites are on an Oxisol hill. A lower area, supporting caatinga forest on Spodosol, lies to the left and bottom of the picture. Photograph reproduced by kind permission of Prof. C.F. Jordan, University of Georgia, USA). (SCR3-1.jpg)

Spatial Resolution

The Tierra Firme study area was approximately 1 ha in size. The Tall Caatinga study area was 10 ha in size (200 x 500 m). The Bana study area was 0.9 ha. See Table 1 for details about study plots.

Table 1. Spatial resolution of the study plots at San Carlos, by parameter

Notes: ¹Uhl and Jordan (1984). ²Jordan (1984). ³Stark and Spratt (1977). ⁴Cuevas and Medina (1988). ⁵Cuevas and Medina (1986). ⁶Jordan and Uhl (1978). ⁷Klinge and Herrera (1983). ⁸Jordan (1989). ⁹Bongers et al. (1985). ¹⁰From two plots in Tall Bana, three in Low Bana, and two in Open Bana. ¹¹Veillon (pers. comm. in Klinge and Herrera (1983). ¹²From two in Tall Bana, five in Low Bana, and two in Open Bana. ¹³Medina and Cuevas (1989). ¹⁴Three replicates per species, per treatment, and per collection. See Methods section for details. ¹⁵Klinge and Herrera (1978).

Temporal Coverage

NPP measurements were made at the Tierra Firme and Bana study sites from 1975 to 1984, and at the Tall Caatinga site from 1975 to 1983. See Table 2 for details.

Climate data are available for several time periods, from 1950 through 1992.

Table 2. Temporal coverage of the studies at San Carlos, by parameter

Notes: ¹Uhl and Jordan (1984). ²Jordan (1984). ³Stark and Spratt (1977). ⁴Cuevas and Medina (1988). ⁵Cuevas and Medina (1986). ⁶Jordan and Uhl (1978). ⁷Klinge & Herrera (1983). ⁸Jordan (1989). ⁹Bongers et al. (1985). ¹⁰Veillon (pers. comm. in Klinge and Herrera (1983). ¹¹Medina and Cuevas (1989). ¹²Jordan (1989). ¹³Klinge and Herrera (1978).

Temporal Resolution

NPP measurements were made on different occasions at each study location (Table 3). All NPP estimates are based on plant dry matter accumulation, expressed as g/m² (dry matter weight).

Climate data are expressed as mean monthly and annual precipitation amounts (mm), mean monthly and annual average temperature (C), and mean monthly and annual maximum/minimum temperature (C) for various time periods.

Table 3. Temporal resolution of the studies at San Carlos, by parameter

Notes: ¹Uhl and Jordan (1984). ²Jordan (1984). ³Stark and Spratt (1977). ⁴Cuevas and Medina (1988). ⁵Cuevas and Medina (1986). ⁶Jordan and Uhl (1978). ⁷Klinge & Herrera (1983). ⁸Jordan (1989). ⁹Bongers et al. (1985). ¹⁰Veillon (pers. comm. in Klinge and Herrera (1983). ¹¹Medina and Cuevas (1989). ¹²Klinge and Herrera (1978).

Data File Information

Table 4. Data files in this data set archive

NPP Data. NPP estimates for the San Carlos site are provided in three ASCII text files (Table 4), one for each sub-site. The variable values are delimited by semi-colons. The first 18 lines are metadata; data records begin on line 19. The value -999.9 is used to denote missing values. All NPP units are in g/m² (dry matter weight).

Table 5. Column headings in NPP files

Sample NPP Data Record <scr1_npp_r1.txt> (Tierra Firme Oxisol Forest)

Table 6. Parameter definitions in <scr1_npp_r1.txt> (Tierra Firme Oxisol Forest)

Notes: ¹Forest aboveground biomass values for this site, as reported in Jordan and Uhl (1978), were overestimated because of drying difficulties. Correct values appear in Table 6 of Uhl & Jordan (1984) and in this archived data file. ²Jordan & Murphy (unpublished); Jordan & Uhl (1978); and Cuevas & Medina (1986). ³From Cuevas & Medina (unpublished). ⁴From Jordan et al. (1982). ⁵From Cuevas (1983).

Sample NPP Data Record <scr2_npp_r1.txt> (Tall Amazon Caatinga)

Table 7. Parameter definitions in <scr2_npp_r1.txt> (Tall Amazon Caatinga)

Notes: ¹Based on Herrera (1979). Data from Klinge and Herrera (1978, 1983) are averages for 13 plots.

Sample NPP Data Record <scr3_npp_r1.txt> (Bana)

Table 8. Parameter definitions in <scr3_npp_r1.txt> (Bana)

Climate Data. Climate data are provided in two ASCII files (.txt format). The first 18 lines are metadata; data records begin on line 19. The variable values are delimited by semi-colons. The value -999.9 is used to denote missing values.

Sample Climate Data Record <scr1_cli.txt>

Sample Climate Data Record <scr2_cli.txt>

3. Data Application and Derivation:

The accumulation of biomass, or NPP, is the net gain of carbon by photosynthesis that remains after plant respiration. While there are many fates for this carbon, this data set accounts for a minimum estimate of NPP at each of three sites, mainly based on above-ground growth or litterfall, and below-ground root production. These are considered the major components of NPP.

Much of the research at San Carlos was conducted in order to fill gaps in ecological knowledge available to support forest management. To address these deficiencies, research projects were organized by the Center of Ecology at the Venezuelan Institute for Scientific Research (IVIC), with the participation of the Institute of Ecology of the University of Georgia (U.S.A.) and the Max-Planck Institute of Limnology (Germany), to study forest composition and structure, and nutrient cycling and conservation in Amazonian forests. Additional research was carried out under the auspices of an international UNESCO Man and the Biosphere (MAB) project.

The tropical forest biomass dynamics data for the San Carlos site are provided for comparison with models and estimation of NPP. Climate data are provided for use in driving ecosystem/NPP models.

4. Quality Assessment:

Comparisons of biomass estimations for the forests at San Carlos should take into account the fact that the methods of determining biomass are not the same in all cases. In mixed forests, estimations were based on regressions developed from felled trees in a size range class (Jordan and Uhl, 1978), while in caatinga and bana forests estimations were based on clear felling and weighing of plots of 100 m² or less (Klinge and Herrera, 1983; Bongers et al., 1985).

The above- and below-ground biomass values for San Carlos caatinga and oxisol forests are compared by Jordan (1989) and Klinge & Herrera (1983) with the biomass estimates for other tropical evergreen forests. The relatively low above-ground biomass and the relatively high below-ground biomass of the San Carlos forests suggest that both the caatinga and oxisol forests are under greater stress than many other tropical forests. Lower biomass accumulation in these forests is probably related to the frequent alternation of flooded and dry periods in the Amazon caatinga, while nutrients seem to be the most important factor in the mixed forest (Medina and Cuevas, 1989).

The above- and below-ground biomass values for San Carlos tall caatinga and bana are compared to data for a seasonal evergreen rain forest and a campina at Manaus, a cloud forest in Venezuela, and the tierra firme tropical rainforest at San Carlos de Rio Negro (Klinge and Herrera, 1978). These four forests are richer in above-ground biomass on average than the Amazon caatinga and have relative root amounts below 20%. This value can be considered normal for tropical forests. The relative amount of root biomass in the Amazon caatinga ranges from 22-33% to as high as 45% in the intergrades between caatinga and bana. The bana has a strikingly higher root percentage which may reach over 60% (62% in Klinge and Herrera, 1978).

Sources of Error

Above-ground biomass data for Tierra Firma site from Jordan and Uhl (1978) were overestimated because of drying difficulties and later revised in Uhl and Jordan (1984). Correct values appear in this archived data set.

5. Data Acquisition Materials and Methods:

Tierra Firme (Oxisol) Forest

Above-ground Biomass and Production

• Algometric Regressions and Tree Biomass (Jordan and Uhl, 1978; Uhl and Jordan, 1984; Jordan, 1989). Forty-two trees were harvested, and leaf and wood biomass were measured and regressed on diameter squared times height times density of wood. Total biomass on a hectare basis was calculated by multiplying average biomass per size class calculated from the regression times the number of trees in each diameter-height-size class per hectare, and summing the products.

In addition to the leaves collected for dry weight determinations, 50-100 leaves were taken at random from every tree and put in a plant press for later weight-area determinations.

Above-ground biomass estimates in Jordan and Uhl (1978) were overestimated because of drying difficulties. That report gave a species-weighted wood density of 0.96. Since then, Solaria (1985) has presented a corrected wood density value of 0.71and the earlier estimates were revised (Uhl and Jordan, (1984); Jordan, 1989). The correct values appear in this data set along with the overestimates, for comparison.

