Fig. 9. Number of days with frozen soil conditions at Helsinki for various soil layers for [A] current temperature conditions (1960-1990), [B] when temperature has elevated +4 degrees, and [C] and [D] at Rovaniemi, respectively.
Background to the study. The frequency of strong winds has increased since the early 1960s especially during unfrozen soil conditions in Finland. This is assumed to be related to changing climatic conditions. At the same time, the structure of the forest ecosystem has become more prone to wind damage; e.g. because of increasing thinning intensities and the preference given to clear felling. This structural change in the forest ecosystem, combined with the increasing frequency of strong winds, especially during those months of the year when the soil is unfrozen, could substantially increase timber losses in the future. Furthermore, the more humid and warmer weather pattern expected in the future is also expected to increase the risk of windthrow for treesbecause of reduced soil freezing, which until now has enhanced trees’ anchorage from late autumn up to the early spring, during the most windy months of the year. In this context, this subproject was aimed at studying the mechanism of wind induced damage on Scots pine (Pinus sylvestris L.), i.e. by developing (i) a simulation model for the mechanism of windthrow and stem breakage of Scots pines in stand edge conditions (for static wind load), by studying (ii) tree swaying as caused by dynamic wind loading in field conditions, by studying (iii) the risk of windthrow for Scots pine in terms of the turning moment caused by dynamic wind loads along the margins of clear-felled areas using a model approach, and by outlining (iv) the implications of the occurrence of soil frost and its depth in forest soils, as modified by the warming climate, and consequent increase of the risk of windthrows due to changes in tree anchorage.
Material and methods. Within the scope of this subproject, a mechanistic model for wind damage of Scots pine was developed in orderto fully describe the mechanistic behaviour of trees exposed to wind loading. The model developed was aimed at determining the windspeed required to uproot single trees or to break tree stems. The model is theoretical and based on the physical properties of trees and vertical wind profiles along stand edges ( Peltola and Kellomäki 1993, Peltola 1995).
Tree swaying caused by dynamic wind loading was also studied along the edge of a stand of Scots pine as well as within the stand by means of two field experiments ( Peltola et al. 1993, Peltola 1995, 1996a). Wind and tree swaying measurements (i.e. mean wind profiles and stem displacement measurements) made along the edge of a stand of Scots pine ( Peltola et al. 1993) were used especially in validating the simulation model thus developed ( Peltola and Kellomäki 1993). Wind and tree swaying measurements conducted both along the stand edge and a distance of two tree heights into a stand of Scots pine, before and after the first thinning (2700 and 1500 stems per ha), concentrated on the relationship between windspeed and the resulting stem displacement using spectral analysis technique ( Peltola 1995, 1996a).
This subproject also involved model computations in order to evaluate the risk of windthrow of Scots pine along the margins of clearfelled areas by evaluating this risk in terms of the turning moment arising from the dynamic wind load (Peltola 1995, 1996b). The turbulent wind field across the forest clearing and within the stands at the clearing margins was simulated using a two-dimensional model developed elsewhere (see Peltola 1996b).
Furthermore, the impacts of frost decrease in forest soils on the risk of windthrow was also evaluated in a stand of Scots pine bothin southern Finland (Helsinki region) and in northern Finland (Rovaniemi region), respectively (Peltola et al. 1996a). Soil frost was simulated using the FinnFor model developed by Kellomäki et al. (1993) and it was compared to current windspeed statistics available for the period 1961-1990. In frost simulations, the present mean annual temperature was assumed to increase by 2-4oC.
Results. In a tree swaying study ( Peltola 1995, 1996a), it was found that nearly equal wind energy transfer and damping of the system occurs between the two stand densities studied. However, a clear difference was observed between trees located along the stand edge and those located within the stand. This means that trees growing along the stand edge (and especially along a newly cut edge) are more liable to wind loading than trees within the stand. On the other hand, with respect to the stand densities studied, neither were trees along the stand edge very likely to be damaged.
