(FY1993-FY1995)
We have triedto reconstruct past climatic conditions by means of alpine-subalpine meadowsoil stratigraphy, which records past snow disappearance time in nivationhollows. Buried peat found on Mt. Zarumori indicates that the date of snowcover disappearance in approximately 1000 y.B.P., was earlier than in recenttimes. Calculations of the past snow melt rate by the degree-day method,indicate that warming in summer and lower winter snowfall accumulation,hastened snow melt and enabled the peat to deposit. This warmer climaticcondition in north Japan corresponds to the so-called "Medieval Warm Period"of Europe and North America. It is suggested that subalpine conifers neverexpand into the alpine vegetation zone which P. pumila has alreadyoccupied, even under the higher temperature conditions.
(Key Words: climate change, alpineand subalpine vegetation, phenology, snow cover)
The second studyarea includes the upper part of the Kitakami Mountains. It is located ona lifted peneplane, and most of the mountains here have gentle slopes.In this study, locations of sample pits were chosen from the mountainsof higher than 1200 m in elevation, and sampled at the stable place frompedoturbation, and these are higher than 1100 m.
The study ofPinus pumila scrub was carried out on Mt. Kinpu (2595 m) in the Okuchichibumountain range of central Japan. The uppermost part of Mt. Kinpu(above ca. 2450 m) was dominated by a dwarf scrub of P. pumila,below which (above ca. 1700 m) was represented by subalpine conifer forestsconsisting mainly of Abies mariesii, A. veitchii and Tsugadiversifolia. Six research plots (1-2 m2 in area)were established in the pine scrub; three plots were on the north facingslope at different elevations (N1: 2575 m, N2: 2560 m, N3: 2540 m)and the other three plots were on the south facing slope (S1: 2580 m, S2:2560 m, S3: 2550 m).
In the AkaishiMountains from Sawarajima to Fujimitaira in Shizuoka Prefecture, edaphicconditions are studied. The elevation of study sites ranges between 1120m (Sawarajima) and 2725 m (Fujimitaira) above sea level. The Akaishi Mountainsmainly consist of Mesozoic sedimentary rocks which were modified by Shimantoactivity. Temperature varies depending on topography and elevation withan annual mean of 4ºC-5ºC at 2000 ml. Annual precipitation isabout 2600 mm at the same elevation. Seven fixed quadrates were setat each elevation, i.e., plot No. 1 at 1120 m, plot No. 2 at 1560 m, plotNo. 3 at 1840 m, plot No. 4 at 2130 m, plot No. 5 at 2470 m, plot No. 6at 2560 m and plot No. 7 at 2725 m. Landforms are classified into fivetypes, i.e., alluvial toeslope, convex creepslope, transportational midslope,seepageslope and interfluve. Soil types which appeared in these experimentalplots are BD, BC (PDIII), BA (PDIII),PDII, PDI. Forest types are classified into Larixkaempferi, Quercus crispula, Tsuga diversifolia, Abiesveittchii, Abies mariesii and Pinus pumila with understoryvegetational types of macro herbs, Dryopteris-Cacalia, moss,and Rhododendron.
The relationshipsbetween change of snow rate and air temperature are studied in Tokamachi,Niigata Prefecture, based on climatic observation from 1963 to 1994.
Changing snow environment andsnow patch grassland
A series of buriedpeat layers were found in the Ou Mountains area, and the paleoenvironmentwhich enabled their deposition was estimated, based on the understandingsof relationships between air temperature, snow melt rate, and vegetationand soil zonings. The simulation of snow melt process estimated thatsnowfall change was more affective than changes of air temperature in summer.This suggests that snowfall in mountain area has changed due to Holoceneclimatic changes. The observations of micro landforms formed by snow gridingand measurements by snow gride-meter showed that snow griding process occurson the treeless slopes steeper than 20 degrees, and it affects distributionof mountain forests. Snow gride was seemed to affect the distributionand expansion of subalpine trees. Abies mariesii, the dominant speciesof subalpine forest in northeastern Japan, tends to distribute on slopeswith less snow pressure. If the warming will be accompanied by decreasein snowfall, it will bring the more suitable environment for expansionof the Abies mariesii, which has continued since ca. 1000 y BP.
Distribution pattern of plantcommunities and phenology around snow patch
At the tops ofMt. Genta, Mt. Yumori and Mt. Chausu in subalpine area of the Ou Mountains,the relationship between soil freezing and distribution pattern of plantcommunities was studied in connection with snow depth. Under Pinuspumila, Tsuga diversifolia, and the windy site communities ( ex. Juniperuscommunis var. nipponica), the surface soils froze deeply, whileAbiesmariesii little. The results of the survey at Mt. Yumori showthat snow depth controls soil freezing. Therefore, in this studyarea, it is presumed that soil freezing corresponds to the distributionpattern of the plant communities.
The phenologyof snow patch vegetation (Primula nipponica, Fauria crista-galliand Hemerocallis dumortieri var. esculenta), snow melt andweather conditions were observed at Mt. Zarumori in Akita Prefecture, duringthe summer in 1993. Snowmelt and subsequent soil temperature werecalculated by a micrometeorological model. They showed two growingpatterns of snow patch vegetation, and an index of climate resource forsnow patch vegetation.
Effects of external factors ongrowth and structure of Pinus pumila scrub
Site differencesin microclimates and productivity due to slope aspect and elevation wereassessed for the alpine dwarf pine (Pinus pumila) scrub on Mt. Kinpu,central Japan. Leaf area index, aboveground biomass and annual productionrates were larger for the pine stands on the south-facing slope than thoseon the north-facing slope. For each slope, these parameters increasedwith decreasing elevation. With respect to environmental factors,the daily photon flux density during the summer months were almost thesame between the N- and S-slopes, but the temperature conditions differeddue to slope aspect, especially soil-temperature, which was 2-4°C higheron the S-slope. The positive correlation between temperature andproductivity indicated that stand development of P. pumila scrub couldbe enhanced under higher temperatures. The effects of recent globalwarming on the Japanese alpine/subalpine vegetation dynamics were discussedby focusing on habitat competition between P. pumila and other subalpineconifers.
Effects of environmental conditionson the site dynamics of a subalpine forest
Data were collectedand measured on microclimates, stand structure and growth, compositionof understory vegetation, litter decomposition rate, soil propertiesand tree regeneration at seven fixed plots. Air temperature in Novemberdecreased linearly with increasing elevation because of adiabatic expansionof air with its ascent. Soil temperature showed a clear decrease when elevationincreased, except in the highest plot (2725 m). The pH values in the topsoilindicated a negative correlation with elevation in which the pH valuesdecreased as elevation increased. The total carbon content in the topsoiltended to increase with elevation, except in the highest plot. The relationshipbetween elevation and the decomposition rate did not show a clear negativecorrelation due to that it is affected by landform and vegetation.
Relationships between change ofsnow rate and air temperature
Using Degree-daymethod and distinction of snowfall and rainfall by ground temperature,we estimate the change of snow water equivalent by ground air temperatureand precipitation for 31 years from 1963 to 1994. Calculated ground airtemperature for distinction was 1.4 degree. And calculated averaged snowmelt coefficient was 4.09, and had relationship with changing ofsnow melt period. We compared snow melt rate calculated by degree-daymethod and heat balance method with observed snow melt obtained by lycimeter. Degree-day method can be used in warm snow area using averaged snow meltcoefficient except with very low snowfall years.
Changing snow environment, snowpatch, and vegetation
How much mustaverage summer temperature and snow melt rate change to affect zoningof soil types in nivation hollows? The degree-day method providesthe best and simplest way to evaluate snow melt rate by changing temperature.We measured air temperature at 1-hour intervals, and surface snow meltto calculate a 'Degree Day Factor'. Effective cumulative temperature(E.C.T.) is the sum of daily average temperatures which exceed -3 degreeplus 3 degrees. The relationship between surface snow melt and E.C.T. isaccurately represented by the linear equation, Y=9.3X , where Y is surface snow melt (mm), X is E.C.T. and the Degree Day Factoris 9.3. Using this equation, we can evaluate snow melt under changingtemperature and snow depth.
