Summary:

The objective of this study was to quantify net primary productivity as a function of rainfall in mesic to wet montane rainforests in Maui, Hawaii. The Maui Moisture Gradient is a sequence of six sites located on the island of Maui that range from 2200-mm to 5050-mm mean annual rainfall, while temperature and all other state factors (parent material, substrate age, organisms, and topography) that control NPP remain relatively constant. This data set contains annual estimates of net primary productivity made in 1996 and 1997.

The data provided are estimates of the accumulation of biomass by plants for a given year, or net primary productivity (NPP). Estimates are given for aboveground and belowground productivity, and the sum as net primary productivity. These data are part of a larger study that focused on the dynamics of carbon cycling and storage in everwet rainforest as a function of changes in rainfall regime. The Maui Moisture Gradient sites were located within a geographic distance of less then 5 km in the Makawao and Koolau Forest Reserves on the north flank of Haleakala volcano. Temperature regimes were similar at the constant altitude of the sites (approximately 1300 m) while mean annual precipitation ranged systematically from 2200 mm/yr (mesic) to over 5000 mm/yr (wet) as a function of aspect relative to the prevailing trade winds. The sites were located on lava flows from the Kula volcanic series (mean age 410,000 years), which was part of the shield-building phase of Haleakala volcano. The original shield surface has been dissected by stream channels, so the study sites were located on shield volcano remnant surfaces on broad, flat (< 5 % slope) interfluve areas to minimize variation in local topography. The soils on this precipitation gradient are classified as Inceptisols and Andisols developed from lava with surface ash deposits. The Hawaiian Islands flora and fauna are relatively species-poor, thus a few species and genera occupy a broad range of environmental conditions. As a result, the forest canopy at all sites was consistently dominated by the native evergreen tree Metrosideros polymorpha (Myrtaceae) which comprises 80% to 100% of basal area in these forests. The understory vegetation was dominated by a variety of ferns and other herbaceous species at all sites, but the dominance of particular understory species shifted among sites. This watershed area has never been cleared by humans and all six sites were located in mature forests stands.

Data Citation:

Cite this data set as follows:

Schuur, E. A. G. 2005. NPP Tropical Forest: Maui, Hawaii, U.S.A., 1996-1997. Data set. Available on-line [http://daacsti.ornl.gov] from Oak Ridge National Laboratory Distributed Active Archive Center, Oak Ridge, Tennessee, U.S.A.

Table of Contents:

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

1. Data Set Overview:

Project: Net Primary Production (NPP)

The net primary productivity measurement presented here is sum of the increase in overstory biomass plus aboveground litterfall (together = ANPP) and belowground root turnover (BNPP). Aboveground tree biomass was calculated using forest inventory and regression techniques. We established a single 25 x 25 m productivity plot at each site and measured the diameter at breast height of all woody vegetation greater than 5 cm diameter. These measurements were converted to aboveground biomass using species-specific allometric equations derived from other studies in the Hawaiian Islands. Because the annual increment of growth was relatively small, dendrometer bands placed on a subset of ten trees were used to record the growth of trees. The average growth of trees with dendrometer bands was applied to the total basal area (> 5 cm dbh) of the study plot to estimate the biomass increment added on an annual basis. We estimated aboveground NPP (ANPP) as the sum of increases in the aboveground biomass plus the litter production in the 25 x 25 m production plots established at each site. We collected litterfall in 15 trays (0.156 m²) per site randomly placed on the forest floor and emptied at monthly intervals. As an indirect measure of belowground turnover, carbon dioxide fluxes from the soil surface were measured at monthly intervals over the period of a year. The surface flux of carbon dioxide consists of autotrophic root respiration plus heterotrophic decomposition of organic matter and has been used in ecosystems approaching equilibrium as a measure of C inputs. Therefore we assume that carbon released from the soil from heterotrophic respiration is approximately equivalent to the annual input of carbon entering the soil and, in combination with litterfall measurements, can be used as an index for belowground NPP.

The investigator was E. A. G. Schuur. You may contact Dr. Schuur at the following address:

University of Florida

Gainesville, FL

(352) 392-7913

tschuur@ufl.edu

2. Data Characteristics:

Spatial Coverage

The plots were located in areas that reasonably represented the local vegetation while keeping local factor such as topography constant among sites.

