Published on December 17, 2007
Slide1: Forecasting of the birch flowering characteristics Factors affecting the amount of pollen emitted in an area: Factors affecting the amount of pollen emitted in an area amount of birch forest (B. pendula, P. pubescens) annual variation of flower production 3. progress of pollen release (flowering), deposition during flowering Factors affecting the amount of pollen emitted in an area: Factors affecting the amount of pollen emitted in an area amount of birch forest (B. pendula, P. pubescens) 1.1. biogeography; dominating broad-leaved genus in N Europe, northern parts of C-Europe (Baltic countries), NW Russia and Byelorussia common in central parts of C Europe, rare in Mediterranean vegetation zone Slide4: Factors affecting amount of birch forest 1.2. land use; large forest areas in sparsely populated areas in N Europe, some parts of C-Europe, large parts of NW Russia and Byelorussia cattle herding has nealy destroyed birch forests in Scotland and Ireland, agriculture in Poland etc. 1.3. forestry practise; clear-cuttings and abandoned fields are regenerating naturally with broad-leaved trees (Baltic countries, NW Russia and Byelorussia), in Nordic countries pine/spruce are favoured Slide6: Annual Average pollen sums Sources of information:: Sources of information: Images of Broadleaved and needle-leaved forests, European Forest Institutes, EFI National forest inventories (National sites & publications, links on EFI www-pages) Digital maps for Forest Tree Species in Europe (Köble & Seufer 2001) Modified digital birch forest map for N Europe by Pilvi Siljamo other forest inventories and remote-sensing inventories, many are freely available 2. Annual variation of flower production: 2. Annual variation of flower production Background: Birch (as most trees in boreal and temperate zones) exhibit masting-flowering amount of flowering varies greatly and irregularly variation is synchronous over large geographic areas (at least radius of 500 km) different tree species often flower synchronously variation is more pronounced in northern latitudes patterns of mean temperatures and precipitation exhibit similar patterns Slide10: Synchronization of flowering Factros affecting the amount of flowers: Factros affecting the amount of flowers Bacground: Catkins (flowers) are formed during spring-early summer one year before pollinating season. The amount of flowers depends on availability of resources by the time of flower production Slide13: Which in turn is affected by weather parameters; temperature and sunshine-hours ---- effectiveness of photosynthesis, (maybe also hormonal regulation of flower initiation) amount of developing seeds they consumes resources, and affect the amount of leaves (one catkin replaces one leave) Slide15: Simply: a very abundant year can’t be preceded by a similar one because of heavy crop of developing seeds. low year is not preceded by a high year if the weather is poor. there may be several moderate or low years in row due to weather conditions Slide16: However, There is always some pollen production because young trees in good sites flower every year. In very abundant years young and old individulas of both birch species flower abundantly also in less favourable site, and that is why the pollen season is also long in very abundant years Slide17: Resource budget model predicts the amount of flowering with monthly mean temperature and amount of sunshine-hours in spring-early summer in year t-1 amount of male flowering in year t-1 amount of male flowering in year t-2 Applicable to wide area: explains over 90 % of annual variation in Japan (Masaka & Magushi, Annals of Botany, 2001) 75-92 % of variation in S and C Finland (Ranta et al., Int. J. Biomet., in press over 80 % of variation in Sweden (Dahl & Strandhede, Aerobiology 1996) Slide18: Variable Coefficient P Xt-1-1 -1515917356 0.0108 Tt-1St-1X t-1-1 291270 0.0025 Xt-2Xt-1-1 -2625 0.0040 Rt-1-1 Xt-1-1 1130861 0.0545 Final model 0.0018 R2 = 0.924 Total F = 28.44 Kangasala, C Finland Slide19: Variable Coefficient P Xt-1-1 -283335425 0.0578 Tt-1St-1X t-1-1 75233 0.0092 Xt-2Xt-1-1 -5716.850 0.0053 Rt-1X t-1-1 613041 0.0578 Final model 0.0102 R2 = 0.763 Total F = 9.07 Kuopio N C Finland Slide20: Sources of pollen data EAN (European Aeroallergen Network) pollen data for many European countries (nordic countries, Baltic countries, Germany, Austria, Switzerland, England, Poland, Tshecz, UK, France, Spain). Baltic data-sets are short, Estonia is problematic, no data for Byelorussia) Russian pollen data; Moscow, (St Petersburg ?) Catkin data Annual inventories of the Finnish Forest Research Institute (Metla’s Phenology project, Hokkanen) Similar data from C Europe ? Slide21: Weather data ECMWF (Europan Centre for Medium range Weather Forecasts) ERA-40 meteorological re-analysis over the past 40 years ECMWF forecast archive. National Oceanic and atmospheric Administration web