The cumulativesurface snow melt curves in summer under the changing temperatures weredrawn using the above equation. They are also simulation curves for snowdisappearance when we regard the snow thickness axis (vertical axis) assnow depth in early summer (June 1st). The average summer temperaturein 1993 was 2°C lower than the past 70 years at Akita. Therefore eachcurve shows the case of +2,0,-2 and -4°C temperature changes from thepast 70 years. If average summer temperature rises +4°C higherthan 1993 (+2°C higher than the past 70 years), snow cover will disappearin latest July , almost the latest time at which peat can deposit. Howevera +2°C rise will be too large a change, considering that a climaticchange usually accompanies a change in snowfall. If the Medieval Period'swarming was with decreased snowfall, a smaller temperature rise will beenough for peat deposition. Fig. 5 shows snow deposition in the Mt. Zarumorinivation hollow, in the summer of 1993 and 1994. The shape of snow depositionresemble a delta, and positions of front-end (forest) are very differentfrom each other. So if the amount of snow fall decrease, the position ofthe front-end will retreat and snowdepth will change dynamically near thefront-end. We should further consider changes in snowfall when wediscuss the past and future vegetation of Japanese mountains.
There also existedmany fine charcoals on most surfaces of the soil profiles. Therewere probably more effects from human activities, especially burn-off,than forest fire. This is because the Kitakami Mountains have been famousas a production center of horses, and in the last part of the Edo period,pastures and meadows were moved from the low lands to the upper part ofthe mountains. Therefore, the natural vegetation was probably burned offfor pasturing.
Judging fromchanges of soil color in the profiles, it appears that the accumulationof humus began sometime around the To-Cu tephra fall. To confirm this hypothesis,carbon in each soil horizon was quantitatively analyzed. As the carboncontent began to increase (more than 5-6%) simultaneously from about theTo-Cu horizon in many profiles, it almost corresponds to the soil color.But there also is another type of profile, Aomatuba A had been more than7% of carbon content from bottom of the profile. According to a studyof soil pollen analysis on the same samples, even the warm period ca. 5500y.B.P., there were only a few shrubs, herbs and ferns in most of the subalpinearea, while at the tops of Sakudogamori and Aomatsuba A were deciduousbroad-leaved forests.
From these results,we believe that the accumulation of humus in most of the study area beganafter the To-Cu tephra fall. The accumulation of humus beganafter the To-Cu tephra fall in most of the subalpine area of the KitakamiMountains. However, in some stable places, it began before the To-Cu tephrafall, and it is believed that Abies mariesii had colonized theseplaces.
Effects of external factors onPinuspumila scrub
In the EuropeanAlps, the effects of microclimates on the photosynthesis of subalpine conifertrees have been well examined on two contrasting slopes (e.g., north- andeast-facing) of the same elevation. Turner et al. (1983) indicatedthat the annual net carbon gains of two pine trees (Pinus montana, P.cembra) were much less on the northern slope than on the eastern slopedue to the limitations of soil-temperature and total photosynthesis periodon the N-slope. Hasler (1982) also showed that the average dailyCO2 uptake rates of Larix decidua seedlings on the northernslope were about half of those on the eastern slope irrespective of similarirradiance input, and suggested that the lower soil temperature level onthe N-slope was the main limiting factor in photosynthesis. Similarly,the productivity of P. pumila stands was greater on the south-facingslope than on the northern slope. This might be due in large partto the higher summer temperatures in the S-slope, especially soil temperature,because the differences in microclimates due to slope aspect were moreconspicuous in temperature than in irradiance regimes.
