Faculty of Natural Sciences

Differences in vocalization between subspecies of songbirds may influence intraspecific interactions including mate choice and territorial behavior, and contribute to their further divergence by strengthening pre-mating reproductive isolation. We have recently shown that songs of two Bluethroat subspecies, Luscinia svecica svecia and L. s. cyanecula, differ in spectral characteristics and structure. Here, in a first experimental test of whether these differences are biologically relevant, we compared responses of L. s. svecica males to both types of songs. Altogether 19 males from a subartic population in Abisko, Sweden, were exposed to playbacks of songs from the two subspecies, accompanied by display of a stuffed Bluethroat lacking subspecies-specific colour marks. These stimuli were presented in random order, separated by over 1 h. We evaluated two largely uncorrelated components of male responses: non-vocal territorial behavior, and singing activity. A significant difference in reaction to subspecies playbacks was observed in aggressive behavior: independently of the song order, males generally responded more strongly to playback of their own subspecies. In contrast, playback order rather than subspecies identity significantly influenced the singing activity, which usually increased in the second trial. A more detailed analysis nevertheless suggested that vocal responses of birds exposed first to playback of their own subspecies were stimulated in the subsequent trial but not vice versa. Our results show that Bluethroats clearly discriminate the two song types, indicating the potential for subspecies recognition. Further experimental work is needed to assess the general validity of these patterns. Lautäußerungs-Unterschiede zwischen Unterarten von Singvögeln beeinflussen möglicherweise intraspezifische Interaktionen wie Partnerwahl und Revierverhalten und tragen zur weiteren Ausprägung der Unterschiede bei, indem sie die reproduktive Isolation vor der Paarung verstärken. Kürzlich haben wir gezeigt, dass sich die Gesänge zweier Blaukehlchen-Unterarten, Luscinina svecica svecica und L. s. cyanecula, in den Charakteristika und der Struktur ihrer Spektren unterscheiden. In einer ersten Untersuchung zur möglichen biologischen Relevanz dieser Unterschiede verglichen wir die Antworten von L. s. svecica-Männchen auf die Gesänge beider Unterarten. Dafür wurden insgesamt 19 Männchen einer subarktischen Population aus Abisko, Schweden, Aufnahmen der Gesänge beider Unterarten vorgespielt und dazu ein ausgestopftes Blaukehlchen ohne die für die betreffende Unterart spezifischen Farbmuster gezeigt. Diese Stimuli wurden in zufälliger Reihenfolge und in Abständen von mehr als einer Stunde dargeboten. Wir evaluierten zwei weitgehend unzusammenhängende Antworts-Komponenten der Männchen: Revierverhalten ohne Lautäußerungen sowie Gesangs-Aktivität. Ein signifikanter Unterschied in den Reaktionen auf das Vorspielen der Gesänge der Unterarten konnte in aggressivem Verhalten beobachtet werden: im Allgemeinen reagierten die Männchen stärker auf die Gesänge ihrer eigenen Unterart, unabhängig von der Reihenfolge, in der ihnen die Gesänge vorgespielt wurden. Im Gegensatz dazu beeinflusste die Reihenfolge stärker als die Identität der Unterart signifikant die Gesangs-Aktivität, die in der Regel beim zweiten Durchgang anstieg. Eine detailliertere Analyse legte jedoch nahe, dass vokale Reaktionen von Vögeln, die als erstes Aufnahmen ihrer eigenen Unterart hörten, im darauf folgenden Durchgang stimuliert wurden, nicht jedoch umgekehrt. Unsere Ergebnisse zeigen, dass Blaukehlchen beide Gesangsarten eindeutig voneinander unterscheiden und die Fähigkeit haben, Unterarten zu erkennen. Weitere Untersuchungen sind notwendig, um die generelle Gültigkeit dieser Verhaltensmuster zu bewerten.