The plot size was 25 x 25 m. The number of trees per hectare ranged from 1,104 to 2,352.

SITE	MAP	LATITUDE	LONGITUDE
Site 1	2200 mm	20 48' 21.0" N	156 15' 19.0" W
Site 2	2450 mm	20 48' 26.5" N	156 15' 10.5" W
Site 3	2750 mm	20 48' 30.0" N	156 15' 0.0" W
Site 4	3350 mm	20 48' 47.5" N	156 14' 49.5" W
Site 5	4050 mm	20 48' 47.5" N	156 14' 25.0" W
Site 6	5050 mm	20 48' 54.5" N	156 13' 47.0" W

 MAP =  mean annual precipitation

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

Site (Region)	Westernmost Longitude	Easternmost Longitude	Northernmost Latitude	Southernmost Latitude	Geodetic Datum
Maui, Hawaii, USA	-156.25528	-156.22972	20.81514	20.80583

Temporal Coverage

All measurements were made from 1996/01/01 to 1997/12/31.

Temporal Resolution

All NPP measurements based on plant dry matter accumulation are expressed on a annual basis (Mg ha-1 yr-1).

All of the NPP Estimates for the Maui Moisture Gradient Sites are in one file.

Net Primary Productivity for Maui Moisture Gradient Sites

Site	MAP (mm)	NPP (Mg C ha-1 yr-1)	ANPP (Mg C ha-1 yr-1)	BNPP (Mg C ha-1 yr-1)
1	2200	9.563	4.735	4.828
2	2450	9.734	5.205	4.529
3	2750	8.513	4.83	3.683
4	3350	8.408	4.625	3.783
5	4050	6.193	3.655	2.538
6	5050	3.877	2.085	1.792

MAP = mean annual precipitation in millimeters

NPP = net primary productivity is the sum of the increase in overstory biomass (ANPP) and belowground root turnover (BNPP)

ANPP = aboveground net primary productivity is the sum of the increase in overstory biomass plus aboveground litterfall

BNPP = belowground net primary productivity is the belowground root turnover

Biomass units are in megagrams (10⁶ grams) of carbon per hectare per year.

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 dataset accounts for aboveground growth, leaf turnover, and root turnover. These are considered the major components of NPP.

This dataset has been applied to a larger review of tropical forest NPP. See Schuur (2003).

4. Quality Assessment:

The data provided are of generally good quality based on the site location and time frame. Data were examined for general consistency and clarity. Measurements were made under a variety of field conditions (rain mostly!) that presumably had little effect on the quality of measurements.

Sources of Error

The greatest potential source of error that other researchers need to be aware of for the Maui Moisture Gradient sites is the location of plots. These were stratified to keep state factors constant, so they may not represent NPP on larger scales with other factors such as local topography varying. Also, given that the measurements are presented on an annual basis, the amount of observation time (span of 2 years) is relatively short, although typical of many NPP studies.

5. Data Acquisition Materials and Methods:

Tree biomass -- Aboveground tree biomass was calculated using forest inventory and regression techniques. We established a single 25 x 25 m productivity plot at each site and measured the diameter at breast height (dbh) of all woody vegetation greater than 5 cm diameter. These measurements were converted to aboveground biomass using species-specific allometric equations derived from other studies in the Hawaiian Islands. For species other than M. polymorpha, allometries included tree height measured with an extension pole. Because tree heights were not available for all M. polymorpha trees, we used an allometric equation that relied on dbh alone. Finally, there were several large Acacia koa trees at Site 2 and 3. For these, a generalized wet forest allometric equation was used to estimate biomass for those individual trees because no species-specific allometric equation that includes A. koa trees of that size exists.

