This story highlights a current research project at the Hubbard Brook Experimental Forest. To read more about research projects at HBEF, visit Current Research page. Check back regularly to learn about new research projects.
Stream Ecosystem Research at Hubbard Brook


Current Stream Research

Linking direct and indirect data on dispersal: isolation by slope in a headwater stream salamanderWinsor H. Lowe (Univ. of Montana), Gene E. Likens (IES), Mark A. McPeek (Dartmouth), and Don C. Buso (IES)

There is growing recognition of the need to incorporate information on movement behavior in landscape-scale studies of dispersal. One way to do this is by using indirect indices of dispersal (e.g., genetic differentiation) to test predictions derived from direct data on movement behavior. Mark-recapture studies documented upstream-biased movement in the salamander Gyrinophilus porphyriticus (Plethodontidae). Based on this information, we hypothesized that gene flow in G. porphyriticus is affected by the slope of the stream.  Specifically, because the energy required for upstream dispersal is positively related to slope, we predicted gene flow to be negatively related to change in elevation between sampling sites. Using amplified DNA fragment length polymorphisms among tissue samples from paired sites in nine streams in the Hubbard Brook Watershed, New Hampshire, USA, we found that genetic distances between downstream and upstream sites were positively related to change in elevation over standardized 1-km distances.  This pattern of isolation by slope elucidates controls on population connectivity in stream networks, and underscores the potential for specific behaviors to affect the genetic structure of species at the landscape scale. More broadly, our results show the value of combining direct data on movement behavior and indirect indices to assess patterns and consequences of dispersal in spatially complex ecosystems.

Lowe WH, Likens GE, McPeek MA, Buso DC.  2006.  Linking direct and indirect data on dispersal: Isolation by slope in a headwater stream salamander.  Ecology 87: 334-339.

Fish Populations in Streams of the HBEFDana Warren (Cornell), Clifford Kraft (Cornell), Gene Likens (IES), Madeleine Mineau (Cornell)

To date, few studies have evaluated fish in streams of the Hubbard Brook Experimental Forest (HBEF).  What fish are present in streams at the HBEF and how are the fish distributed?  This project is designed to address these important initial questions that can set the stage for future research on fish in this well-studied ecosystem.  In summer 2005, we conducted fish surveys in all Hubbard Brook tributaries and at three locations on the mainstem of Hubbard Brook.  Brook trout (Salvelinus fontinalis) were the only fish species present in all HBEF tributaries except Norris Brook, which also contained slimy sculpin (Cottus cognatus).  In most cases the upstream movement of fish was clearly constrained by a waterfall or other large physical barrier.  In a few cases, the mechanism for a loss of fish as one moved upstream was not immediately apparent.  We are exploring the possibility that episodic acidification during snowmelt moves fish downstream in the spring and that the presence of groundwater seeps are particularly important in maintaining fish in these streams.  One chronically acidified stream, contained no fish.  The mainstem was surveyed in three places from approximately 200 m downstream of the Forest Service bridge upstream to its confluence with Crazy Brook and then upstream of its confluence with Crazy Brook to its headwaters in a beaver pond.  In all sections, brook trout were the only species present.  No surveys were conducted below the gorge in Hubbard Brook.  This feature may act as a barrier to fish movement and upstream recolonization of species such as dace or sculpin following their local extirpation due to stream acidification in the 1980’s.  Norris brook, which does contain sculpin, enters the mainstem below this waterfall. 

Landscape-level controls on microbial diversity in the HBEFNoah Fierer (Univ. of Colorado), Rob Jackson (Duke), Emily Bernhardt (Duke)

Fine particulate organic matter (FPOM) is the primary reservoir of organic nutrients (C, N, and P) in forested 1st order streams. FPOM is rich in bacterial and fungal biomass and these microorganisms largely control the decomposition of FPOM, releasing soluble nutrients from FPOM into the stream channel. The microbes decomposing FPOM are likely to be diverse, however few studies have attempted to characterize this diversity and examine which biotic and abiotic factors have the most important influence on the composition of these microbial communities. The goal of this study is to expand our knowledge of FPOM microbial communities by examining the ecological factors that influence their community structure across the Hubbard Brook watershed. We expect that this study at HBEF will represent a starting point for a more comprehensive continental-scale survey of within-stream microbial community ecology. For this study, we will compare the microbial communities inhabiting stream FPOM and the associated riparian soils from 11 streams at HBEF. These streams differ from one another in terms of their watershed characteristics and stream chemistries and we expect the composition of FPOM and riparian soil microbial communities to reflect these differences. The results of a continental-scale analysis of microbial communities in unsaturated soils show that pH is a very good predictor of microbial diversity and community composition. For this reason, we expect that streamwater pH, along with vegetation type (which dictates the characteristics of the detrital inputs) will be good predictors of stream microbial community composition. By relating the ecology of bacteria, which are microns in size, with ecological patterns that can be observed at larger spatial scales (meters to kilometers), this study will link two distinct fields of ecology, microbial ecology and landscape ecology.

Woody Debris and Nitrogen Dynamics in Headwater Stream EcosystemsDana Warren (Cornell), Darren Bade (IES), Kristi Judd (IES), Clifford Kraft (Cornell), and Gene Likens (IES).