• Control and Experimental Plot Biomass (Uhl and Jordan, 1984; Jordan, 1989). To obtain data on entire forest structure, two 1 ha study areas containing mature forest similar in species composition, height, and above-ground biomass were selected. One site was designated as a "control" plot for studies of nutrient input and output, tree growth, and litter accumulation; the other plot was designated "experimental" for experimental treatments and studies on secondary succession following forest cutting and burning. The control plot was used to determine if any variation in productivity or nutrient dynamics during cutting and cultivation of the experimental plot were attributed to variables extrernal to the experiment. The total number of trees in each 5 cm diameter class for trees > 10 cm dia. was counted and their height estimated in the control, and the experimental plot before cutting. Trees 5-10 cm dbh were measured over transects totalling 4000 m², and trees 1-5 cm dbh over transects totalling 1525 m², in the control and experimental plots. Standing dead trees were counted separately. Putz (1983) determined liana biomass in the control forest.

• Above-ground Wood Production (Jordan, 1989).
  • Single 1981-1983 Growth Increment. A method employing a diameter tape precisely positioned on the trees > 10 cm dia. was used for a single 1981-1983 growth increment interval. Diameter increment was converted to above-ground biomass increment using allometric equations which predict biomass as a function of diameter, height and wood density (Jordan and Uhl, 1978). Height increment was determined as a function of diameter increment using Uhl's (unpublished) data on the height and diameter of 1,123 trees on the Tierra Firme site.
  • Four Year Growth Increment. The annual diameter of 120 trees on the Tierra Firme control plot (and 120 trees on the Tall Caatinga plot) were measured between 1975 and 1981 with a dendrometer for trees > 10 cm dia., and veneer calipers for trees < 10 cm dia.

• Leaf and Wood Litterfall (Jordan and Escalante (1980). Annual leaf litterfall (=leaf sloughing, in source) was determined by sampling 42 litter traps monthly for 4 years. Annual wood litterfall (=sloughing of bark and branches, in source) was determined by sampling a 2 x 800 m transect quarterly for 3 years.

• Total Litterfall and Litterfall Nutrient Content (Cuevas and Medina, 1986; Medina and Cuevas, 1989). Ten 0.5 m wire mesh baskets 15 cm above the forest floor were distributed over 400 m² plots in the Tierra Firme site (as well as in Tall Caatinga and Bana sites). Accumulated litter was collected weekly for a period of 57 weeks. Frequent collections were carried out in order to reduce nutrient losses due to leaching and decomposition in the basket. Collected litter was oven-dried at 60 C for a minimum of 5 days and later separated into leaves, flowers + fruits, and small wood fractions. Ten replicates of the leaf fraction per collection were analyzed for N, P, Ca, Mg, and K concentrations.

• Monthly and Annual Litterfall (Jordan, 1989). The rate of fall of leaves and of twigs < 1 cm dia. was determined from monthly collections in 42 plastic receptacles (32.5 cm x 37.7 cm x 20 cm deep) between 1975 and 1980 in the Tierra Firme (as well as in the Tall Caatinga site). The average monthly and annual dry weight leaf + twig litterfall accumulation is given in this data set.

• Leaf Litter (Jordan 1989; Medina and Cuevas, 1989). Dry mass of leaf and twig litter on the soil surface was estimated by collecting this material in six randomly located 0.25 m² plots in the pre-burn experimental mature forest. Subsamples were taken from three plots for nutrient analysis.

• Litter and Deadwood (Uhl and Jordan, 1984). Litter biomass and dead wood biomass data are from the pre-burn experimental mature forest (C. F. Jordan, pers. comm.; subsequently published in Jordan, 1989). See Nutrients in Plant Biomass section.

• Leaf Area Index (Jordan and Uhl, 1978). In addition to the leaves collected for dry weight determinations, 50-100 leaves were taken at random from every tree and pressed for weight-area determinations. The pressed leaves were spread on top of light sensitive paper which was then exposed. The leaf images were then cut and the weight compared to an uncut sheet of paper of known area, and from this, leaf area was calculated.

Below-ground Biomass and Production

• Pre-burn Forest Root Biomass (Uhl and Jordan, 1984). Root biomass values are the sum of below-ground root mass measured in soil pits in the mature forest site (see Stark and Spratt 1977 below) and above-ground root mass measured in the pre-burn "experimental" site (C. F. Jordan, pers. comm.; Jordan, 1989).

