Introduction
The Great Bustard Otis tarda is a globally threatened bird species classified as ‘Vulnerable’ by IUCN (2010). Its West-Pannonian population declined significantly during the second half of the last century but increased again from 130 birds in 1995 to 376 individuals in the winter 2008/2009 due to the implementation of various conservation measures (Raab et al. Reference Raab, Kollar, Winkler, Faragó, Spakovszky, Chavko, Maderič, Škorpíková, Patak, Wurm, Julius, Raab and Schütz2010). However, some remaining threats still cause substantial mortality of Great Bustards. Particularly, power line collisions occurred quite frequently during the last decade and were responsible for 41% of all West-Pannonian bustards found dead (Raab unpubl. data).
Man-made structures such as tall masts for television and mobile phone, wind turbines or power lines can cause harm to birds in many ways: habitat loss due to disturbance, disruption of local (or migratory) movements or − even more severe − injury and death through collisions (Ballasus and Sossinka Reference Ballasus and Sossinka1996, Barrios and Rodríguez Reference Barrios and Rodríguez2004, Bevanger and Brøseth Reference Bevanger and Brøseth2004, Haas et al. Reference Haas, Nipkow, Fiedler, Handschuh, Schneider-Jacoby and Schneider2006, Newton Reference Newton2007, Drewitt and Langston Reference Drewitt and Langston2008, Rollan et al. Reference Rollan, Real, Bosch, Tintó and Hernández-Matías2010). Mortality due to collisions with power lines involves a broad range of bird species (Janss Reference Janss2000, Bevanger and Brøseth Reference Bevanger and Brøseth2004, Frost Reference Frost2008), but evidence exists that large species considered as “poor flyers” are particularly sensitive to collisions with man-made structures. Small and broad wings combined with high wing load prevent rapid and appropriate reactions to unexpected obstacles (Bevanger Reference Bevanger1995, Reference Bevanger1998, Janss Reference Janss2000). Therefore bustards are among the commonly recorded victims of collisions (Janss Reference Janss2000, Reiter Reference Reiter2000, Lane et al. Reference Lane, Alonso and Martín2001, Alonso et al. Reference Alonso, Martín, Palacín, Martín and Magña2005, Martín et al. Reference Martín, Alonso, Alonso, Palacín, Magña and Martín2007).
To mitigate the high collision risk for these species, enhancing the contrast of wires against the background by using power line markers or avian flight diverters (Alonso et al. Reference Alonso, Alonso and Muñoz-Pulido1994, Brown and Drewien Reference Brown and Drewien1995, De La Zerda and Rosselli Reference De La Zerda and Rosselli2003, Frost Reference Frost2008, Yee Reference Yee2008) and underground cabling (Drewitt and Langston Reference Drewitt and Langston2008) have been recommended as conservation measures.
Power line collisions represent a major mortality factor for Great Bustards Otis tarda, although they are able to adapt their flight behaviour (Raab et al. Reference Raab, Spakovszky, Julius, Schütz and Schulze2011). In the framework of the LIFE project LIFE05 NAT/A/000077 for protecting the West-Pannonian population of the Great Bustard (www.grosstrappe.at) two conservation measures were implemented to decrease the risk of collision with power lines: (1) extensive underground cabling (starting in 2005) and (2) marking of power lines (starting in 2006). In this article, we present a comparative evaluation of both mitigating measures.
Methods
Study area
The study area of 686.5 km2 covered most of the current distribution area of the West-Pannonian Great Bustard population in Eastern Austria and Western Hungary (Fig. 1). A more detailed description of the area is provided in Raab et al. (Reference Raab, Kollar, Winkler, Faragó, Spakovszky, Chavko, Maderič, Škorpíková, Patak, Wurm, Julius, Raab and Schütz2010, Reference Raab, Spakovszky, Julius, Schütz and Schulze2011).
Figure 1. Austrian and Hungarian parts of the West-Pannonian distribution area (marked black) of the Great Bustard considered in this study.
Underground cabling and marking of power lines
To decrease collision risk of Great Bustards in Eastern Austria, high-voltage power lines were equipped with different markers within the LIFE project LIFE05 NAT/A/000077 (www.grosstrappe.at). To enhance visibility of 380 kV power lines for birds, the earth wire was marked with double black and white aviation marker balls (30 cm diameter; one marker per 25–30 m), the conductors were marked with 30 x 30 cm marker plates (one per 40–50 m) fixed between the duplex conductors, alternating between black and white in colour (Fig. 2a). Double black and white aviation marker balls were also used for 220 kV power lines (one marker per 30–35 m earth wire and conductor; Fig. 2b); 110 kV power lines were marked with 40 x 10 cm marker plates (alternating in colour between black and white; one marker per 15 m earth wire and per 30–90 m conductor; Fig. 2c).