site (ftp://ftp.ncdc.noaa.gov/pub/data.ghcn) Slide22: Herrera C M, Jordano P, Guitian J & Traveset 1998. Annual variability in seed production by woody plants and the masting consept: reassessment of principles and relationship to pollination and seed dispersal. The American Naturalist, 152: 576-594. Koenig W D & Knops J M H 2000. Patterns of annual seed production by northern hemisphere trees: A global perspective. The American Naturalist, 155: 59-69. Masaka K & Maguchi S. (2001) Modelling the masting behaviour of Betula platyphylla var japonica using the resource budget model. The Annals of Botany 88:1049-1055. Ranta H, Oksanen A, Hokkanen T, Bondestam K & Heino S. Masting by Betula-species; applying the resource budget model to North European data sets. International Journal of Biometeorology, in press. Schauber EM, Kelly, D, Turchin P, Simon C, Lee WG, Allen RB, Payton IJ, Wilson PR, Cowan PE & Brockie RE. (2002) Masting by eighteen New Zealand plant species: the role of temperature as a synchronizing cue. Ecology 83:1214-1225. 3. Progress of pollen release (flowering), deposition during flowering: 3. Progress of pollen release (flowering), deposition during flowering weather conditions during the flowering; Consentrations peak quickly if the weather is warm and sunny Chilly weather prolongs flowering , rain spells interrupt it Synchrony of the two species Conclusions: Conclusions amount of birch forest defines the limits for pollen production in an area (minimum and maximum annual pollen sum from long data-sets give limits what to expect) latitude and heigth gradient are important in extreme climate conditions (small trees in northernmost Europe Slide27: preceding years weather and flowering can be used to predict the magnitude of next years’ flowering weather conditions affect the speed of pollen release during the flowering How accurate the emission source data must be ?: How accurate the emission source data must be ? view point of allergic patient ?: Very low concentrations (< 10 pollen grains in m-3) cause symptoms to most sensitive patients ---- very small amounts of LTD-pollen may be important. They can be produced during flowering in any year in nearly any area in Europe ? Counts over 100 gains in m-3 cause symptoms to nearly all patients ----- exceeded during several days every year in Finland Slide29: dose dependency in allergic diseases ? problems: in reality low concentration LTD-episodes can be mixed with local and LTD alder and hazel pollen, even Salix ----- they have similar allergens, flower earlier release of small allergenic particles from local birches and other species there may be LTD-pollen in areas where birch does not flower yet Suggestion: Two-stage warning system for LTD birch pollen: Suggestion: Two-stage warning system for LTD birch pollen 1. Risk of LTD-pollen, no reference to amount (air currencies come from any area where birch forests exist and where flowering is in progress) 2. Specified warning with reference to amount, for example: low, moderate, abundant ? - areas with large birch forests - areas where abundant flowering is expected (min/max annual pollen sums of long data-sets give limits what to expect; they very very much in different parts of Europe) Objects to be studied : Objects to be studied areal variation of pollen production variation in annual flowering becomes more pronounced towards north (east ?); - do we have to make annual flowering predictions for all parts of Europe or can the amount of birch forest be directly used as a measure of expected flowering ? spatial correlation of synchronization in birch flowering – for how large areas flowering forecasts are valid ? Slide32: - average pollen sums in different parts of Europe, minimum and maximum pollen loads; how much pollen can be expected from certain area TOOLS: spatial and temporal autocorrelations (database: EAN pollen data, latitude-longitude co-ordinates, distance data WHERE ?, should we analyse grid points or transects) Slide33: areal prediction models of the amount of flowering the relationship between areal variation in annual pollen sums and amount of birch forest comparison of modelled data versus real weather data; can modelled data be used to predict the amount of coming flowering Tools: Resource budget model, database; real and modelled weather data (monthly mean temperatures and sums of sunshine hours and precipitation), EAN pollen data Slide34: translation of pollen data into catkin data, do we have to do this ? can we use expected average pollen loads in different areas of Europe (average, min/max) specified with prediction of flowering intensity ? the relationship between percentage of birch forest and annual pollen loads to calibrate areas with no pollen data Tools: Correlation between average pollen loads and amount of birch forest, correlations between Finnish pollen data and catkin inventories (they correlate), similar data from C Europe ?