Aboveground net primary productivity -- We estimated ANPP as the sum of increases in the aboveground biomass plus the litter production in the 25 x 25 m production plots established at each site. All litter was dried at 70 degrees C to obtain a consistent dry weight measurement. We collected litterfall in 15 trays (0.156 m²) per site randomly placed on the forest floor and emptied at monthly intervals. Three sites (Sites 3, 5, 6) contained understory species whose litter production could not be collected with littertraps because senesced leaves did not fall to the forest floor. In Site 3 and Site 5, production of the fern Dicranopteris linearis (N.L. Burm.) Underw. (Gleicheniaceae) was measured by tagging 30 individual fronds, tagging new segments as they grew, and collecting and weighing the tagged frond segments that senesced during the study period. At 10 random points in each production plot we counted frond density in 1 x 1 m quadrats. Total litter production by D. linearis was calculated by multiplying the density of fronds by the annual litter produced by each frond. At Site 6, we measured the litter production of Carex alligata, an understory sedge, in a similar fashion. Leaves from 30 individual tillers were tagged as they grew, collected and weighed after they senesced, and multiplied by an estimate of tiller density.

Because the annual increment of growth is relatively small, we placed aluminum, spring-loaded dendrometer bands on a subset of ten trees randomly selected from all trees in the production plot. Growth increments were recorded at six month intervals over 18 months, starting after the bands had been allowed to settle on the trees for six months. The average growth of trees with dendrometer bands was applied to the total basal area (> 5 cm dbh) of the study plot to estimate the biomass increment added on an annual basis, using the regression equations previously described.

Belowground Productivity -- We measured carbon dioxide flux from the soil surface at monthly intervals over the period of a year. Soil carbon dioxide fluxes from all measurement points were averaged to estimate an average daily efflux of carbon dioxide that incorporates seasonal fluctuations. The surface flux of carbon dioxide consists of autotrophic root respiration plus heterotrophic decomposition of organic matter and has been used in ecosystems approaching equilibrium as a measure of C inputs. The sites on this precipitation gradient cover a broad climatic range of unmanipulated, mature forests, meeting the general assumptions of this method. Therefore we assume that carbon released from the soil from heterotrophic respiration is approximately equivalent to the annual input of carbon entering the soil and, in combination with litterfall measurements, can be used as an index for belowground NPP.

To measure carbon dioxide flux, we used static chambers placed over tins containing soda lime for a 24 hour period, following established methods. Initially, ten randomly placed rings per site were pressed 1 cm into the surface of the soil and remained in the field during the course of this study so that fine roots would not be disturbed during measurements. During the 24 hour measurement periods, rings were removed and replaced with chambers of the same diameter such that a seal formed between the chamber and the soil. Tins of soda lime were dried at 105 degrees C and weighed before and after the field measurements, and blanks were used to correct for the drying process. We calculated grams of CO2 adsorbed using a revised correction factor of 1.69 to account for weight differences between H2O and CO2. Soda lime has been shown in laboratory studies to overestimate very low (zero) fluxes and underestimate high fluxes but it is linear within a range of moderate fluxes where most of our measurements occurred. Furthermore, field studies have shown that it can be reliable compared to more sophisticated techniques over the range of carbon dioxide fluxes from soil, given the natural variability of soil fluxes.

6. Data Access:

This data is available through the Oak Ridge National Laboratory (ORNL) Distributed Active Archive Center (DAAC) or the EOS Data Gateway.

Data Archive Center:

Contact for Data Center Access Information:

E-mail: uso@daac.ornl.gov
Telephone: +1 (865) 241-3952
FAX: +1 (865) 574-4665

Product Availability:

Requested data can be provided electronically on the ORNL DAAC's anonymous FTP site or on CD-ROM.

7. References:

Schuur, E.A.G. 2003. Productivity and global climate revisited: the sensitivity of tropical forest growth to precipitation. Ecology 84:1165-1170.

Schuur, E.A.G., O.A. Chadwick, and P.A. Matson. 2001. Carbon cycling and soil carbon storage in mesic to wet Hawaiian montane forests.

Schuur, E.A.G. 2001. The effect of water on decomposition dynamics in mesic to wet Hawaiian montane forests. Ecology 82:3182-3196.

Schuur, E.A.G., and P.A. Matson. 2001. Aboveground net primary productivity and nutrient cycling across a mesic to wet precipitation gradient in Hawaiian montane forest. Oecologia 128:431-442.

Miller A.J., E.A.G. Schuur, and O.A. Chadwick. 2001. Redox control of phosphorus pools in montane forest soils in Hawaii. Geoderma 102:219-237.

8. Document Information: