Deep Soil Frost Research Articles

Snowpack affects soil microclimate throughout the year

Variations in snow depth have complex effects on soil microclimate. Snow insulates soil and thus regulates, along with air temperature, the nature, and extent of soil freezing. There is great uncertainty about the main drivers of soil freezing, which have important effects on ecosystem carbon and nitrogen cycling processes and might change as climate warms and snowfall decreases as part of climate change. Here, we utilitze sites from a variety of elevations and aspects within the northern hardwood forest at the Hubbard Brook Experimental Forest (New Hampshire, USA) to investigate relationships between seasonal snowpack, soil freezing, and soil microclimate across this gradient using 8 years of bi-weekly snowpack and soil frost-depth measurements, and continuous soil climate monitoring. We utilize a time-integrated snowpack descriptor and find that snowpacks with lower seasonal snow water equivalents result in more soil temperature variation and deeper soil frost but have no effect on variation in soil moisture. Seasonal snow water equivalent of the snowpack influences the date of rapid soil warming in the spring, which in turn influences both summer soil moisture and an index of annual cumulative soil heat. These results show that snowpack dynamics, which are highly sensitive to changes in climate, have wide-ranging effects on soil microclimate year-round and thus could have important implications for ecosystem carbon and nitrogen cycling processes.

Late Glacial/Early Holocene slope deposits on the Swiss Plateau: Genesis and palaeo-environment

On the Swiss Plateau, glacial and glaciofluvial sediments deposited during MIS 2 are widespread. They are generally overlain by younger sediments of different origin, mainly related to periglacial and slope wash processes. Depending on their genesis and composition, 3 types of slope deposits can be distinguished, which are classified as units 1, 2a and 2b. Unit 1 is characterized by its wide spatial occurrence and a constant thickness of 40–60cm, high silt contents, and a sharp lithological discontinuity to the lying sediments, generally forming the parent material for the Ah- and Al-horizons of the Luvisols. Unit 2 represents erosion rills, filled up with loess loam and some gravel (<5%) due to slope wash. Locally, periglacial deformations led to cryoturbatic structures in these sediments, which is used to differentiate between unit 2a (no indications for periglacial activities) and unit 2b (with cryoturbations).Based on stratigraphic assumptions, the formation of unit 1 has been tentatively attributed to the Younger Dryas. Optical Stimulated Luminescence and radiocarbon dating of unit 2 indicate morphodynamic activity during the Early Holocene up to 7.5ka, suggesting an open landscape. This scenario is in conflict with the established notion that reforestation in Middle Europe took place during the Bölling-Alleröd warm period. In order to explain this apparent contradiction for the Younger Dryas to Early Holocene environment, we suggest a continental climate with dry summers and drought stress on the vegetation, in combination with very cold winters and deep soil frost at the Pleistocene/Holocene transition, caused by the orbital precession maximum. After 7.5ka, with the spread of oceanic woodland and reduced seasonal contrasts, the landscape stabilized and soils developed.

Optimum soil frost depth to alleviate climate change effects in cold region agriculture

On-farm soil frost control has been used for the management of volunteer potatoes (Solanum tuberosum L.), a serious weed problem caused by climate change, in northern Japan. Deep soil frost penetration is necessary for the effective eradication of unharvested small potato tubers; however, this process can delay soil thaw and increase soil wetting in spring, thereby delaying agricultural activity initiation and increasing nitrous oxide emissions from soil. Conversely, shallow soil frost development helps over-wintering of unharvested potato tubers and nitrate leaching from surface soil owing to the periodic infiltration of snowmelt water. In this study, we synthesised on-farm snow cover manipulation experiments to determine the optimum soil frost depth that can eradicate unharvested potato tubers without affecting agricultural activity initiation while minimising N pollution from agricultural soil. The optimum soil frost depth was estimated to be 0.28–0.33 m on the basis of the annual maximum soil frost depth. Soil frost control is a promising practice to alleviate climate change effects on agriculture in cold regions, which was initiated by local farmers and further promoted by national and local research institutes.

Nematode community resistant to deep soil frost in boreal forest soils

As global climate change advances, shifts in winter precipitation are becoming more common in high latitude ecosystems, resulting in less insulating snow cover and deeper soil frost. Long-term alterations to soil frost can impact on ecosystem processes such as decomposition, microbial activity and vegetation dynamics. In this study we utilized the longest running, well-characterized soil frost manipulation experiment in a boreal forest. We measured nematode family composition and feeding group abundances at four different soil layer depths from plots that had been subjected to deep soil frost for one and 11 years. The overall abundance of nematodes and the different feeding groups were unaffected by deep soil frost. However, a higher Maturity Index was weakly associated with deep soil frost (indicative of lower nutrient enrichment and more persister nematode (i.e., K-strategist) families), likely due to the loss of nutrients and reduced inputs from inhibited decomposition. Multivariate and regression analyses showed that most nematode families were weakly associated with dominant understory plant species and strongly associated with soil organic matter (SOM). This is probably the result of higher resource availability in the control plots, which is favorable to the nematode community. These results indicate that the nematode community was more strongly driven by the long-term indirect effects of deep soil frost on SOM as opposed to the direct effects. Our findings highlight that the indirect effects of altered winter precipitation and soil frost patterns may be more important than direct winter climate effects. Further, such indirect effects on SOM and the plant community that may affect the nematode community can only be seen in long-term experiments. Finally, given the critical role nematodes play in soil food webs and carbon and nutrient cycling, our results demonstrate the necessity of considering the response of nematodes to global climate change in boreal forest soils.

Hydrologic flowpaths during snowmelt in forested headwater catchments under differing winter climatic and soil frost regimes

Changes in hydrologic flowpaths have important impacts on the timing, magnitude and hydrochemistry of run-off during snowmelt in forested catchments, but how flowpaths are affected by variation in winter climate and the irregular presence of soil frost remains poorly understood. The depth and extent of soil frost may be expected to increase as snowpack decreases or develops later because of climate change. In this study, we used end-member mixing analysis to determine daily contributions of snow, forest floor soil water and groundwater to stream run-off during snowmelt under different soil frost regimes resulting from interannual and elevational variation at the Hubbard Brook Experimental Forest in New Hampshire, USA. We observed greater routing of run-off through forest floor flowpaths during early snowmelt in 2011, when the snowpack was deep and soil frost was minimal, compared with the early snowmelt in 2012 under conditions of deep and extensive soil frost. The results indicate that widespread soil frost that penetrated the depth of the forest floor decreased the flow signal through the shallowest subsurface flowpaths, but did not reduce overall infiltration of melt waters, as the contribution from the snow-precipitation end-member was similar under both conditions. These results are consistent with development of granular soil frost which permits vertical infiltration of melt waters, but either reduces lateral flow in the forest floor or prevents the solute exchange that would produce the typical chemical signature of shallow subsurface flowpaths in streamwater. Copyright © 2016 John Wiley & Sons, Ltd.

Spatial and temporal variation in weather events critical for boreal agriculture: III Frost and winter time fluctuation

In the boreal zone of Europe, differences between the four seasons are considerable. Also, the within-season variation in climatic conditions is substantial. This has many impacts on agriculture that are exceptional when compared to any other environmental zone in Europe. All the meteorological data were based on weather observations made by the Finnish Meteorological Institute. Likelihood (%) for soil frost (≤ 0 °C at 20 cm soil depth) at nine weather stations, and late snow cover (&gt; 1 cm) (10 km × 10 km grid) were estimated for late spring. Probabilities (%) of night frost at the ground surface (March-September) were calculated at nine weather stations by frequencies of the lowest observed night-time temperature: a) between –2 and –5 °C (mild), b) ≤ –5 °C (moderate) and c) ≤ –9 °C (severe). Also, the probabilities (%) of night frost in mid-summer were estimated (≤ –1 °C for at least five hours). Furthermore, a significant shift from mild to below-freezing conditions was measured in winter as a period of at least ten days with daily maximum temperatures above 0°C followed by at least a 10-day period with daily mean temperatures below –5°C in order to characterize high fluctuating winter conditions. All these except late snow cover constitute high risks to crop production. Deep soil frost may postpone sowings, while in advanced springs, night frost may cause damage. For winter crops and perennials, shifts from mild to cold spells outside the growing season are particularly detrimental. Again the data may have many other applications beyond the assessments highlighted in this paper.

Influence of soil frost on the character and degradability of dissolved organic carbon in boreal forest soils

AbstractRecent studies suggest that increases in extent and duration of winter soil frost increases dissolved organic carbon (DOC) concentrations in boreal riparian soils and connected aquatic systems during the subsequent spring and summer. However, little is known about the impact of frost on DOC character and its degradability. We applied three experimental treatments to riparian soils in northern Sweden—shallow soil frost (insulated), deep soil frost (snow removed) and control plots—to test the effect of different soil frost regimes on the chemical characteristics and degradability of soil DOC. Soil pore water samples were analyzed using excitation‐emission fluorescence (parallel factor analysis) combined with biological and photochemical degradation experiments. We found that the absolute bacterial metabolic rates were significantly lower in samples from the shallow soil frost treatments, compared with the other treatments. Explorative multivariate analyses indicate that increasing soil frost is contributing to increased protein‐like fluorescence and to increased biological degradability of the DOC. Our study shows that decreases in riparian soil frost due to climate warming may not only contribute to decreased riparian DOC concentrations but also lead to shifts in the DOC composition, resulting in decreased biodegradability (yet similar photodegradability) of the DOC that is exported from riparian soils to streams.

Impacts of snow season on ground-ice accumulation, soil frost and primary productivity in a grassland of sub-Arctic Norway

Europe’s and the World’s northernmost agriculture is very vulnerable to harsh overwintering conditions. It is important from both an economic and societal standpoint to have accurate methods of predicting the severity and impact of the current snow season. Technology has advanced to enable such measurements to be regularly recorded but despite this, a detailed assessment, involving remote sensing , of the impacts of various types of snow season on agricultural yields in northernmost Europe has not previously been undertaken. Here we characterize variation in snow types and concomitant soil frost and ground-ice accumulation at a Norwegian sub-Arctic, maritime-buffered site (Tromsø, Troms County, 69 °N) during the period 1989/90 to 2013/14 and analyse how winter conditions affect agricultural productivity (both measured in the field and using remote sensing). These data were then used to build important predictive modelling approaches. In total, five contrasting types of snow season were identified, from snow-rich with no soil frost and no ground-ice to low snow and considerable soil frost and ground-ice. Conditions of low snow and low soil frost and ground-ice that result from numerous warming events were rare within the time period studied but are predicted to become the dominant snow season type. Agricultural productivity was lowest and claim settlements paid to farmers were highest after winters with high accumulation of plant-damaging, hermetic ground-ice. Deep soil frost per se did not affect primary productivity. Overall, our results together with information from other sources, suggest that icy, low snow conditions are the most challenging of all seasonal types for both the environment and livelihoods in sub-Arctic Norway. Winters with extremely deep snow also cause considerable problems. As winters are expected to warm more than summers, it is likely that the winter climate will become an even stronger regulator of northern primary productivity. To better understand the physical and biological effects of the changing winter climate, there is a requirement for continued and increasing monitoring of winter processes, especially related to frost and ice in the rhizosphere, as this is currently not well covered in national monitoring programs. Continued monitoring will enable further refinement of predictions and will support the better community planning for greatest agricultural benefit.

Snow removal reduces annual cellulose decomposition in a riparian boreal forest

Kreyling, J., Haei, M. and Laudon, H. 2013. Snow removal reduces annual cellulose decomposition in a riparian boreal forest. Can. J. Soil Sci. 93: 427–433. Decomposition is a key process in carbon and nutrient cycling. However, little is known about its response to altered winter soil temperature regimes in boreal forests. Here, the impact of soil frost on cellulose decomposition over 1 yr and soil biotic activity (bait-lamina sticks) over winter, in spring, and in summer was investigated using a long-term (9-yr) snow-cover manipulation experiment in a boreal Picea abies forest. The experiment consisted of the treatments: snow removal, increased insulation, and ambient control. The snow removal treatment caused longer and deeper soil frost (minimum temperature −8.6°C versus −1.4°C) at 10 cm soil depth in comparison with control, while the increased insulation treatment resulted in nearly no soil frost during winter. Annual cellulose decomposition rates were reduced by 46% in the snow removal manipulation in comparison with control conditions. Increased insulation had no significant effect on decomposition. The decomposition was mainly driven by microorganisms, as no significant difference was observed for containers enclosed with a 44-µm and a 1-mm mesh. Soil biotic activity was slightly increased by both the snow removal and the increased insulation treatment in comparison with control conditions over winter. However, this effect disappeared over spring and summer. We conclude that soil frost can have strong effects on decomposition in boreal ecosystems. Further studies should investigate to which degree the observed reduction in decomposition due to reduced snow cover in winter slows or even offsets the expected increase in decomposition rates with global warming.

