PROJECT SUMMARY

• Establish experimental treatments of decreased snowpack at the Hubbard Brook Experimental Forest (HBEF) in the White Mountains of New Hampshire. We have established experimental treatments in single-species stands of sugar maple and yellow birch. Plots are instrumented with thermistors and time domain reflectometry (TDR) probes to characterize snow and soil temperature and moisture dynamics.



• Examine the effects of decreases in snowpack accumulation and increases in soil freezing on fine root dynamics. We are evaluating root dynamics using minirhizotrons and are quantifying nutrient uptake and litterfall.



• Examine the effects of decreases in snowpack accumulation and increases in soil freezing on microbial biomass C and N dynamics. We are characterizing changes in the size and activity of the soil microbial biomass and quantifying N mineralization, nitrification, denitrification and respiration rates throughout the year.



• Examine the effects of decreases in snowpack accumulation and increases in soil freezing on soil solution chemistry. We are using zero tension lysimeters to sample soil solutions.



• Quantify the effects of decreases in snowpack accumulation and increases in soil freezing on soil-atmosphere fluxes of methane (CH₄) nitrous oxide (N₂O) and carbon dioxide (CO2) using in-field chambers.



• Conduct laboratory and greenhouse studies to evaluate the specific effects of soil freezing and very low soil temperatures on plant root and soil microbial dynamics.



• Evaluate the potential regional scale effects of snow reduction on ecosystem processes. We are calibrating an existing model (SOILTHERM) of soil freezing using experimental data. This model will then be used to evaluate previous soil freezing events that have occurred over the last 33 years at HBEF and to predict changes in snow depth and soil frost given climate change scenarios. Experimental data will then be used to evaluate the long-term effects of climate change-induced snow reduction on community composition, N losses, drainage water acidification and trace gas fluxes in northern hardwood forests.

PROGRESS IN THE FIRST THREE YEARS OF WORK (SEPTEMBER 1996 - AUGUST 1999)

Site selection and instrumentation. We selected four sites for study within the HBEF; two nearly pure stands of sugar maple and two nearly pure stands of yellow birch. At each site, we have delineated two 10 m by 10 m plots. One of the plots will received a six week snow removal treatment in winter 1997/98 and 1998/99 while the other served as a reference.

In fall and winter of 1996, we cleared understory vegetation from all plots, and installed soil solution samplers (lysimeters), thermistors for soil temperature monitoring, water content reflectometers (for measuring soil moisture), soil atmosphere sampling probes, minirhizotron access tubes and trace gas flux measurement chambers. All plots are equipped with dataloggers to allow for continuous monitoring of soil moisture and temperature. Our objective, which was accomplished, was to have all instruments installed in fall and winter of 1996 to allow for any installation-related disturbance effects to subside before our treatment began in fall/winter of 1997.

Develop and test methods for snow removal. After much discussion, we decided to abandon our original idea to construct snow "shelters" to keep plots snow free. We felt that if the shelters were complete enough to keep out drifting snow, they would have significant effects on the thermal regime of the plots. We therefore decided to test the plausibility of keeping the plots snow free by shoveling. Key factors that needed to be evaluated were 1) the amount of labor required to keep a plot snow free by shoveling and 2) does shoveling disturb the forest floor.

A "test site" was set up in a mixed hardwood stand near the Pleasant View Farm dormitory facility at Hubbard Brook. Two 10 m x 10 m plots were delineated, instrumented with thermistors and kept snow free from mid-January through February 1997. More than 50 cm of snow fell during this period.

Results from the test site were very encouraging in several regards. First, it is quite feasible to keep plots snow free by shoveling, i.e. it takes about an hour to remove several inches of snow from a 10 m by 10 m plot. Second, it is possible to shovel plots without disturbing the forest floor. The key to avoiding disturbance is to leave a small (e.g. 2 cm) layer of snow on the surface. This layer does not inhibit freezing, preserves ground vegetation and leaves the forest floor intact. Third, snow removal induced soil freezing as hoped. Even though air temperatures have not been extremely low during our study period, soil on the shoveled plots has frozen.

Laboratory studies

A series of studies to evaluate the effects of severity of freezing (e.g. -10 versus - 3^o C), length of freezing (e.g. 1 week versus three weeks) and freeze-thaw cycles (e.g. three cycles versus continuous freezing over a three week period) on potential net N mineralization and nitrification and trace gas fluxes were carried out during summer and fall 1997. In 1998 and 1999, these studies were expanded to address effects on plants and microbial biomass and activity.

Preliminary results from the first two years of treatment, winter 1997/98 and winter 1998/99

The winters of 1997/98 and 1998/99 had relatively mild temperatures. However, all of the shoveled plots froze to at least 10 cm depth (Figure 1). Plots remained frozen even after shoveling stopped and snow began to accumulate. The yellow birch sites had lower temperatures than the sugar maple plots, especially yellow birch site #2, which is north facing, while all the other sites are south facing. The SOILTHERM model was able to accurately depict the temperature data in the winter of 1997/98 (Figure 2).