• Below-ground Root Biomass (including roots in both sand and clay horizons; Stark and Spratt, 1977). In order to not to disturb the Jordan (1989) control and experimental areas, Stark and Spratt (1977) studied root biomass on two additional half-hectare plots on two nearby oxisol hills. Roots located below 50 x 50 cm rigid quadrats (0.25 m²) were removed from 18 soil pits to the surface of the mineral soil, to the bottom of the sand layer, and to 40-50 cm depth into the oxisol. Each horizon was weighed wet in the field and subsampled for moisture content and fine roots. The roots and other organic matter were separated by hand and weighed separately. The root samples were air dried, shaken lightly to remove sand and day, and then dried at around 100 C for 48 hours. The dried roots were re-weighed.

• Root Growth Experiments (Jordan and Escalante, 1980). Root growth in the surficial root mat was measured by determining the rate at which roots grew into fresh leaf and wood litter placed on the surface of the root mat for 9 months (experiment 1), 10-month root growth through 1600 cm² screens (with litter placed on top of screens to depth equal to the depth of litter in the surrounding area; experiment 4), and lateral root growth into pits cleared of the root mat (root growth into litter which accumulated over 30 months in the 40 cm on a side x 40 m deep pits; experiment 6). The results in this data set represent the total average root growth from experiments 1, 4, and 6.

• Root Production in Ingrowth Cylinders (Cuevas and Medina, 1988). Plastic cylinders with 4 mm² mesh (10 cm tall x 7.5 cm diameter) filled with medium-sized vermiculite were used to measure fine root production. The experiments were conducted twice at the Tierra Firme and Tall Caatinga sites and once at the Low Bana site. The first experiment was carried out at the three sites from September 1980 until January 1981, and the second experiment was carried out in Tierra Firme from May until August 1981, and in Tall Caatinga from May until November 1981. A total of 128 cylinders per site were used in the first experiment, 32 per nutrient treatment plus control. Half the cylinders were placed within the root mat in Tierra Firme or on top of the soil in Tall Caatinga and Low Bana. The remaining cylinders were inserted in the top 10 cm of soil. Eight cylinders were collected per treatment (4 from above and 4 from within the soil) approximately every 30 days. During collection, roots that had penetrated the cylinders were carefully cut before lifting. Within two days after collection the cylinders were transported to the laboratory in Caracas and frozen until further analysis. For root separation, cylinders were thawed, and the vermiculite plus roots taken out for processing under a 10 x magnifying lens. The roots were dried in a ventilated oven at 60 C for a minimum of 72 h, weighed, and stored for chemical analyses. The second experiment was carried out in a similar manner except that the replicate number increased from 4 to 6. To calculate root production rates, root biomass accumulated inside the cylinders was divided by the basal area of the cylinder. This production values correspond only to the root mat and the top 10 cm of soil (measured by cylinder height), so it is only a fraction of fine root production in the whole profile.

Nutrients in Plant Biomass

• Pre-burn Experimental Forest Biomass Nutrients (Uhl and Jordan, 1984; Jordan, 1989). Subsamples were taken from 42 harvested trees in the Jordan (1989) experimental plot. Cross sectional samples of the trunk near the base and at the first branch were subsampled for bark, sapwood, and heartwood. Two cross sections of one large branch from each tree were similarly sampled. Approximately one kg of leaves and one kg of twigs (branches above the lowermost leaf) also were collected by sampling approximately equal amounts from the top, middle, and lower canopy. For roots, 12 subsamples were taken from the roots from each of the 18 soil pits analyzed by Stark and Spratt (1977). Cores taken from the root mat were used for litter (partially decomposed leaves and fine humus) and fine root analysis. Samples of every fifth decomposing log encountered during the survey of fallen log biomass were taken for analysis of nutrients in dead wood.

• Nutrient Dynamics in Fine Roots and Leaf Litter (Cuevas and Medina, 1986; Cuevas and Medina, 1988; and Medina and Cuevas, 1989). Root samples from the ingrowth cylinder experiment (Cuevas and Medina, 1988) and leaf litter samples from litterfall collections (Cuevas and Medina, 1986) were analyzed for nutrient content using standard laboratory methods. Leaves collected for the leaf decomposition experiment were also analyzed for nutrient content (i.e., the initial nutrient content of freshly dried leaves) (Cuevas and Medina, 1988).