Snow removal reduces annual cellulose decomposition in a riparian boreal forest

Kreyling, J., Haei, M. and Laudon, H. 2013. Snow removal reduces annual cellulose decomposition in a riparian boreal forest. Can. J. Soil Sci. 93: 427-433. Decomposition is a key process in carbon and nutrient cycling. However, little is known about its response to altered winter soil temperature regimes in boreal forests. Here, the impact of soil frost on cellulose decomposition over 1 yr and soil biotic activity (bait-lamina sticks) over winter, in spring, and in summer was investigated using a long-term (9-yr) snow-cover manipulation experiment in a boreal Picea abies forest. The experiment consisted of the treatments: snow removal, increased insulation, and ambient control. The snow removal treatment caused longer and deeper soil frost (minimum temperature -8.6°C versus -1.4°C) at 10 cm soil depth in comparison with control, while the increased insulation treatment resulted in nearly no soil frost during winter. Annual cellulose decomposition rates were reduced by 46% in the snow removal manipulation in comparison with control conditions. Increased insulation had no significant effect on decomposition. The decomposition was mainly driven by microorganisms, as no significant difference was observed for containers enclosed with a 44-µm and a 1-mm mesh. Soil biotic activity was slightly increased by both the snow removal and the increased insulation treatment in comparison with control conditions over winter. However, this effect disappeared over spring and summer. We conclude that soil frost can have strong effects on decomposition in boreal ecosystems. Further studies should investigate to which degree the observed reduction in decomposition due to reduced snow cover in winter slows or even offsets the expected increase in decomposition rates with global warming.

Soil frost enhances stream dissolved organic carbon concentrations during episodic spring snow melt from boreal mires

AbstractChanges in winter time conditions at high‐latitude ecosystems could severely affect the carbon exchange processes. Using a 15 year stream record combined with winter field measurements and laboratory experiment, we studied the regulation of dissolved organic carbon (DOC) concentration in stream water draining boreal mire during snow melt. The most unanticipated finding was that cold soils with deep soil frost resulted in increased snow melt DOC concentrations in the stream runoff. Wintertime field measurements of DOC concentrations below the mire soil frost showed that this phenomenon could be explained by freeze‐out of DOC giving higher levels of DOC in the soil water below the ice as the soil frost developed downwards in the mire. Experimental freezing of water with a certain DOC concentration in the laboratory further corroborated the freeze‐out of DOC. In this experiment, as much as 96% of the DOC was excluded from the ice, whereas the freeze‐out in the mire was less effective (60%). The difference between the proportion of DOC retained in pure water relative to the proportion retained in peat water during freezing is probably due to trapped DOC in the solid peat soil matrix. A simple mass‐balance model showed that to explain the higher stream DOC concentrations during a winter with deep soil frost, approximately 0.5% of the mire area needed to be hydrologically connected to the stream discharge. In the stream records, we also found that the DOC concentrations during snow melt episodic runoff were negatively related to increasing intensity of the snow melt episodes (dilution by low DOC snow melt water) and higher previous export of DOC.

Absence of snow cover reduces understory plant cover and alters plant community composition in boreal forests

Snow regimes affect biogeochemistry of boreal ecosystems and are altered by climate change. The effects on plant communities, however, are largely unexplored despite their influence on relevant processes. Here, the impact of snow cover on understory community composition and below-ground production in a boreal Picea abies forest was investigated using a long-term (8-year) snow cover manipulation experiment consisting of the treatments: snow removal, increased insulation (styrofoam pellets), and control. The snow removal treatment caused longer (118 vs. 57 days) and deeper soil frost (mean minimum temperature -5.5 vs. -2.2°C) at 10 cm soil depth in comparison to control. Understory species composition was strongly altered by the snow cover manipulations; vegetation cover declined by more than 50% in the snow removal treatment. In particular, the dominant dwarf shrub Vaccinium myrtillus (-82%) and the most abundant mosses Pleurozium schreberi (-74%) and Dicranum scoparium (-60%) declined strongly. The C:N ratio in V. myrtillus leaves and plant available N in the soil indicated no altered nitrogen nutrition. Fine-root biomass in summer, however, was negatively affected by the reduced snow cover (-50%). Observed effects are attributed to direct frost damage of roots and/ or shoots. Besides the obvious relevance of winter processes on plant ecology and distribution, we propose that shifts in the vegetation caused by frost damage may be an important driver of the reported alterations in biogeochemistry in response to altered snow cover. Understory plant performance clearly needs to be considered in the biogeochemistry of boreal systems in the face of climate change.

Winter soil frost conditions in boreal forests control growing season soil CO2 concentration and its atmospheric exchange

AbstractThe impact of changes in winter soil frost regime on soil CO2 concentration and its atmospheric exchange in a boreal Norway spruce forest was investigated using a field‐scale soil frost manipulation experiment. The experiment comprised three treatments: deep soil frost, shallow soil frost and control plots (n= 3). Winter soil temperatures and soil frost distribution were significantly altered by the different treatments. The average soil CO2 concentrations during the growing season were significantly lower in plots with deep soil frost than in plots with shallow soil frost. The average CO2 soil–atmosphere exchange rate exhibited the same pattern, and differences in soil respiration rates among the treatments were statistically significant. Both the variation in soil CO2 concentration and the CO2 soil–atmosphere exchange rate could statistically be explained by the differences in the maximum soil frost depth during the previous winter. A response model for growing season soil respiration rates suggests that every 1 cm change in winter soil frost depth will change the emission rates by ca. 0.01 g CO2 m−2 day−1, corresponding to 0.2–0.5% of the estimated net ecosystem productivity (NEP). This suggests that the soil frost regime has a significant influence on the C balance of the system, because interannual variations in soil frost up to 60 cm have been recorded at the site. We conclude that winter climate conditions can be important in controlling C balances in northern terrestrial ecosystems, and also that indirect effects of the winter season must be taken into account, because these can affect the prevailing conditions during the growing season.

Infiltration in Icelandic Andisols: the Role of Vegetation and Soil Frost

ABSTRACTSoil frost formation, snow distribution, and winter/spring/summer terminal infiltration rates (TIRs) were quantified in Icelandic Andisols with contrasting vegetation cover types (grassland, spruce and birch woodland, lupine, and sparsely vegetated lava site). TIRs (mm h−1; determined with double-ring infiltrometers) were generally higher in unfrozen than in frozen soils (102–369 vs. 9–306, respectively in sandy soils; 28–94 vs. 3–72 in finer-textured soils) and differed between land cover types, being consistently highest in birch woodlands. TIR was an inverse function of soil frost depth. Lowest TIRs were associated with deep and dense soil frost, which formed in spruce woodland and grassland communities where snow depth was shallow. Results suggest conditions conducive to erosion by water are most likely to occur during winter warm spells and in spring in vegetation types where snow cover is low or ephemeral. Threefold increases in TIRs occurred one year after livestock grazing was discontinued, suggesting Andisols are hydrologically resilient where vegetation cover is relatively continuous and soil organic carbon content is high.

The influence of seasonally frozen soil on the snowmelt runoff at two Alpine sites in southern Switzerland

In spring, snowmelt releases huge quantities of meltwater, affecting the hydrology of Alpine areas. Seasonal soil frost influences these water fluxes by locally decreasing the infiltration capacity of the soil, resulting in an increased amount of surface runoff. The main goal of this study was to investigate the spatial variability of the seasonal frost depth and to quantify by how much this seasonal soil frost affects the snowmelt discharge. For this purpose, an extensive field study was run for two winter seasons (2000/2001 and 2001/2002) at Gd St Bernard (2470 m) and Hannigalp (2090 m) in the southern Swiss Alps. The different components of the water balance (lateral runoff, deep percolation, liquid soil water content) were measured on delimited plots of 5 m 2. The two winters investigated had opposing weather and soil frost conditions: in the first winter a thick snowpack prevented the formation of soil frost, whereas in the second winter little snow fell until January, which produced a deep and persistent soil frost. We classified the snowmelt events into several classes (mid-winter, late winter, spring and post-spring) and analysed the significance of the different water flow components for each melt situation. While 90–100% of melt water infiltrated into the ground during the first winter, 25–35% of melt water ran off laterally in the second, mainly during late winter and spring snowmelt events. In that second winter, the soil infiltration capacity was primarily reduced by the presence of a basal ice sheet after mid-winter melt events.