ABSTRACT This study was conducted to characterise macrogeographic variation in the vocalisation of the corncrake Crex crex, a bird species with a non-learned and highly stereotyped call. We also examined: (1) whether call characteristics remained stable across successive breeding seasons within two of the study populations and (2) whether call similarity was related to distance between populations. Recordings of 352 males from 8 populations were analysed. The analyses focused on variation in (1) temporal characteristics (duration of syllables and intervals, duration of the intervals between consecutive maximal amplitude peaks within syllables, called pulse-to-pulse duration (PPD)), and (2) spectral characteristics (minimal and maximal frequency, frequencies below which 25%, 50%, and 75% acoustic energy of signal is distributed). We found significant differences in most of the temporal and all of the spectral characteristics between populations. No differences were found in PPD. Significant interannual differences in spectral characteristics were found in both of the populations examined, whereas differences in temporal characteristics were only observed in one population. In general, geographic variation in calls showed clinal distance-dependence, where distant populations showed larger differences in call than neighbouring populations. Our results show that geographic variation in corncrake calls may be very dynamic in the short term and that within-population variation may occur on the same scale as between-population variation. This finding is surprising because call characteristics in non-learners are essentially inherited, and genetic transmission should be very slow. We suggest that the social interactions between males and/or the specific dispersal patterns of this species and the low site fidelity of adult and young birds may be responsible for such pattern.

The common versions (referred to as self-calibrated here) of the Standardized Precipitation Index (SPI) and the Palmer Drought Severity Index (PDSI) are calibrated and then applied to the same weather series. Therefore, the distribution of the index values is about the same for any weather series. We introduce here the relative SPI and PDSI, abbreviated as rSPI and rPDSI. These are calibrated using a reference weather series as a first step, which is then applied to the tested series. The reference series may result from either a different station to allow for the inter-station comparison or from a different period to allow for climate-change impact assessments. The PDSI and 1–24 month aggregations of the SPI are used here. In the first part, the relationships between the self-calibrated and relative indices are studied. The relative drought indices are then used to assess drought conditions for 45 Czech stations under present (1961–2000) and future (2060–2099) climates. In the present climate experiment, the drought indices are calibrated by using the reference station weather series. Of all drought indices, the PDSI exhibits the widest spectrum of drought conditions across Czechia, in part because it depends not only on precipitation (as does the SPI) but also on temperature. In our climate-change impact experiments, the future climate is represented by modifying the observed series according to scenarios based on five Global Climate Models (GCMs). Changes in the SPI-based drought risk closely follow the modeled changes in precipitation, which is predicted to decrease in summer and increase in both winter and spring. Changes in the PDSI indicate an increased drought risk at all stations under all climate-change scenarios, which relates to temperature increases predicted by all of the GCMs throughout the whole year. As drought depends on both precipitation and temperature, we conclude that the PDSI is more appropriate (when compared to the SPI) for use in assessing the potential impact of climate change on future droughts.

The main objective of this study was to determine the vulnerability of current agricultural cropping systems in the Marchfeld region to climate change. The investigation area Marchfeld is located in the north-eastern (NE) part of Austria and is characterized by a semi-arid climate with low annual rainfall. It is one of the driest regions in the country, but also one of the main field crop production areas. The soil conditions in Marchfeld demonstrate a significant spatial variability, which include soils with low to moderate water-storage capacities. Higher temperatures in the next decades imply higher evaporation and consequently higher water demand for the crops. The phenological development rates of the cultivars will accelerate and an increase of heat stress as well as drought stress can be expected. These points influence intense the water balance and subsequently the yield of the crops in the investigation area. In order to improve water use efficiency under those changing conditions, a shift of average sowing dates and an adjustment of tillage were analyzed. The DSSAT cropping system model was applied for winter wheat and spring barley to assess potential yield under climate scenarios for NE Austria. The scenarios were carried out with ECHAM5, HadCM3 and NCAR PCM global circulation models (GCMs) for present conditions (reference period 1961-1990) and 2035's (2021-2050), based on SRES-A1B emission scenarios. Yield model simulations were done for all defined scenarios (climate, management, crop) and different soil classes. The simulations contain the CO2 fertilizing effect, rain fed farming, adapted sowing date and contemporary crops without consideration of potential profit cuts caused by pest or diseases. Simulation results indicate that climate change will force a delay of the sowing date for winter wheat of maximal 14 days in October. In case of spring barley, climate change allows an earlier sowing date in spring (up to 14 days). Both crops show a decrease of potential yield, where spring barley demonstrates higher yield damage. Mainly NCAR PCM presents a slightly increase of winter wheat yield on medium soils. At the same time the interannual yield variability of both crops increased for all soils, leading to a higher economic risk for farmers. A replacement of ploughing by minimum tillage and direct drilling would lead to a lower decrease of mean yield for winter wheat (up to 5% in comparison with ploughed soil) and spring barley (up to 6% in comparison with ploughed soil) in 2035. This effect is mainly a result of an increase of plant available field capacity, a better water supply for the crops as well as a decrease of unproductive water losses.