The latest in a series of stream manipulation studies conducted at the HBEF is one examining the role of woody debris dams in controlling the cycling and export of nitrogen (N).  Woody debris dams are prominent features in streams of the HBEF, and may influence stream biogeochemistry through effects on organic matter retention, stream geomorphology and residence time, and by serving as hot spots for microbial activity.  To investigate the role of debris dams on N retention in streams, we have manipulated debris dams (dams removed from Crazy Brook and dams added to W3).  We are measuring responses in several processes related to N cycling (e.g., nitrate uptake length, uptake by mosses and microbes, denitrification, dissolved organic N and carbon production) and comparing these responses to a reference stream (W6) before and after manipulations.  Tracer additions of the stable isotope 15NO3- allow us to quantify the role of debris dams in N retention and assess how dynamics change from spring to fall.  Background sampling occurred during 2005 and streams were manipulated after the fall sampling.  Post-treatment measurements of stream nitrate dynamics will be conducted in summer 2006, and 2007.

In conjunction with field studies, we are developing and testing a conceptual model to explain the amount and distribution of large wood in streams running through mixed hardwood-conifer forests at the HBEF and elsewhere across the northern forest.

Long-term change in stream ecosystemEmily Bernhardt (Duke),  Gene Likens (IES), Robert Hall, Jr. (U WY), Bill McDowell (UNH), Dana Warren (Cornell) 

There has been very little work to date on ecosystem development in lotic ecosystems, despite a tremendous body of work in Temperate Zone terrestrial ecosystems. At the HBEF there has been a change in how stream ecosystems process nitrogen (N). In the 1970's Joanna Richey and Bill McDowell were unable to measure any uptake of nitrate from stream water, instead stream ecosystems served as sources of inorganic N through high rates of nitrification. In the late 1990's, work by Emily Bernhardt, Bob Hall, and Gene Likens showed that HBEF streams actively process nitrate, and can dramatically reduce watershed export of N. To test whether these ecosystems have changed, B. McDowell and E. Bernhardt repeated in 2000, experiments that McDowell conducted in the late 1970's, adding leaf leachate and determining the effect on DIN concentrations. They found that while adding leachate carbon to streams in the 1970's had no impact on stream N processing, the same releases in the 1990's dramatically stimulated nitrate uptake. With the support of The A.W. Mellon foundation a workshop was held of all stream researchers that had conducted research at HBEF since the early 1960's. The purpose of the workshop was to synthesize data and knowledge about how HBEF streams have changed over the past 40 years.  We demonstrated that changes in HBEF streams over the period of record could have important implications for interpreting the long-term record of streamwater N losses and propose new approaches for the long-term monitoring of watershed studies that will allow us to better understand stream impacts on N cycling in the future (see Bernhardt et al. 2005).

Bernhardt ES, GE Likens, RO Hall, DC Buso, SG Fisher, TM Burton, JL Meyer, MH McDowell, MS Mayer, WB Bowden, SEG Findlay, KH Macneale, RS Stelzer, and WH Lowe.  2005.  Can't see the forest for the stream? - In-stream processing and terrestrial nitrogen exports.  BioScience 55: 219-230.  

Links between dissolved organic carbon and nitrogen export from forest watershed Kristi Judd (IES), Darren Bade (IES), and Gene Likens (IES)

Nitrate and dissolved organic carbon (DOC) are linked through several potential biogeochemical processes, and understanding the mechanisms underlying these links is needed to improve our understanding of C and N export from watersheds.  Negative relationships have been observed between DOC and nitrate concentrations at both large (Goodale et al. 2005) and small (Hedin et al. 1998) scales, suggesting that DOC concentration may drive nitrate export.  However, increased nitrate concentrations in experimental studies has been found to increase DOC export.  These contrasting patterns highlight our lack of understanding of the causal relationship between DOC and nitrate.  Understanding the mechanisms linking DOC and nitrate in forested watersheds in the Northeastern U.S. is particularly interesting because nitrate export from forested catchments has unexpectedly declined, despite forest maturation and continued inputs of atmospheric N (Goodale et al. 2003).  DOC export from the HBEF, on the other hand, has been relatively constant over the long-term record, in contrast to increases observed in some ecosystems in the Northeastern U.S. and Western Europe.  Within the Hubbard Brook Valley, streams differ in their nitrate and DOC concentrations.  To better understand the links between DOC and N export, we are combining field surveys and experimental approaches.  We are measuring DOC, dissolved organic N, and nitrate in streamside wells in various landscape types (e.g., seeps, hardwood, and coniferous).  To determine the direction of the causal relationship between nitrate and DOC export, we are amending soil cores from each landscape type with three treatments: 1) + DOC, 2) + nitrate, and 3) +DOC + nitrate. 



Goodale CL, JD Aber, PM Vitousek, WH McDowell.  2005.  Long-term decreases in stream nitrate: Successional causes unlikely; Ecosystems 8: 334-337.


Goodale CL, Aber JD, Vitousek PM.  2003.  An unexpected nitrate decline in New Hampshire streams.  Ecosystems 6: 75-86.


Hedin LO, JC von Fischer, NE Ostrom, BP Kennedy, MG Brown, GP Robertson.  1998.  Thermodynamic constraints on nitrogen transformations and other biogeochemical processes at soil-stream interfaces.  Ecology 79: 684-703.






Watershed nutrient export in response to widespread soil frost due to low snow accumunlation Gene Likens (IES), Don Buso (IES), Kristi Judd (IES), Darren Bade (IES).