According to computations made using a mechanistic wind damage model developed within this subproject by Peltola and Kellomäki (1993), the windspeed required to blow down a tree or break the stem of a tree located along the stand edge decreased if the height to diameter ratio or the crown to stem weight ratio of the trees increased (as well as when the tree size increased ). The windspeed required to uproot a tree was much smaller than that required to cause the stem to break. On the other hand, even windspeeds of 12-14 ms-1 were found to be strong enough to uproot Scots pines (slender individuals ) located along the stand edge ( Peltola and Kellomäki 1993).
In addition, based on the model computations by Peltola (1995, 1996b), stand density and height were found to affect mostly the windspeed and turning moment on trees located along the stand edge, i.e. it decreased as stand density increased and increased as stand height increased. Thus, the risk of uprooting increases also sharply with increasing tree height and the differences between various stand densities increases also along with height increase. On the other hand, the difference in windspeed between various clearings of different sizes (0.04-4.0 ha) was only some percent for the same stand height and stand density along the stand edge. However, the turning moment decreased quite substantially when the distance from the stand edge increased, and the decrease was greatest at the dense margin and within the distance of one tree height from the edge into the stand (30%). According to the results obtained in this study (Peltola 1995, 1996b), the risk of uprooting might be even greater for trees at the margins of smaller clearings, because of the much greater length of perimeter at risk.
Until now, frozen soil has increased trees’ anchorage during the time of year usually characterised by strong winds, i.e. from late autumn to early spring. In the future, especially in southern Finland, the duration of soil frost may decrease from 4-5 months down to 2-3 months, if 2-4oC is added to the present mean annual temperature (Peltola et al. 1996a). Furthermore, it seems that the number of days when the soil is frozen may decrease substantially more in the deeper soil layers (40-60 cm) than near the ground surface (0-20 cm), especially in southern Finland. Similarly, in northern Finland, the number of months when the soil is frozen may decrease from 5-6 months down to 4-5 months (Fig. 9). In northern Finland, the same kind of dramatic change in the number of days when the soil is frozen as in southern Finland is not evident, not even in deeper soil layers. On the whole, the improved stability of forest trees from late autumn to early spring due to soil frost may substantially decrease in the future, thereby evidently increasing the risk of windthrow. This is because the number of strong winds during in unfrozen soil conditions seems to substantially increase. Nowadays, up to 45 % of the strong winds occur during months when the soil is frozen (i.e. >15 days per month when soil frost occurs in soil layers of 0-40 cm) in southern Finland, whereas in the future this percentage is expected to be only ca. 20 %. In northern Finland, the corresponding percentage of days today is 60 %, and in the future 50 %.
Discussion of results. A more humid and warmer weather pattern than today can make Scots pines (as well as other tree species), and especially in southern Finland, far more liable to windthrow during winter and spring storms than is the case nowadays because of a substantial decrease in soil frost and thus of weakening of the anchorage to the soil of trees. This risk will be even more evident especially if the air temperature during winter months increases by as much as 6-8oC, as has been suggested, thereby further decreasing the occurrence of soil frost. Furthermore, changing climate may also increase the frequency (as well as intensity) of storm activity in northern latitudes and increase the risk of wind-induced damage even more than can be expected based on the current wind climate.
In the future, the mechanistic wind damage model( Peltola and Kellomäki 1993, Peltola 1995, Peltola et al. 1996b) developed within this subproject can, for example, be used to study how thinning intensity and its timing affect critical windspeeds under various stand conditions. The model will also be applicable to tree species other than Scots pine (e.g. Norway spruce and birch species) with different tree, stand and site characteristics applying to trees located along the stand edge as well as within the stand, through the modification of the controlling equations and parameters.In addition to wind loading, the model can also be used to determine the snow load required to damage single trees. The model will be validated by tree pullings (static force) made mostly in the autumn of 1995 and windspeed profile and tree swaying measurements from the years 1991-1996.
Fig. 9. Number of days with frozen soil conditions at Helsinki for various
soil layers for [A] current temperature conditions (1960-1990), [B] when
temperature has elevated +4 degrees, and [C] and [D] at Rovaniemi,
respectively.