Litter Decomposition (Jordan, 1989; Cuevas and Medina, 1988; and Medina and Cuevas, 1989). Adult leaves from three dominant species (Caryocar glabrum, Aspidosperma megalocarpum and Lieania heteromorpha) were collected and dried without petioles at the field station in a ventilated oven at 60 C during 3-4 days. Around 10 g of leaves of each species were placed in 20 x 20 cm plastic bags with 4 mm² mesh (same mesh as the root cylinders), and individually tagged. The mesh size was chosen to allow fine roots characteristic of the root mat and the upper soil layer to enter the bags. Initial chemical composition was measured on all species. Groups of bags for each of the species were exposed to one of three treatments: (1) left undisturbed on top of the intact soil (root mat + litter) to allow root penetration and attachment, (2) placed on top of the intact soil and lifted weekly in order to prevent root penetration, and (3) put on a 1 m² wire basket with 1 cm² mesh elevated 15 cm from the soil surface. There were three replicates per species, per treatment, and per collection. The bags were randomly placed in areas approximately 30 x 30 m. Replicates were collected at 14, 32, 71, 132, 242, and 482 days. Bags and their contents were oven-dried at 60 C. Leaves and roots were separated and weighed. Decomposition was analyzed following an exponential model. See Cuevas and Medina (1988) for details.

Tall Amazon Caatinga

Above-ground Biomass, Production, and Nutrients

• Leaf Biomass (Klinge and Herrera, 1978; 1983). The leaf biomass estimate is the fraction of leaves in total above-ground phytomass of trees > 1 cm dbh based on harvest methods described for Tierra Firme below.

• Above-ground Phytomass (Klinge and Herrera, 1978; 1983). The 200 m x 500 m study site was marked with a 10 x 10 m reference grid and 13 plots phytomass plots were selected exclusively in the tropaquod area which had ben inventoried by Bruinig (1980) and delineated. The plots were subjectively distributed, in order to cover a wide range of the vegetation along the topographical gradient from Tall Amazon Caatinga to Bana woodland. Trees, palm-trees and climbers > 1 cm dbh were harvested individually. Boles were cut at the forest floor and total length, bole length, and maximum crown diameter were measured after felling. Branches and twigs, and roots when pulled out, were severed from the bole. Leafy twigs were separated from the branches, and foliage was stripped off by hand. Each fraction was weighed separately. Plants below 1 cm in dbh were pulled out and counted separately for each plot; their phytomass was separated into foliage, shoot and root and weighed. Epiphytes (aroids, bromeliads, orchids), parasites (Loranthaceae),
  bryophytes (growing mostly on the lower portions of the boles), and Selaginella sp. of the forest floor were separately harvested and weighed. Cross sections of boles > 5 cm dbh were taken for the estimation of sapwood and hardwood area. Cross sections of a few selected species were studied anatomically. Standing dead wood was harvested and weighed. Dead branches and boles at the forest floor in each plot were heaped and weighed. Fine litter was collected under an area of 50 cm x 50 cm, in five localities distributed randomly in each plot before phytomass harvesting began. All phytomass fractions were sampled. The samples were ground and oven dried at 105 C. The moisture percentages obtained were used for the conversion of fresh weight to dry weight. Altogether 651 individuals (629 trees and 22 lianas) were sampled for the estimation of dry weight and nutrients, while fresh weight data were obtained for additional 849 individuals.

• Litterfall and Litterfall Nutrient Content (Cuevas and Medina, 1986). Methods same as for Tierra Firme site.

• Leaf Litterfall (Klinge and Herrera, 1983; Medina and Cuevas, 1989). Klinge and Herrera, 1983 data cannot be verified; Methods not indicated. CHECK Klinge and Herrera, 1978 CHECK Proctor

• Monthly and Annual Litterfall (Jordan, 1989). Methods same for Tierra Firme site.

• Above-ground Wood Production (Jordan, 1989). Methods same as for Tierra Firme site (four year growth increment study, Tall Caatinga site).