Snowpack insulates soils, and soils at the HBEF freeze only rarely.  Soil freezing alters watershed nitrogen (N) dynamics.  The long-term record at the HBEF shows increased stream nitrate export following years of increased soil freezing (e.g., 1969-70 and 1973-74).  Snow manipulation studies at HBEF showed that freezing increased root mortality and nitrification, but did not alter N-mineralization, indicating that changes in snow accumulation and soil freezing could dramatically alter N cycling and export.  Low snow accumulation in the winter of 2006 resulted in widespread soil frost throughout the Hubbard Brook Valley; initial surveys indicate that this soil frost event was the most extensive since 1955.  This event provides an excellent opportunity to understand how changes in winter snow pack alter nutrient and base cation retention in forest watersheds.  To quantify responses to soil frost, we will intensively monitor stream export over the spring and summer of 2006.  We will conduct longitudinal studies in Watershed 6 (the biogeochemical reference watershed) to determine how responses in export change with elevation, and in W1, which received Ca2+ and Si addition (as wollastonite) in 1999.  We will also sample all of the tributaries to Hubbard Brook each month to better understand spatial patterns in response to soil frost.  Concentrations of nitrate and base cations will be compared to “non-frost&rdquo years.  In 1998, stream responses to a severe ice storm that caused extensive damage to the forest canopy were documented (Houlton et al. 2003, Bernhardt et al. 2003).  Comparing responses to the soil frost event (below-ground disturbance) to the ice-storm event (above ground disturbance) will help us better understand the role of disturbances in controlling nutrient export from forest watersheds. 



Bernhardt ES, GE Likens, DC Buso, and CT Driscoll.  2003.  In-stream uptake dampens effects of major forest disturbance on watershed nitrogen export.  Proceedings of the National Academy of Sciences 100: 1034-1038.


Houlton BZ, Driscoll CT, Fahey TJ, Likens GE, Groffman PM, Bernhardt ES, Buso DC.  2003.  Nitrogen dynamics in ice storm-damaged forest ecosystems: Implications for nitrogen limitation theory.  Ecosystems 6: 431-444.




Recent Publications on Stream Research at Hubbard Brook

Bernhardt ES, GE Likens, RO Hall, DC Buso, SG Fisher, TM Burton, JL Meyer, MH McDowell, MS Mayer, WB Bowden, SEG Findlay, KH Macneale, RS Stelzer, and WH Lowe.  2005.  Can't see the forest for the stream? - In-stream processing and terrestrial nitrogen exports.  BioScience 55: 219-230.  

Bernhardt ES and GE Likens.  2004.  Controls on periphyton biomass in heterotrophic streams.  Freshwater Biology 49: 14-27.

Bernhardt ES, GE Likens, DC Buso, and CT Driscoll.  2003.  In-stream uptake dampens effects of major forest disturbance on watershed nitrogen export.  Proceedings of the National Academy of Sciences 100: 1034-1038.

Bernhardt ES and GE Likens.  2002.  Dissolved organic carbon enrichment alters nitrogen dynamics in a forest stream.  Ecology 83: 1689-1700.

Bernhardt ES, RO Hall, and GE Likens.  2002.  Whole-system estimates of nitrification and nitrate uptake in streams of the Hubbard Brook Experimental Forest.  Ecosystems 5: 419-430. 

Judd KE, GE Likens, and PM Groffman.  Submitted.  High nitrate retention during winter in soils at the Hubbard Brook Experimental Forest. 

Hall RO, ES Bernhardt, and GE Likens.  2002.  Relating nutrient uptake with transient storage in forested mountain streams.  Limnology and Oceanography 47: 255-265. 

Hall RO, GE Likens, HM Malcom.  2001.  Trophic basis of invertebrate production in 2 streams at the Hubbard Brook Experimental Forest.  Journal of the North American Benthological Society 20: 432-447.

Hall RO, KH Macneale, ES Bernhardt, M Field, and GE Likens.  2001.  Biogeochemical responses of two forest streams to a 2-month calcium addition.  Freshwater Biology 46: 291-302.

Likens GE, DC Buso, BK Dresser, ES Bernhardt, RO Hall, KH Macneale, and SW Bailey.  2004.  Buffering an acidic stream in New Hampshire with a silicate mineral.  Restoration Ecology 12: 419-428.

Lowe WH, GE Likens, ME Power.  2006.  Linking across spatial scales in stream ecology BioScience (in press).

Lowe WH, Likens GE, McPeek MA, Buso DC.  2006.  Linking direct and indirect data on dispersal: Isolation by slope in a headwater stream salamander.  Ecology 87: 334-339.

Lowe, W.H., and G.E. Likens.  2005.  Moving headwater streams to the head of the class. BioScience 55:196-197.

Lowe W., KH Nislow, and GE Likens. 2005. Forest structure and stream salamander diets: implications for terrestrial-aquatic connectivity. Verh. Internat. Verein. Limnol. 29:279-286.

Macneale KH, BL Peckarsky, and Likens.  2005.  Stable isotopes identify dispersal patterns of stonefly populations living along stream corridors.  Freshwater Biology 50: 1117-1130.

Macneale KH, BL Peckarsky, and GE Likens.  2004.  Contradictory results from different methods for measuring direction of insect flight.  Freshwater Biology 49: 1260-1268.

Macneale KH, GE Likens, and BL Peckarsky. 2002. Feeding strategy of an adult stonefly (Plecoptera): implications for egg production and dispersal. Verh. Internat. Verein. Limnol. 28: 1140-1146.

Stelzer RS and GE Likens.  2006.  The effects of sampling frequency on estimates of dissolved silica export by streams: the role of hydrological variability and concentrations-discharge relationships.  Submitted.

Stelzer RS, J HeffernanJ, and GE Likens.  2003.  The influence of dissolved nutrients and particulate organic matter quality on microbial respiration and biomass in a forest stream.  Freshwater Biology 48: 1925-1937.