Increased soil nitrate losses under mature sugar maple trees affected by experimentally induced deep frost

This work reports and discusses data gathered during soil solution monitoring that was part of an experiment conducted in the Duchesnay Experimental Forest (Quebec, Canada) to study the effects of induced deep frost, superficial frost, and superficial frost plus drought on mature sugar maple trees (Acersaccharum Marsh.). Frost treatment was applied by preventing snow from accumulating under the canopy. Soil solution chemistry was modified when mature sugar maple trees declined after exposure to a severe deep frost. The first vegetation period after treatment showed that losses of NO3− below the rooting zone were greatly increased under affected trees. The leaching rate of NO3− and basic cations was directly related to the level of change in canopy dieback and transparency. A mean NO3− concentration of 630 μmolc•L−1 (53 times the controls; max. 4500 μmolc•L−1) was measured in soil solution under the deep frost treated trees. The leaching rate of K+ (18×, relative to the control) and Mn2+ (11×) was higher than that of Mg2+ (5×) and Ca2+ (2.6×). Acidification of the soil solution (50% more H+) as a result of intense nitrification caused an increase in aluminum concentration (5×) and a decrease in SO42−. The acidification during the year after treatment was equivalent to decades of atmospheric acid deposition. The seasonal mean of SO42− did not differ between treatments, but there was evidence of a significant correlation between pH and SO42− in soil solution. Concentration of NH4+ was also enhanced but to a lesser degree (10×) than that of NO3−. Specific conductivity was a good predictor of NO3−, Ca2+, Mg2+, and total Al in soil solution. The ion balance shifted from an anion deficit to a strong cation deficit when NO3− concentrations were high. Superficial frost with or without induced summer drought did not cause any significant change in soil solution chemistry compared with the controls. These results indicated the necessity to consider perturbations induced by extreme climate conditions, like deep soil frost, for the interpretation of soil solution chemistry data in the context of acid deposition studies and forest health.

Cold-induced dieback of montane spruce forests in the Swedish Scandes - a modern analogue of paleoenvironmental processes.

In 1987 particularly, but also in 1988, primaeval montane coniferous forests of northern Scandinavia experienced needle loss on an unprecedented scale, which affected the character of the entire landscape in certain regions. This study elucidates the dieback event For Norway spruce [Picea abies (L.) Karst.] in an area of the Swedish Scandes. The study includes visual assessment of percentage needle loss, age structure, radial growth and various environmental parameters, along altitudinal transects and within a larger sample plot. Frequency and magnitude of foliage damage were found to be related to elevation. The dieback process of closed forest stands was interpreted as a response to short-term climatic variability. Damage mainly developed during one of the coldest early winters (December January 1986/7) of this century. In combination with sub-normal depth of the snow caver, the severe cold promoted exceptionally deep and persistent soil frost in spruce forest stands. It is hypothesized that damage was mainly due to winter/spring drought (frost-drought). Phenotypically closed forest stands may change in just a few years in response to cooling, a feature which has not been realized previously. The possibility of major retrogressive responses lo short-term climatic events should therefore be considered in studies of historical biogeography and when modelling vegetation succession. The forest studied was probably established after a severe wind storm in 1837. The restocking process was facilitated by the post-Little Ice Age warming, but the stands are at present suffering severe dieback due to the recent cooling of the climate. The results stress the potential importance of short-term climatic disturbance in boreal forest succession. Reasonably, rapid cold-induced dieback of forest stands may have been a recurrent feature of Holocene forest history in the Boreal zone.

Soil Water Movement as Affected by Deep Freezing

Knowledge of water movement into and through frozen soils is needed for studying the hydrologic behavior of northern watersheds. Soil water was logged by the neutron method in both sandy and silty soils over four winters in southwestern Wisconsin. Bonded frost depth, which ranged from 8 to 120 cm at maximum penetration, was measured concurrently by frost probe or resistance blocks. Changes in soil water took place throughout much of the frost season, even with deep soil frost. One series of data showed that water may infiltrate and percolate through more than 60 cm of hard-frozen ground. However, frozen ground did impede percolation, causing a buildup of water in the frozen zone during spring melt. Although neutron readings increased at upper levels during the frost season, most of the increases could not be accounted for by decreases at lower levels. Thus, they appeared to result more from infiltration than from upward migration as reported by others.