• Leaf Area Index (Klinge and Herrera, 1978; 1983). Methods not indicated. CHECK Klinge and Herrera, 1978

• Leaf Litter Carbon:Nitrogen and Carbon:Phosphorus Ratios (Klinge and Herrera, 1978; Jordan, 1989). Methods not indicated. CHECK Klinge and Herrera, 1978

Below-ground Biomass, Production, and Nutrient Content

• Below-ground Phytomass (Klinge and Herrera, 1983). In the center of each of the 13 plots a soil pit was dug until the groundwater level was reached. The soil of each genetic soil horizon dug out in an area of 50 cm x 50 cm was spread on a plastic sheet. The roots were handpicked, and roots and soil were weighed separately. The technique is described in Klinge and Herrera (1978).

• Root Production in Ingrowth Cylinders (Cuevas and Medina, 1988). Methods same as for Tierra Firme site.

• Nutrient Dynamics in Fine Roots (Cuevas and Medina, 1988; Medina and Cuevas, 1989). Methods same as for Tierra Firme site (root samples from the ingrowth cylinder experiment).

Litter Decomposition (Jordan, 1989; Cuevas and Medina, 1988; and Medina and Cuevas, 1989). Methods same as for Tierra Firme site except vegetation different. From the Tall Amazon Caatinga forest, the following species were selected: Micrandra sprucei, Manilkara sp. and Eperua leucantha. These species are representative of the dominants in each of the three ecosystems in terms of litter input to the forest floor (Cuevas and Medina, 1986).

Bana

Above-ground Biomass, Production, and Nutrient Content

• Above-ground Phytomass (Klinge and Herrera, 1978). Methods same as for Tall Amazon Caatinga.

• Above-ground Phytomass (Bongers et al., 1985). The phytomass study was carried out in March and April 1978 in 9 plots laid out along two lines. Two immediately adjacent plots were established in Tall Bana, five in Low Bana, and two in Open Bana. The litter layer, where present, was sampled in two to five 0.25 m² areas per plot. Herbaceous and woody species <1 cm dbh were uprooted by hand. Bigger plants were mapped and measured for dbh. After felling, crown diameter and bole length were measured. The foliage was removed by hand. Foliage, branches and boles were weighed separately. Herbaceous plants were separated mostly by species and then into above-ground and below-ground fractions. The above-ground fractions of woody plants >1 cm dbh were kept separately for each individual tree or climber. All weighed fractions were sampled for the estimation of dry weight and nutrient content.

• Litterfall and Litterfall Nutrient Content (Cuevas and Medina, 1986). Methods same as for Tierra Firme site.

• Leaf Area Index (Bongers et al., 1985). Ten to 15 dry leaves were selected at random from the leaf samples, outlined on paper and weighed.

Below-ground Biomass and Production

• Root Biomass (Bongers et al., 1985). Two soil pits, one on raised ground and one in a depression, were dug in 7 of the plots using the same metal frame as used in sampling the litter. All roots of each soil layer were handpicked and weighed fresh. Soil was also weighed, and soil samples were taken, for careful inspection, for fine root fragments in the laboratory. All weighed fractions were sampled for the estimation of dry weight and nutrient content.

• Root Production in Ingrowth Cylinders (Cuevas and Medina, 1988). Methods same as for Tierra Firme site.

• Nutrient Dynamics in Fine Roots (Cuevas and Medina, 1988; Medina and Cuevas, 1989). Methods same as for Tierra Firme site (root samples from the ingrowth cylinder experiment).

Litter Decomposition (Jordan, 1989; Cuevas and Medina, 1988; and Medina and Cuevas, 1989). Methods same as for Tierra Firme site except vegetation different. From the Bana forest the following species were selected: Rodognaphalopsis discolor, Remijia involucrata and Macairea rufescens. These species are representative of the dominants in the ecosystem in terms of litter input to the forest floor (Cuevas and Medina, 1986).

Climate

The climate data in this data set are available from a government weather station in the village of San Carlos.

Figure 4. Cutting an experimental plot at the San Carlos de Rio Negro tropical forest site, Venezuela. (Experiment to determine productivity and nutrient dynamics during slash-and-burn agriculture on an Oxisol. Photograph taken September 1976; reproduced by kind permission of Prof. C. F. Jordan, University of Georgia, USA). (SCR1-1.jpg)

Figure 5. Burning an experimental plot at the San Carlos de Rio Negro tropical forest site, Venezuela. (Experiment to determine productivity and nutrient dynamics during slash-and-burn agriculture on an Oxisol. Photograph taken December 1976; reproduced by kind permission of Prof. C. F. Jordan, University of Georgia, USA). (SCR4-1.jpg)

Figure 6. Profile of top 40 cm of soil on an Oxisol hill sub-site at the San Carlos de Rio Negro tropical forest site, Venezuela. (Lateritic concretions are visible mixed in with the sand. Note mat of fine roots and humus on soil surface and concentration of larger roots directly above. Photograph reproduced by kind permission of Prof. C. F. Jordan, University of Georgia, USA). (SCR5-1.jpg)

6. Data Access:

This data set is available through the Oak Ridge National Laboratory (ORNL) Distributed Active Archive Center (DAAC).