In subarcticregions, the upward shifts of tree-limits due to extensive seedling establishmentin the warmer years of the 20th century have been observed for Piceaglauca (Payette & Filion, 1985) and P. abies (Kullman, 1986).Temperature increase caused by recent global warming, which was noted (Hansenet al., 1981) in the 1980s, may promote the migration of Japanese subalpineconifers (A. veitchii, A. mariessii and Tsuga diversifolia)into higher elevations. However, the positive correlation betweentemperature and productivity mentioned above indicates that stand developmentof the P. pumila scrub could be enhanced under higher temperatures. Thus, in the Japanese alpine regions, the regeneration success of subalpineconifers primarily depends on habitat competition with P. pumila.
Kimura (1963)showed that the relative light intensities on the floors of subalpine fir(A. veitchii, A. mariessii) and hemlock (T. diversifolia)forests averaged 5.9-6.5% where seedlings of each species could survive. Kohyama (1983) suggested that A. veitchii and A. mariesiiseedlings required a relative light intensity of 5-8% for survival and27-28% for maintaining normal growth-rates to reach the sapling stage. Generally, natural regeneration of the Abies and Tsugacan progress only in open sites after gap-formation. These resultsin relatively even-aged forests, such as "wave-regenerated Abiesforests" and "mosaic-patch Tsuga forests" (Kanzaki, 1984). In the P. pumila stands at Mt. Kinpu, light conditions under thecanopies, 15-39% in relative PFD, may allow the establishment of such coniferseedlings.
Kajimoto (1989)showed that mature P. pumila stands (ca. 100 yrs-old in the abovegroundage) in the Kiso Mountains had LAI values exceeding 5 m2m-2,in which the relative PFD values on the floors were below 10%. He also indicated that gap-formation due to individual mortality had rarelyoccurred within the mature P. pumila scrub because of its vegetativepropagation by layering). Such canopy structure and reproduction habitsin the well-developed pine scrub probably act defensively against invasionfrom other subalpine conifers. Thus, it is impossible to predictfuture vegetational changes in the Japanese alpine/subalpine areas, i.e.,upward expansion of the subalpine conifer forests to succeed the P. pumilazone, based solely on temperature increases resulting from global warming.
Effects of environmental conditionson the site dynamics of a subalpine forest
The air and soiltemperature conditions in alpine and subalpine zone have been discussedtaking into account the freeze-thaw cycles and the formation process ofmicrolandform. Although air temperature decreases nearly linearly at thelapse rate with the elevation, its decreasing tendence fluctuated dependingon season in this study. Air and soil temperature conditions are affected by direct sun light, snow fall, vegetation, soil water condition,organic matter and landform. The results of this study showed the factthat soil temperature of plot 6 where is windy and has lower and sparseforests as lower than that of plot 7 with higher elevation.
Temperature andsoil fauna and flora affect litter decomposition rate. Among of thesefactors, temperature is most important. As already described above, airand soil temperatures decreased with increasing the elevation. Therefore,litter decomposition rate decreased gradually with increasing the elevation.We found that pH in the topsoils decreases with the elevation, affectingthe litter decomposition rate. Values of pH also decreasedwith the elevation in the alpine of the Beartooth Plateau of Wyoming andMontana, U.S.A.(Johnson & Billings, 1962). It is considered that airand soil temperatures decrease with elevation and litter decompositionrate also decreases according increasing elevation. According these facts,Organic matter accumulated at higher elevation and soil water and soilbecomes acidic.
If the globalwarming occurs, litter decomposition rate will be accelerated at higherelevation in subalpine area. Therefore, the acidity of topsoil will becomemore neutral and then subalpine site condition (Akaishi mountains) willchange, habitat of Tsuga diversifolia which distributes at the ridge ofPodozolic soil will change and Abies and other species will invade thishabitat. Vegetational changes around snow patch may be little ifthe accumulation of snow was not decreased so much. Distributionof Pinus pumila scrub closely correlates with the distribution of frozensoil. Certain species of alpine meadow are affected by the effectof day length more than the increasing temperate.