Contact information: 

Dr. Darren Bade
Institute of Ecosystem Studies
65 Sharon Turnpike, Box AB
Millbrook, NY 12545
(845) 677-5343

Dr. Emily S. Bernhardt
Nicholas School of the Environment
PO Box 90328
Duke University
Durham, NC 27708-0328

Dr. Noah Fierer
University of Colorado
Boulder, CO 80309
(303) 492-5615

Dr. Kristin Judd
Institute of Ecosystem Studies
65 Sharon Turnpike, Box AB
Millbrook, NY 12545
(845) 677-5343

Dr. Gene E. Likens
Institute of Ecosystem Studies
65 Sharon Turnpike, Box AB
Millbrook, NY 12545

Dr. Winsor Lowe
Division of Biological Sciences
The University of Montana
Missoula, MT 59812
(406) 243-4375

Dana Warren
Department of Natural Resources
Cornell University
Ithaca, NY

Recent Stream Ecosystem Projects at the Hubbard Brook Experimental Forest
Controls on solute export across multiple spatial and temporal scales
G.E. Likens, D.C. Buso, R.S. Stelzer, J.H. McCutchan, Jr., E.S. Bernhardt, and K.H. Macneale.

Processes operating at multiple spatial and temporal scales affect the export of solutes from watersheds (Likens 1999). In order to better understand what controls solute export within landscape mosaics, it is necessary to study the export of solutes from watersheds of varying sizes and positions within the landscape and to understand how the timing of sampling affects estimates of solute export.

Valley-Wide Stream Survey Previously, most studies of solute export at the HBEF have been focused on small, headwater streams (numbered watersheds, Figure 1), with some work also examining export from the entire valley. Export of solutes from watersheds of intermediate size (e.g., 100-1000 ha) has, however, received less attention. Thus, we have expanded the list of permanent, routine sampling locations for streamwater chemistry at HBEF to include some intermediate-sized watersheds.

We began a valley-wide survey of stream chemistry and morphology at the HBEF, New Hampshire in April, 2000. This survey is a contribution to the ongoing Valley-Wide Streamwater Initiative at HBEF that we began in 1996, and data from this survey will be used in the selection of new routine stream sampling locations.

Between 4 May and 8 June 2000 we surveyed 8 tributaries to the main Hubbard Brook, draining south-facing watersheds (Figure 1). The 8 tributaries account for approximately 985 ha (33%) of the 3000 ha Hubbard Brook Valley. For each tributary, we surveyed physical and chemical parameters from the main Hubbard Brook to the first running water. Those streams with large (>20 ha watershed area) perennial branches were treated as separate watersheds and each branch was surveyed. At junctions of smaller streams, we surveyed only the section with the higher flow.  The result of this study is now in press (Likens et al. in press)

  Map of HBEF  
  Figure 1. Map of streams at HBEF. Streams draining watersheds 1-9 are sampled routinely for solute concentrations and are monitored continuously for discharge. Streams draining shaded watersheds were surveyed in 2000; all tributaries were sampled twice (Spring and Fall) in 2001.  

For each 100-m interval along the tributaries, we quantified channel form, substrate, presence of moss and organic debris dams, and identified nearby vegetation with a predetermined series of index (US Forest Service 1995; US Forest Service 1978). We also made field measurements of pH, conductance, and water temperature at 100-m intervals using a laboratory-calibrated field meter. Water samples were collected at discrete points where the field meter indicated significant changes in any of the 3 parameters above. These discrete samples were later measured in the lab for pH, conductance, and acid neutralizing capacity (ANC) to confirm the field readings.

Chemical parameters, especially pH, varied considerably within and among watersheds (Table 1). The pH and conductivity generally changed gradually from first running water downstream, but we did find large differences in chemical composition for some streams draining adjacent watersheds (e.g., the east and west branches of Beaver Brook).

Table 1. Range of pH and watershed area for the 8 streams sampled in 2000. For each stream, the pH range is from the lowest to highest sampling location. Areas are approximate and rounded to nearest 5 ha.


pH range

Area, ha


Norris Brook

6.5 - 5.5


nutrient manipulations

Paradise Brook

5.5 - 4.0


HBEF W 1-4

Bear Brook

5.5 - 4.5

HBEF W 5-6

Falls Brook

5.8 - 5.3


Nupert experiment sites

Cushman Brook

4.7 - 4.5


steep, pristine; coniferous cover

Beaver Brook E

5.7 - 4.5


active beaver ponds

Beaver Brook W

4.5 - 5.0


abandoned beaver ponds

Lost Brook

5.0 - 4.7


steep, pristine; coniferous cover

Sampling Frequency and Estimation of Solute Export In addition to our survey of spatial variation in stream chemistry at HBEF, we began a study of the effects of sampling frequency on the estimation of solute export. Preliminary analyses revealed that the frequency of sampling for solute concentration could affect the accuracy of annual estimates of solute export at HBEF. For some solutes (e.g., dissolved silica), sampling monthly or seasonally resulted in substantial biases in estimates of annual export compared with estimates based on weekly sampling. For other solutes (e.g., calcium), however, estimates of export were essentially unaffected by sampling frequency. These findings led us to examine the effects of sampling frequency on the estimation of solute export in a general context.

As part of the Hubbard Brook Ecosystem Study, solute concentrations have been measured weekly for selected streams at HBEF for 43 years. Using data from HBEF, we have re-sampled measured solute concentrations to simulate the effects of less frequent sampling. In order to make comparable comparisons across a wide range of stream types and to make comparisons between daily and weekly sampling at HBEF, we have employed a simulation approach to predict changes in solute concentration on a daily basis.

Over a wide range of streams, we determined relationships between solute concentration and discharge as described in Johnson et al. (1969). Concentration and discharge data are from the long-term record of the Hubbard Brook Ecosystem Study or from Alexander et al. (1996). We then predicted daily changes in concentration from measurements of mean daily discharge and re-sampled predicted concentrations at intervals of 7, 14, 28, 56, and 91 days. For each resampling interval, estimates of annual solute export were compared to estimates based on daily measurements.

The simulation and re-sampling procedure was repeated for streams that vary in mean annual discharge, flashiness, and predictability of flow. For flashy streams (i.e., those with hydrographs dominated by precipitation), estimates of annual export for dissolved silica are substantially biased (>20% error) at sampling intervals greater than 14 days. For snowmelt-dominated streams and others with predictable hydrographs, estimates of annual export were less subject to bias at long sampling intervals. Final analyses are underway and the findings of this study are being prepared for publication (Stelzer et al. 2001). One of the ultimate goals of this project is to establish sampling guidelines for the estimation of solute export for a given type of stream and solute.

Results from the Valley-Wide survey have generated a variety of hypotheses related to the distribution of organisms, including salamanders, fish, and microbes and processes controlling stream chemistry and the export of nutrients and energy from forest watersheds. 



Alexander, R. B., J. R. Slack, A. S. Ludtke, K. K. Fitzgerald, and T. L. Schertz. 1996. Data from selected U.S. Geological Survey national stream water quality monitoring networks (WQN). C-D ROM digital media, U.S. Geological Survey.

Johnson, N. M., G. E. Likens, F. H. Bormann, D. W. Fisher, and R. S. Pierce. 1969. A working model for the variation in stream water chemistry at the Hubbard Brook Experimental Forest, New Hampshire. Water Resources Research 5: 1353-1363.

Likens, G.E. 1999. The science of nature, the nature of science: long-term ecological studies at Hubbard Brook. Proceedings of the American Philosophical Society 143:558-572.

Stelzer, R. S., J. H. McCutchan, Jr., G. E. Likens, and D. C. Buso. 2001. Effects of sampling frequency on the estimation of solute export. In preparation.

US Forest Service. 1978. Ecological land types of the White Mountains National Forest. Ecosystem Inventory Report. USDA Forest Service, Region 9, WMNF, Laconia, NH.

US Forest Service. 1995. Basin-wide Stream Inventory Handbook. USDA Forest Service, Region 9, WMNF, Laconia, NH. 20p.


Food sources for stream invertebrates: estimates from gut contents and stable isotope ratios J.H. McCutchan, Jr., R.O. Hall Jr., and G.E. Likens

Aquatic insects and other invertebrates often are the dominant primary consumers in small streams and form an important link between primary producers and higher trophic levels. Low rates of aquatic primary production and large inputs of terrestrial plant detritus to canopied streams have led to the general conception that algae are nutritionally unimportant to invertebrates in most small streams. Algae often are a better food source than terrestrial plants, however, and may be more important to consumers in small, canopied streams than is currently recognized.

One way to test the dependence of invertebrates on algae is to estimate how much algae and detritus they eat. Hall et al. (2001) measured consumption of algae and detritus in Bear Brook and the main Hubbard Brook for one year and found that algae supported 5% of secondary production in Bear Brook and 20% in Hubbard Brook. However, analyses of gut contents reveal what invertebrates have eaten, but may not reveal what was assimilated because of uncertainty in estimates of assimilation rates for different food sources. Also, some food items can be difficult to identify. Stable isotope ratios (see below) act as a natural tracer of organic matter and can provide information about what has been assimilated by invertebrates.

In June, 2000, we began a year-long study of the relative contribution of algal carbon to consumer production at three sites. This study combined measurements of C and N stable isotope ratios for consumers and their potential food sources with estimates of secondary production. At two sites, we used natural-abundance stable isotope ratios to estimate carbon sources and trophic position for consumers. At a third site, we added 13C-bicarbonate to increase the isotopic separation between terrestrial plants and algae. At each site, the relative contribution of algal carbon to macroinvertebrate production is being compared to the relative availability of algal carbon in the stream; the purpose of this comparison is to test the assumption that food resources are used by consumers in proportion to their availability. Even though algal production contributed less than 1-2% of the total supply of organic matter to these streams, algae carbon supported up to 20% of the macroinvertebrate production. Contrary to expectations, algal carbon supported a larger fraction of total macroinvertebrate production at the tributary sites than at Hubbard Brook, even though the tributaries were more heavily shaded than Hubbard Brook.




Hall, R. O., Jr., G. E. Likens, and H. M. Malcolm. 2001. Trophic basis of invertebrate production in 2 streams at the Hubbard Brook Experimental Forest. J. N. Am. Benthol. Soc. 20: 432-437.

History of Stream Ecosystem Research at the Hubbard Brook Experimental Forest
Gene E. Likens, Robert O. Hall, Jr., Emily S. Bernhardt, Kate H. Macneale, Kristi Judd


Map of HBEF


Stream ecosystem research at the Hubbard Brook Experimental Forest (HBEF) started in the mid 1960's, and has included: surveys of invertebrate taxa; long-term studies of stream chemistry, temperature and hydrology; flux, cycling and mass-balance approaches to element cycling; and manipulative experiments to examine stream processes. The following summary demonstrates how different research programs have contributed to our integrated understanding of the stream as an ecosystem.

The Stream Community While much of the research on stream ecosystems at the HBEF has focused on ecosystem processes, and a number of studies have examined the distribution and abundance of aquatic and riparian organisms.  Early studies, such as McConnochie and Likens' 1969 survey that found 50 species of caddisflies (Trichoptera), hinted at the great diversity of aquatic invertebrates, and perhaps invertebrate habitats, within the watersheds. Fiance (1978, 1979) surveyed the distributions and emergence patterns of mayflies (Ephemeroptera) and stoneflies (Plecoptera), and correlated those patterns with environmental factors such as stream pH.

Marilyn Mayer (1986) studied the diet of a common species of caddisfly (Neophylax aniqua) found in Bear Brook. She discovered that algae made up about half of the total consumption of these caddisflies and supported ~75% of their growth. Work done by Thomas Burton and others (1988) showed that grazers accounted for ~ 9% of the insect biomass in Bear Brook. These studies indicated that algal production was higher than the 0% assumed by Fisher and Likens (1973). An important question emerged: Had there been a change in the food web of these ecosystems since the late 1960’s, or was the assumption incorrect?

Burton et. al. (1988) described the community dynamics for stream insects, suspended bacteria and benthic algae in the Watershed 6 stream over the course of a year (Oct. 1984 to 1985). They examined changes in chlorophyll a, bacterial biomass and macroinvertebrate densities and species composition in both leaf packs and sediment baskets. They found that the fauna was dominated by shredders and predators, that algal biomass remained low throughout most of the year with peaks immediately after leaf fall and prior to canopy development. Despite this low algal biomass a few grazing species were able to derive most of their energy from autotrophic production (e.g. Neophylax sp. and Tanytarsus guerlus).

Burton and Likens (1975) (along with a small army of undergraduate students) surveyed the salamander communities of the HBEF. They found that the biomass of salamanders (1770 g/ha wet weight) within the Hubbard Brook Valley was about twice that of birds during the breeding season and about equal to the biomass of small mammals. One terrestrial species Plethodon cinereus accounted for 93.5% of the total biomass, with the remaining 6.5% consisting of the streamside species, Desmognathus fuscus, Eurycea bislineata, and Gyrinophilus porphyriticus. Gene and Phyllis Likens have been repeating the 1970 salamander surveys in recent years, and more recently, Winsor Lowe has studied factors controlling dispersal of G.  porphyriticus using genetic analyses.

Energy Budgets and Element Flux and Cycling The hallmark of Hubbard Brook stream research is studies examining the controls on ecosystem energy and nutrient fluxes and cycling.  The Fisher and Likens (1973) energy budget for Bear Brook was one of the first studies to consider a stream as an ecosystem. Almost all energy in this stream was derived from allochthonous inputs from the watershed. Dissolved organic matter was the primary form of exported carbon. Algal production was considered to be negligible in Bear Brook, however more recent studies have shown while standing stocks are low and production may be low its relative importance to some invertebrates is significant (e.g., caddisflies (Neophylax sp. and Tanytarsus guerlus) Mayer 1986, Burton et al. 1988), possibly suggesting a shift in stream food web structure over time.  A later study found that sediment respiration (measured as CO2 production) was strongly correlated with sediment organic matter (Hedin 1990).

Meyer and Likens (1979) developed a budget for phosphorus, an element that is transported mostly as organic particles. Dissolved P concentrations are very low in headwater streams of the HBEF, < 1-2μg/L. Particle concentration in streams at HBEF increase exponentially with discharge, hence most phosphorus is exported at the highest discharges.

Meyer et al. (1981) compared the annual (fall of 19968-1969) mass balances of phosphorus, nitrogen, and organic carbon in Bear Brook. The organic C budget was assumed to be balanced (Fisher and Likens 1973), but the annual exports of P exceeded inputs by 12% while inputs of N exceeded exports by 8%. Inputs of all elements were processed in the stream: 69% of coarse (>1mm) particulate matter C, 78% of coarse particulate organic N, and 73% of coarse particulate organic P inputs were exported in the fine particulate (0.45 μm - 1mm), dissolved (<0.45μm), or gaseous fraction. Organic C was exported primarily in the gaseous or dissolved fraction, P in the fine particulate fraction, and N in the dissolved fraction. The N:P atomic ratio in fluvial exports was lower during the growing season (16:1) than during the rest of the year (130:1).

Several researchers have compared stream processes among forested and deforested watersheds. Burton and Likens (1973) studied the effect of strip cutting on stream temperatures of Watershed 4 of the HBEF. They found that stream water temperatures could increase by 4-5°C in 25-m wide strips cut in the canopy, with subsequent and equivalent cooling in adjacent uncut strips, on sunny days during the summer.  Thornton (1974) described the impacts of the Watershed 2 deforestation on stream invertebrate communities. He found that following cutting the amount of organic debris within the stream was reduced and that invertebrate secondary production was lower. In a forested watershed, Bilby and Likens (1980) examined the controls on particle export by removing all organic debris dams from a stream. Debris dam removal dramatically increased organic matter export (632% for fine particles) in streams, showing the importance of in-stream debris in regulating carbon export from the stream. Likens and Bilby (1982) proposed a model for long-term loss and recovery of organic debris dams following deforestation.  Stream order and other characteristics and vegetation type and succession were considered in this model.  After surveying debris dam density in the experimental watersheds, Hedin et al. (1988) constructed another model predicting long-term changes in particle export from streams at different stages of recovery following clearcutting as a function of debris dam density. These debris dams are hotspots for metabolic activity, as indicated by elevated levels of community respiration (Hedin 1990). Steinhart et al. (2001) found that denitrification potential was markedly higher in accumulations of organic matter in debris dams than in other stream habitats, suggesting that these sites may have an important influence on stream nitrogen transformations.

Whole Stream Manipulations Stream reach manipulations have been an important means of furthering our understanding of stream ecosystem structure and function.  A number of such experimental manipulations have been conducted in streams of the HBEF and in conjunction with the whole watershed- ecosystem manipulations, a hallmark of the Hubbard Brook Ecosystem Study, these studies shed light on the roles of streams in controlling the retention and export of nutrients and energy from watersheds. 

Hedin et. al. (1990) experimentally tested the hypothesis that in some areas affected by acid rain the deposition of strong mineral acids has been buffered by concurrent losses in DOC and organic acids. They experimentally acidified a brownwater stream (Watershed 9) and found that DOC concentrations were not reduced by acidification and that organic acids have little capacity to buffer inputs of strong mineral acids.

Ronald Hall and colleagues conducted and acid addition experiment to Norris Brook in 1978 (Hall et al. 1980).  They added dilute sulfuric (target pH was 4, reduced from a background of 6) and measured increased streamwater concentrations of Al, Ca, Mg, and K, but no change in dissolved organic carbon, Na+, NO3-, NH4+, and several other metals. They found an increase in invertebrate drift, lower insect biomass, and reduced numbers of emerging insects during the five-month continuous acid addition. Many macroinvertebrates, especially collectors, appeared to respond to the acidification by leaving the section during the first week of the treatment. Higher periphyton biomass and lower numbers of fungal hyphomycetes were measured following acidification.

One consequence of acidic deposition is the mobilization of aluminum (Al) from soils to streams.  To investigate the effects of episodic increases in Al and acidification, e.g., during the spring melt period, Hall et al. (1985) performed Al additions (as AlCl3) to Norris Brook.  These experiments showed that Al affected stream physical, chemical, and biological characteristics.  Physical-chemical effects included reduced DOC concentrations (36% reduction within the first hour of the experiment), reduced surface foam accumulation due to lower surface tension, and reduced pH.  The strongest biological response was increased drift of macroinvertebrates with increased Al concentration and exposure time.     

Meyer (1979) added phosphate to Bear Brook to examine mechanisms of phosphate retention. She found that phosphate was removed quickly from the water column by abiotic processes, and that phosphorus concentration in stream water is held in equilibrium by abiotic exchange. Richey et al. (1985) added ammonium, nitrate and urea to Bear Brook. They found that much ammonium and urea were removed from the water column and nitrified to form nitrate but that nitrate was not consumed.  More recent research by Steinhart et al. (2001) suggests that denitrification may be a much more important process in streams of the HBEF than previously thought.  Laboratory studies of denitrification potential suggest that rates of denitrification in Bear Brook may remove as much as 25-110% of the nitrate output in stream water during July and October. McDowell (1985) added DOC in the form of extracts of maple leaves to measure DOC uptake, and found rapid uptake of DOC, which was likely abiotic. Interestingly, DOC uptake in a stream (Watershed 5) with much organic matter removed was similar to Bear Brook which has a high organic matter standing stock, suggesting that microbial processes associated with organic matter is not important to DOC uptake in Watershed 5.

McDowell and Likens (1988) furthered our understanding of DOC dynamics at HBEF with a synthesis paper, describing DOC concentrations, composition and flux for the entire Hubbard Brook Valley. They found that water flowing through the HBEF landscape shows a distinct increase in DOC concentration with passage through the canopy and upper soil horizons. They postulated that the qualitative composition of DOC within the Valley is the result of biotic processes while the flux of DOC is regulated by direct abiotic processes and linkages between abiotic and biotic processes.

Findlay et al. (1993) studied bacterial--algal relationships in Bear Brook and the nearby clearcut Watershed 5 stream. They found that rates of primary production in the clearcut streams were as much as 5 times higher than those measured for Bear Brook but that standing crops of algae (chlorophyll a) did not differ between the two streams. Bacterial abundance and growth did not differ between the two streams and bacterial abundance did not appear to be correlated with algal biomass in either stream. Findlay et al. (1993) also manipulated light levels in several field shading experiments and nutrient additions in a number of laboratory experiments. The shading reduced algal abundance and growth but did not affect bacteria. Similarly, the nutrient amendments stimulated algal abundance but did not lead to increased bacterial biomass. They concluded that there was not a tight trophic connection between algae and bacteria for these streams.

Recently, whole-stream manipulations have been used to investigate the effects of alkalinization on stream ecosystems and to trace the biological pathways of carbon and nitrogen. To test the role of streams in processing calcium lost from watershed soils, Hall et al. (2001) added 240 μeq Ca l-1 as Calcium chloride to two second order streams for two months. Sodium bicarbonate also was added to study the effects of simultaneous increases in pH and alkalinity. In the high-pH stream (Ca + bicarbonate), 10-50% of the added calcium was removed from the water column in the 80-m study reach, but after the addition, Ca was released very slowly from the sediments. In the low-pH stream (Ca only), less Ca was removed by the water column and desorption of Ca was not measured following the addition. The algal community in these streams was not measurably affected by changes in calcium or pH but the addition of Calcium chloride negatively affected the emergence of the stonefly Leuctra ferruginea. Mulholland et al. (2000) used a 15-N tracer addition to study food resources for macroinvertebrates in Bear Brook. They found that some mayflies (e.g., Stenonema sp. and Baetis sp.) assimilated primarily epilithon. McCutchan and Likens added 13C bicarbonate to Falls Brook for 7 months to estimate the contribution of algal carbon to the growth of stream consumers. Preliminary results from this study indicate that some mayflies (e.g., Baetis sp., Ameletus sp., and Heptageniidae) were strongly dependent on algal carbon as a food source. These results are consistent with the results of Mulholland et al. (2000) and are quite surprising considering the fact that algae account for less than 1% of the input of organic matter to Falls Brook.

Atmospheric input of acidic anions has depleted calcium from soil of the Hubbard Brook Experimental Forest (Likens et al. 1998).  To test the effects of inputs of soil Ca to streams and the role of streams in retaining Ca lost from soils, Hall et al. (2001) conducted a Ca addition experiment under buffered (NaHCO3 added to simulate increased alkalinity) and unbuffered conditions.  Stream uptake of Ca was positively related to pH and between 10 and 50% of the added Ca was retained in the 80 m study reach of the buffered stream. 



Bilby, R. E. and G. E. Likens. 1980. Importance of organic debris dams in the structure and function of stream ecosystems. Ecology 61:1107-1113.