Data Archive:

Web Site: http://daac.ornl.gov

Contact for Data Center Access Information:

E-mail: uso@daac.ornl.gov
Telephone: +1 (865) 241-3952

7. References:

Bongers, F., D. Engelen, and H. Klinge. 1985. Phytomass structure of natural plant communities on spodosols in southern Venezuela: the Bana woodland. Plant Ecology, 63(1): 13-34.

Brunig, E. F. 1980. Structure and function of a tropical rainforest in the Amazon MAB-ecosystem project at San Carlos de Rio Negro. 5. In: J. 1. Furtado (ed.) Proc. Vth lnternat. Syrup. Trop. Ecol., Kuala Lumpur 1979 1:33 45.

Cuevas, E., and E. Medina. 1986. Nutrient dynamics within amazonian forests. I. Nutrient flux in fine litter fall and efficiency of nutrient utilisation. Oecologia 68: 466-472.

Cuevas, E., and E. Medina. 1988. Nutrient dynamics within amazonian forests. II. Fine root growth, nutrient availability and leaf litter decomposition. Oecologia 76: 222-235.

Jordan, C. F., and C. Uhl. 1978. Biomass of a "tierra firme" forest of the Amazon basin. Oecologia Plantarum 13: 387-400.

Jordan, C. F., ed. 1989. An Amazonian Rain Forest: structure and function of a nutrient-stressed ecosystem and the impact of slash-and-burn agriculture. UNESCO-MAB/Parthenon, Carnforth, UK. 176 pp.

Jordan, C. F., and G. Escalante. 1980. Root productivity in an Amazonian rain forest. Ecology 61: 14-18.

Jordan, C. F., and R. Herrera. 1981. Tropical rain forests: are nutrients really critical? The American Naturalist 117(2): 167-180.

Jordan, C. F., and J. R. Kline. 1977. Transpiration of trees in a tropical rainforest. J. appl. Ecol. 14: 853 860.

Jordan, C.F., W. Caskey, G. Escalante, R. Herrera, F. Montagnini, R. Todd, and C. Uhl. 1982. The nitrogen cycle in a "Tierra Firme" rainforest in the Amazon territory of Venezuela. Plant and Soil, 67: 325-332.

Herrera, R. 1979. Nutrient distribution and cycling in an Amazonian Caatinga forest on spodosols in southern Venezuela. Ph.D. thesis, University of Reading, 241 pp.

Klinge, H., and R. Herrera. 1978. Biomass studies in amazon caatinga forest in southern Venezuela. I. Standing crop of composite root mass in selected stands. Tropical Ecology 19: 93-110.

Klinge, H., and R. Herrera. 1983. Phytomass structure of natural plant communities on spodosols in southern Venezuela: the tall Amazon Caatinga forest. Vegetatio 53: 65-84.

Medina, E., and E. Cuevas. 1989. Patterns of nutrient accumulation and release in Amazonian forests of the upper Rio Negro basin. In: Mineral Nutrients in Tropical Forest and Savanna Ecosystems. (J. Proctor, ed.) British Ecological Society Special Publication No. 9: Blackwell Scientific, Oxford. pp. 217-240.

Medina, E., M. Sobrado, and R. Herrera. 1978. Significance of leaf orientation for leaf temperature in an amazonian sclerophyll vegetation. Radiation and Environmental Biophysics 15(2): 131-140.

Putz, F. E. 1983. Liana biomass and leaf area of a "Tierra Firme" forest in the Rio Negro Basin, Venezuela. Biotropica 15(3): 185-189.

Stark, N., and M. Spratt. 1977. Root biomass and nutrient storage in rain forest oxisols near San Carlos de Rio Negro. Tropical Ecology 18: 1-9.

Uhl, C., and C. F. Jordan. 1984. Succession and nutrient dynamics following forest cutting and burning in Amazonia. Ecology 65: 1476-1490.