Determining Flowering Timing and Intensity

Considering the seasonal development of allergenic plants, a special attention must be dedicated to early-flowering tree species and herbaceous taxa. For trees, the genera in Fagales (Betula, Alnus, Carya, Corylus, Carpinus, Juglans, Quercus) and the genus mulberry (Moraceae; Morus) together with sugi (Taxodiaceae: Cryptomeria japonica) in the boreal and temperate zones and olive (Oleaceae: Olea) in the Mediterranean climate are among the most common agents of seasonal allergies. Important herbaceous taxa include one of the largest plant family, the grasses (Poaceae) with about 640 genera and over 10,000 species (Mabbelrey 1987), widely distributed weeds, such as mugwort (Artemisia), ragweed (Ambrosia), pigweed (Amaranthus) and goose foot (Chenopodium), as well as pellitory (Parietaria) growing predominantly in the Mediterranean climate (www.pollen.com/Pollen. com.asp, www.polleninfo.org).

For many anemophilous plants, effective pollination and reproduction requires adaptations that make plants release pollen simultaneously. Photoperiod, especially the length of the diurnal dark period, has an important part in regulating the development of herbaceous plants (Flood and Halloran 1982; Hay 1990; Mahoney and Kegode 2004). Temperature is also important for the growth rate of herbaceous plants and the discrimination of seasons (Yanovsky and Kay 2003). For trees, a complicated interplay of changes in the light environment and temperature conditions controls the timing of events related to flowering and leaf bud burst. Dormancy, the state of arrested growth in autumn and winter, is released by exposure to chilling temperatures while the following bud development is driven by increasing temperature in spring. There is also recent evidence of importance of light environment for the onset of the bud development (Heide 1993a, b; Partanen et al. 1998, 2001, 2005; Linkosalo and Lechowicz 2006).

Flowering of wind-pollinated trees for most broadleaf species takes place before the leaf bud burst. The advantage is obvious: leaves would hinder the movement of pollen within the tree while the flowering before the bud burst enables more efficient spread of pollen. The timing of leaf bud burst poses a crucial optimization problem between maximizing the period of active photosynthetic production and minimising the risk of damage due to spring frost (Chabot and Hicks 1982; Kikuzawa 1989; Reich et al. 1992). As the flowering precedes the leaf bud burst, its time window is quite narrow and requires proper regional synchronization between the plants. High correlation between these phenological events within and between many boreal tree species seems to suggest that the trees utilise similar mechanisms to control the timing of the spring phenological events. Yet the timing of flowering between species varies somewhat. For example, in Southern Finland the mean date of birch flowering is in mid-May, while aspen flowers about two, and alder up to 4 weeks earlier (Linkosalo 2000).

For herbaceous species, the studies of common ragweed (Ambrosia artemisii-folia) and two mugwort species suggest a juvenile phase, during which it is not sensitive to photoperiod, but after that the light for about 14 h a day is needed for flower bud development and seed production (Deen et al. 1998; Sheldon and Hewson 1960; Edmonds 1979; Ferreira et al. 1995; Mahoney and Kegode 2004). In northern Europe, the pollen season of mugwort starts earlier in northern locations where the length of darkness grows more rapidly in late summer (Ekebom et al. 1997).

For all wind-pollinated plants, the weather parameters have significant impact on daily and diurnal patterns of pollen release. The release of mature grains is largely a physical process. Dry conditions, sometimes just a temporary decrease in relative air humidity, are needed for anther walls to dry out and anthers to split, thus allowing the pollen to release. A warm night accelerates the release while it can be almost inhibited by a cold one, especially in herbaceous species (Edmonds 1979; Sheldon and Hewson 1960; Solomon 2002).

After the pollen liberation, air flows agitate the source plant and/or lift the pollen free from the bounder layer surrounding the surface on which it rests. A few species liberate the grains actively, for example, mulberries and nettles catapult their pollen into the air (Solomon 2002). During the pollination period pollen dispersal is favored by warm and windy conditions, low relative humidity and absence of precipitation. Rain causes a break in pollination, removes pollen from the air and may, if continuing for several days or in combination with night frosts, even spoil catkins thus decreasing the total pollen yield. In general, any strong deviation of external factors from the regional climatologic norms (e.g. temperatures or rainfall above or below normal level) during the critical points of developmental phase may result in lowered growth and flower production, or abortion of flower buds (Sharp and Chisman 1961). In a long run, the pollination seasons of species may be able to become adapted to the new conditions through natural selection (Flood and Halloran 1982; Hay 1990).