Burton, T. M. and G. E. Likens. 1975. Salamander populations and biomass in the Hubbard Brook Experimental Forest, New Hampshire. Copeia 3: 541-546.

Burton, T. M. and G. E. Likens. 1973. The effect of strip-cutting on stream temperatures in the Hubbard Brook Experimental Forest, New Hampshire. BioScience 23(7): 433-435.

Burton, T. M., K. E. Ulrich and S. K. Haack. 1988. Community dynamics of bacteria, algae and insects in a first order stream in New Hampshire, USA. Verh. Internat. Verein. Limnol. 23:1125-1134.

Fiance, S. B. 1978. Effects of pH on the biology and distribution of Ephemeralla (Eurylophella) funeralis (Emphemeroptera: Ephemerellidae). Oikos 31:332-339.

Fiance, S. B. 1979. Effects of lowered pH on the composition and structure of stream invertebrate communities. Ph.D. Thesis. Cornell University. 165 pp.

Findlay, S., K. Howe and D. Fontvielle. 1993. Bacterial -algal relationships in streams of the Hubbard Brook Experimental Forest. Ecology 74(8):2326-2336.

Fisher, S. G. and G. E. Likens. 1973. Energy flow in Bear Brook, New Hampsh Brook, New Hampshire: an ecosystem metabolism. Ecological Monographs 43:421-439.

Hall, R. J., G. E. Likens, S. B. Fiance, and G. R. Hendrey. 1980. Experimental acidification of a stream in the Hubbard Brook Experimental Forest, New Hampshire. Ecology 61:976-989.

Hall, R. J., C.T. Driscoll, G. E. Likens, and J. M. Pratt.  1985.  Physical, chemical, and biological consequences of episodic aluminum addition to a stream.  Limnology and Oceangraphy 30: 212-220.

Hall, R. O., Jr., K. H. Macneale, E. S. Bernhardt, M. Field, and G. E. Likens. 2001. Biogeochemical responses of two forest streams to a two-month calcium addition. Freshwat. Biol. 46: 291-302.

Hedin, L. O. 1990. Factors controlling sediment community respiration in woodland stream ecosystems. Oikos. 57: 94-105.

Hedin, L. O., G. E. Likens, K. M. Postek and C. T. Driscoll. 1990. A field experiment to test whether organic acids buffer acid deposition. Nature. 345: 798-800.

Hedin, L. O., M. S, Mayer and G. E. Likens. 1988. The effect of deforestation on organic debris dams. Verh. Internat. Verein. Limnol. 23:1135-1141.

Likens, G. E., D. C. Buso, B. K. Dresser, E. S. Bernhardt, R. O. Hall, K. H Macneale, and S. W. Bailey.  2004.  Buffering an acidic stream in New Hampshire with a silicate mineral.  Restoration Ecology 12: 419-428.

Likens, G. E., C. T. Driscoll, D. C. Buso, T. G. Siccama, C. E. Johnson, G. M. Lovett, T. J. Fahey, W. A. Reiners, D. F. Ryan, C. W. Martin, S. W Bailey.  1998.  The biogeochemistry of calcium at Hubbard Brook.  Biogeochemistry 41: 89-173.

 Likens GE and RE Bilby.  1982.  Development, maintenance and the role of organic-debris dams in New England streams.  Pp. 122-128.  In: FJ Swanon, RJ Janda, T Dunne and DW Swanston (eds.).  Sediment budgets and routing in forested drainage basins.  USDA Forest Service General Technical Report PNW-141.

Mayer, M. S. 1986. The importance of algae as food for insects in Bear Brook, a small woodland stream in New Hampshire. M.S. Thesis, Cornell University. 70 pp.

Mayer, M. S. and G. E. Likens. 1987. The importance of algae in a shaded headwater stream as food for an abundant caddisfly (Trichoptera). Journal of the North American Benthological Society. 6(4): 262-269.

McConnochie, K. and G. E. Likens. 1969. Some Trichoptera of the Hubbard Brook Experimental Forest in Central New Hampshire. The Canadian Field-Naturalist 83:147-154

McDowell, W. H. 1985. Kinetics and mechanisms of dissolved organic carbon retention in a headwater stream. Biogeochemistry 1:329-352.

McDowell, W.H. and G.E. Likens. 1988. Origin, composition and flux of dissolved carbon in the Hubbard Brook valley. Ecological Monographs. 58(3):177-195.

Meyer, J. L. 1979. The role of sediments and bryophytes in phosphorus dynamics in a headwater stream ecosystem. Limnology and Oceanography 24:365-375.

Meyer, J. L. and G. E. Likens. 1979. Transport and transformation of phosphorus in a forest stream ecosystem. Ecology 60:1255-1269.

Meyer, J. L., G. E. Likens and J. Sloane. 1981. Phosphorous, nitrogen and organic carbon flux in a headwater stream. Arch. Hydrobiol. 91(1):28-44.

Mulholland, P. J., J. L. Tank, D. M. Sanzone, W. M. Wolheim, B. J. Peterson, J. R. Webster, and J. L. Meyer. 2000. Food resources of stream macroinvertebrates determined by natural-abundance stable C and N isotopes and a N-15 tracer addition. J. N. Am. Benthol. Soc.. 19(1):145-157.

Richey, J. S., W. H. McDowell, and G. E. Likens. 1985. Nitrogen transformations in a small mountain stream. Hydrobiologia 124:129-139.

Steinhart, G. S., G.E. Likens and P. M. Groffman. 2001. Denitrification in stream sediments in five northeastern USA streams. Verh. Internat. Verein. Limnol. 27: 1331-1336.

Thornton, G. A. 1974. Some effects of deforestation on stream macroinvertebrates. M.S. Thesis. Cornell University. 84 pp.