One of the possible approaches (employed also in this contribution) to downscale the low-resolution output from Global Climate Model (GCM) into spatial and temporal scales required by impact models consists in using the stochastic weather generator (WG), whose parameters are modified according to the GCM-based climate change scenario. In the first part, we present a methodology for probabilistic climate change impact assessment based on daily and monthly outputs from multiple GCMs and two step stochastic weather generator M&Rfi. In the second part, we apply this methodology for assessing the impacts on selected climatic and agroclimatic characteristics in Europe and U.S.A. Methodology: to account for the changes in variability on various time scales, the daily WG (dWG) is coupled to the monthly WG (mWG). Parameters of dWG and mWG are derived from the observed series and then modified according to the GCM based climate change scenarios. To obtain more comprehensive scenarios, we merge information from GCM outputs at two temporal resolutions: monthly (benefiting from the length of the available output series) and daily (benefiting from the finer temporal resolution allowing to determine changes in climatic characteristics not derivable from the monthly series). The monthly GCM data (1961-2100) are used to derive changes in means, and standard deviations and intermonthly variability of monthly averages of temperature, precipitation and solar radiation. The daily GCM series (1961-1990 and 2081-2100) are used to derive changes in interdiurnal weather variability, daily temperature range and probability of precipitation occurrence. The climate change scenarios for a given future period, emission scenario and climate sensitivity are then defined by the pattern scaling method, in which the standardised changes (related to 1K rise in global mean temperature) derived from the GCM outputs are multiplied by the change in global mean temperature projected by simple climate model MAGICC. The experiment will (i) show impact of the climate change on selected climatic and agroclimatic characteristics in multiple European and US stations, and (ii) demonstrate effect of merging information extracted from monthly and daily GCM series (effect of changes in characteristics derived from the daily data will be assessed). The probabilistic impact assessment will be based on comparison of probability distribution functions of the climatic and agroclimatic characteristics derived from present vs. future climate synthetic weather series and assuming scenarios derived from multiple GCMs (IPCC-AR4 database). The climatic characteristics will include extreme precipitation and temperature characteristics and characteristics of wet/dry/hot/cold spells. The agroclimatic characteristics will include characteristics related to vegetation period (temperature sums over 5 and 10°C, sum of effective global solar radiation, number of effective growing days, number of days with low evapotranspiration) and the date of last frost day. Acknowledgements: The study is supported by the GAAV Grant Agency (project IAA300420806), Minstry of Education (KONTAKT project ME10128) and National Agency for Agric. Research (QI91C054).

Met&Roll is a WGEN-like parametric four-variate daily weather generator (WG), with an optional extension allowing the user to generate additional variables (i.e. wind and water vapor pressure). It is designed to produce synthetic weather series representing present and/or future climate conditions to be used as an input into various models (e.g. crop growth and rainfall runoff models). The present contribution will summarize recent experiments, in which we tested the performance of the interpolated WG, with the aim to examine whether the WG may be used to produce synthetic weather series even for sites having no meteorological observations. The experiments being discussed include: (1) the comparison of various interpolation methods where the performance of the candidate methods is compared in terms of the accuracy of the interpolation for selected WG parameters; (2) assessing the ability of the interpolated WG in the territories of Czechia and Nebraska to reproduce extreme temperature and precipitation characteristics; (3) indirect validation of the interpolated WG in terms of the modeled crop yields simulated by STICS crop growth model (in Czechia); and (4) indirect validation of interpolated WG in terms of soil climate regime characteristics simulated by the SoilClim model (Czechia and Nebraska). The experiments are based on observed daily weather series from two regions: Czechia (area = 78864 km2, 125 stations available) and Nebraska (area = 200520 km2, 28 stations available). Even though Nebraska exhibits a much lower density of stations, this is offset by the state's relatively flat topography, which is an advantage in using the interpolated WG. Acknowledgements: The present study is supported by the AMVIS-KONTAKT project (ME 844) and the GAAV Grant Agency (project IAA300420806).