The phenological phases and their key features can be quantified and simulated by means of models. The basis for phenological modelling was laid more than 270 years ago, when French scientist Reaumur (1735) first discovered, while trying to find out why cereals ripen at different times in different years, that the heat sum accumulated from April until the ripening date always reaches about the same value. This model is essentially similar to the so-called Thermal Time or Growing Degree Day model, that is still the most common model describing various phenological events. As Reaumur found out, most developmental events proceed at rate depending on ambient air temperature. Several experimental studies have since verified this finding, including the classical work by Sarvas (1972), who found that the relationship between ambient temperature and rate of progress of meiosis in pollen mother cells is constant for several tree species (Fig. 5.3a). The relationship follows a sigmoidal curve, and is rather similar to the linear dependence suggested by the Thermal Time models (Fig. 5.3b).

Sarvas claimed that the same relationship applies to the flower development of several boreal tree species. Even further, the good performance of similar models in describing both flowering and leaf bud burst of various species seems to suggest that the boreal trees utilise rather similar control mechanisms to drive all their spring phenological events. More support for this claim was given by Linkosalo (1999), who found using data from historical phenological time series, that the correlation between flowering and leaf bud burst of Betula is 0.97, while the two events take place on the average 1.1 days apart. Similar figures for the flowering of Betula and Populus are 0.83 and 22.7 days. The lower correlation in the latter case is due to the longer time difference between the two events.

Leaf and flower buds of boreal and temperate trees develop already during the previous summer. To keep the buds viable until the next spring, they need to reach a frost-resistant state in the autumn, keep it throughout the winter and avoid premature onset of development during that time. Various studies starting from the classical work of Coville (1920) have shown that buds fall dormant in the autumn - a state where the development is hindered even if the environmental conditions, most of all air temperature, are feasible. Dormancy is released by exposure to chilling temperatures, to certain changes in the light environment, or to a combination of the two (e.g. Sarvas 1974; Heide 1993a, b; Myking and Heide 1995; Leinonen 1996; Linkosalo and Lechowich 2006). Sarvas (1974) found the optimal temperature for

Fig. 5.3 The rate of ontogenetic bud development (forcing) as a function of air temperature according to Sarvas (1972) (a) and according to Thermal Time model (b). The rate of dormancy development (chilling) as a function of temperature according to Sarvas (1974) (c)

dormancy release to be around 3°C (Fig. 5.3c). He formulated a model to describe the sequence of chilling followed by the bud development. Thus the model is called "Sequential" (Hunter and Lechowicz 1992; Hänninen 1995).

Curiously, some more recent work (e.g. Häkkinen et al. 1998; Hannerz 1999; Linkosalo et al. 2000; Hänninen et al. 2006) have found that the standard Thermal Time model, where the bud development is assumed starting from a fixed date in spring, fits empirical data better than the Sequential model. It seems that even though chilling, as described by Sarvas, is required, it is not a sufficient condition for the dormancy release, but an additional environmental forcing is needed. Indeed, some studies on Betula indicate that light environment affects the dormancy release (Heide 1993a, b; Linkosalo and Lechowicz 2006). These results are for leaf bud burst, but the above-mentioned high correlation of leaf bud burst and flowering suggests that the events are controlled by same mechanisms, and thus the use of the same models is justified.

In the temperate climatic zone the role of chilling in dormancy release seems to have much more importance, as chilling temperatures are not as prevailing as in the boreal zone, so that the sufficient amount of chilling can be reached rather late in winter. Yet also Thermal Time model has been used with good success to describe the bud development of temperate trees (e.g. Kramer 1994; Chuine et al. 1999).

The reasons causing the year to year variation of the flowering intensity are less known. Yet the intensity may fluctuate from year to year by more than an order of magnitude. The variation is irregular but synchronised over large geographic areas (Koenig and Knops 2000). Accumulated heat sum of previous summer seems to correlate positively with flowering intensity, which is also physiologically reasonable as the flower buds are developed during the previous summer. However, even after taking this into account the variability of the results still remains large. There are several possible reasons for this. Firstly, environmental stress of non-meteorological origin can affect the flowering intensity. Secondly, the resource allocation within the tree varies from year to year, for example the allocation to the ripening seeds that affects the development of catkins and forces less pollen production in years following the strong flowering (Tuomi et al. 1982; Masaka & Maguchi, 2001). Thirdly, the pollination success depends heavily on stochastic environmental features, such as wind, rain and air humidity during the flowering (Peternel et al. 2005). These are hard to take into account in the phenological model and therefore they are often ignored despite potentially considerable impact on the pollen release. Finally, models describing the pollination intensity are typically based on empirical data collected with pollen traps, which are unable to distinguish between local pollen and pollen originating from more distant regions (Ranta et al. 2005).

Was this article helpful?

0 0
51 Ways to Reduce Allergies

51 Ways to Reduce Allergies

Do you hate the spring? Do you run at the site of a dog or cat? Do you carry around tissues wherever you go? Youre not alone. 51 Ways to Reduce Allergies can help. Find all these tips and more Start putting those tissues away. Get Your Copy Of 51 Ways to Reduce Allergies Today.

Get My Free Ebook


Post a comment