The freeze treatment produced a series of significant (p < 0.05) effects on soil biogeochemical processes. Soil nitrate concentrations and in situ N mineralization and nitrification rates (Figure 3) were higher just after snowmelt (April) in the freeze plots than in the control plots. Fluxes of nitrous oxide were higher and uptake of methane was lower in the freeze plots than the control plots (Figure 4). There were marked increases in solute concentrations in drainage water sampled in zero tension lysimeters in the freeze plots relative to the control plots (Figure 5).

Soil freezing caused increased over-winter root mortality in both sugar maple (Figure 6) and yellow birch plots during two successive winters. During the growing seasons following treatment, the fine root production peaked earlier in experimental vs. control plots in both sugar maple and yellow birch plots. The magnitude of this effect was surprising considering the mildness of the freeze treatments. The shift in the phenology of fine root production following freezing has important ramifications for N cycling in this system. While fine roots of both species reacted similarly to these freeze events, leachate chemistry from these plots differed substantially. More detailed study of soil freezing effects on nitrogen cycling in these plots may help explain the mechanism by which this species effect occurred, and this mechanism may involve mycorrhizal relationships.

We were surprised at the marked response that we observed to the relatively minor freezing event that we were able to induce with treatment of our main plots in the relatively mild winter of 1997/98. Eventhough soil temperatures never dropped below –4^o C, we observed significant increases in N cycling and loss in the freeze plots relative to the controls. These data strongly suggest that snow depth and soil freezing events are important regulators of N availability and solute delivery to streams in forested watersheds in the northeastern U.S. These results support observations by Mitchell et al. (1996) who suggested that regional increases in streamwater NO₃^- in the northeastern U.S. during summer 1990 were caused by widespread soil freezing in December 1989.

The significant response to mild freezing that we observed was especially surprising given laboratory studies that we conducted that suggested that we would not observe any significant effects on N cycling without very hard freezing (Figure 7, Nielsen et al., in preparation). In those studies, where soils were frozen at -3 and -13^o C, we did not observe any stimulation of N mineralization at -3^o C, and we did not observe any stimulation of nitrification at either temperature. The marked response that we observed in the field, where temperatures barely reached -3^o C, suggests that plant microbial interactions (which were eliminated in the laboratory studies that had only soil material) are critical to the freeze response. We are currently conducting another round of laboratory studies, with sugar maple seedlings planted in reconstructed soil profiles. We will expose these seedlings to the same temperatures we used in the “soil only” laboratory experiments. We expect to observe much more dramatic responses that we did with the soil only experiments.

In addition to our surprising biotic results, we were also surprised that soils in the freeze plots did not thaw once we stopped shoveling and snow began to accumulate. These observations suggest that one of the most important effects of soil freezing may be to increase runoff and decrease infiltration of snowmelt. We will conduct detailed analysis of leachate volume in lysimeters and continuous soil moisture data from water content reflectometers and will make observations of the nature and extent of soil frost in our plots (e.g., granular versus concrete frost) to quantify this effect.

Literature cited

Driscoll, C.T. and R. Van Dreason. 1993. Seasonal and long-term temporal patterns in the chemistry of Adirondack Lakes. Water Air and Soil Pollution 67:319-344.

Mitchell, M.J., C.T. Driscoll, J.S. Kahl, G.E. Likens, P.S. Murdoch and L.H. Pardo. 1996. Climatic control of nitrate loss from forested watersheds in the northeastern United States. Environmental Science and Technology 30:2609-2612.

Nielsen, C.B., Groffman, P.M., and Hamburg, S.P. 'Changes in microbial respiration and nitrogen cycling following soil freezing.' Manuscript in preparation.

Figure 1.	Soil temperatures at 10 cm depth in two yellow birch (YB) and two sugar maple (SM) sites with freeze and control plots in winter 1997/98 (top) and 1998/99 (bottom).

Figure 2.	Measured and modeled soil temperatures at depths of 10, 20 and 30 cm in a yellow birch site, winter 1997/98.

Figure 3.	Soil nitrate (top) and in situ net nitrification (middle) and mineralization (bottom) in the forest floor of two sugar maple (SM) and yellow birch (YB) sites with freeze and control plots, April – May, 1999. Nitrate values are mean with standard error. Nitrification and mineralization values are single plot values from composite final and initial values.

Figure 4.	Soil atmosphere fluxes of nitrous oxide (top), methane (middle) and carbon dioxide (bottom) in two sugar maple (SM) and yellow birch (YB) sites with freeze and control plots, April 1998. Values are mean with standard error of four sampling dates, with four or five in situ flux chambers at each site.

Figure 5.	Soil solution concentrations of nitrate (NO3-), dissolved organic carbon (DOC), dissolved organic N (DON) and soluble reactive phosphorus (SRP) sampled with zero tension lysimeters placed beneath the forest floor in a sugar maple site with freeze and control plots from December 1997 – June 1999.

Figure 6.	Fine root mortality in the forest floor of a sugar maple site with freeze and reference plots, July 1997 – May 1999.

Figure 7.	Potential net N mineralization (top) and nitrification (bottom) in soils from yellow birch and sugar maple sites exposed to –3o and –13o C freezes in a laboratory study.