The common versions (referred to as self-calibrated here) of the Standardized Precipitation Index (SPI) and the Palmer Drought Severity Index (PDSI) are calibrated and then applied to the same weather series. Therefore, the distribution of the index values is about the same for any weather series. We introduce here the relative SPI and PDSI, abbreviated as rSPI and rPDSI. These are calibrated using a reference weather series as a first step, which is then applied to the tested series. The reference series may result from either a different station to allow for the inter-station comparison or from a different period to allow for climate-change impact assessments. The PDSI and 1–24 month aggregations of the SPI are used here. In the first part, the relationships between the self-calibrated and relative indices are studied. The relative drought indices are then used to assess drought conditions for 45 Czech stations under present (1961–2000) and future (2060–2099) climates. In the present climate experiment, the drought indices are calibrated by using the reference station weather series. Of all drought indices, the PDSI exhibits the widest spectrum of drought conditions across Czechia, in part because it depends not only on precipitation (as does the SPI) but also on temperature. In our climate-change impact experiments, the future climate is represented by modifying the observed series according to scenarios based on five Global Climate Models (GCMs). Changes in the SPI-based drought risk closely follow the modeled changes in precipitation, which is predicted to decrease in summer and increase in both winter and spring. Changes in the PDSI indicate an increased drought risk at all stations under all climate-change scenarios, which relates to temperature increases predicted by all of the GCMs throughout the whole year. As drought depends on both precipitation and temperature, we conclude that the PDSI is more appropriate (when compared to the SPI) for use in assessing the potential impact of climate change on future droughts.

The main objective of this study was to determine the vulnerability of current agricultural cropping systems in the Marchfeld region to climate change. The investigation area Marchfeld is located in the north-eastern (NE) part of Austria and is characterized by a semi-arid climate with low annual rainfall. It is one of the driest regions in the country, but also one of the main field crop production areas. The soil conditions in Marchfeld demonstrate a significant spatial variability, which include soils with low to moderate water-storage capacities. Higher temperatures in the next decades imply higher evaporation and consequently higher water demand for the crops. The phenological development rates of the cultivars will accelerate and an increase of heat stress as well as drought stress can be expected. These points influence intense the water balance and subsequently the yield of the crops in the investigation area. In order to improve water use efficiency under those changing conditions, a shift of average sowing dates and an adjustment of tillage were analyzed. The DSSAT cropping system model was applied for winter wheat and spring barley to assess potential yield under climate scenarios for NE Austria. The scenarios were carried out with ECHAM5, HadCM3 and NCAR PCM global circulation models (GCMs) for present conditions (reference period 1961-1990) and 2035's (2021-2050), based on SRES-A1B emission scenarios. Yield model simulations were done for all defined scenarios (climate, management, crop) and different soil classes. The simulations contain the CO2 fertilizing effect, rain fed farming, adapted sowing date and contemporary crops without consideration of potential profit cuts caused by pest or diseases. Simulation results indicate that climate change will force a delay of the sowing date for winter wheat of maximal 14 days in October. In case of spring barley, climate change allows an earlier sowing date in spring (up to 14 days). Both crops show a decrease of potential yield, where spring barley demonstrates higher yield damage. Mainly NCAR PCM presents a slightly increase of winter wheat yield on medium soils. At the same time the interannual yield variability of both crops increased for all soils, leading to a higher economic risk for farmers. A replacement of ploughing by minimum tillage and direct drilling would lead to a lower decrease of mean yield for winter wheat (up to 5% in comparison with ploughed soil) and spring barley (up to 6% in comparison with ploughed soil) in 2035. This effect is mainly a result of an increase of plant available field capacity, a better water supply for the crops as well as a decrease of unproductive water losses.