Figure 2. Design of wire markers at 380 kV (a), 220 kV (b) and 110 kV (c) high voltage power lines in Eastern Austria and design of rectangular (d) and circular (e) bird flight diverters used at high- and medium voltage power lines in Western Hungary, attached to the earth wire and conductor, respectively.
In Western Hungary, the earth wires as well as the conductors of high- and medium voltage power lines were marked within the Hungarian LIFE project LIFE04 NAT/HU/000109 (www.tuzok.hu) with bird flight diverters, which can reflect light under low light conditions. Two different types of bird flight diverters were used: a rectangular one (9 x 15 cm, one marker per 17–100 m), which spins even in light wind (Fig. 2d) and a circular one (13 cm diameter, one marker per 17–100 m), which fluoresces up to 10 hours after sunset (Fig. 2e).
For all areas marked in Figure 1, high-voltage and medium-voltage power lines were digitised using the software package (ArcMap 9.1, ESRI) and total power line length was measured annually. Due to the small sample size, we could not differentiate between different marker types in our analyses, but only classified power lines as marked and unmarked. At the beginning of this study in 2002 a total of 10.6 km of high-voltage power lines were equipped with markers. Between 2005 and 2011 a further 89.7 km (86.7 km high-voltage, 3.0 km medium-voltage power line) were marked. Additionally, a total of 43.1 km of underground cabling of medium-voltage power line was installed by 2011, decreasing the total power line length in our study area from 356.5 km (249.2 km medium-voltage and 107.3 km high-voltage power line) in 2002 to 313.4 km in 2011.
Data collection
Although mortality of West-Pannonian Great Bustards has been documented in great detail since 1996, in the present analysis we only considered data collected systematically since 2002, when intensive activities to protect the West-Pannonian population of the Great Bustard began within the framework of Rural Development Programmes in Lower Austria (since 2002: LE) and Burgenland (2002-2006: INTERREG; 2007-2010: LPF), and the LIFE projects LIFE04 NAT/HU/000109 (2004-2008; www.tuzok.hu), LIFE05 NAT/A/000077 (2005-2010), and LIFE09 NAT/AT/000225 (since 2010; www.grosstrappe.at). Due to increasing awareness among hunters and farmers, bustards found dead in the field were reported with a much higher likelihood than before 2002.
Data analyses
Annual mortality rate of the Great Bustard population in our study area was defined as the number of individuals that were found dead in the respective year, divided by the total number of bustards present in the study area between April and May (Raab et al. Reference Raab, Kollar, Winkler, Faragó, Spakovszky, Chavko, Maderič, Škorpíková, Patak, Wurm, Julius, Raab and Schütz2010). We did not include young bustards found dead, which were still unable to fly. To calculate the annual mortality due to power line collisions, we considered only power line collision casualties. One year was not defined as calendar year but as the period from 1 June until 31 May of the following year.
Linear regression analyses were calculated to test if (1) total annual mortality rate, (2) annual mortality caused by power line collisions and (3) annual mortality caused by other reasons, changed from 2002 to 2011. Spearman rank correlations were used to test for effects of underground cabling (quantified as underground power line length) and length of marked power lines in the respective years on the annual mortality rate (considering power line collision victims only). Additionally, we used a paired Wilcoxon test to evaluate differences in the mean number of bustards colliding yearly with marked and unmarked power lines. Finally, we calculated GLMs (with normal error distribution and log-link function) testing for effects on annual mortality rate due to power line collisions, including the variables annual length of underground cables and marked power lines, and both variables separately. The three models were ranked according to their corrected Akaike information criterion (AICc) values.
Results
Between 1 June 2002 and 31 May 2011 a total of 78 dead individuals were reported, of which 41.0% (32 birds) referred to Great Bustards which collided with power lines. Mean annual mortality rate (± SD) between 2002 and 2010 was 3.5 (± 1.6)%, mean annual collision rate was 1.6 ± (1.3)%. The significant decline in the total mortality rate observed from 2002 to 2011 (Fig. 3a) is mainly due to a strongly decreasing proportion of bustards colliding with power lines (Fig. 3b). The proportion of birds that died due to other reasons remained rather similar over the study period (Fig. 3c).
Figure 3. Change of (a) total mortality rate, (b) mortality caused by power line collisions and (c) other casualties of West-Pannonian bustards between 2002 and 2011.
Annual mortality rate due to power line collisions declined significantly with increasing length of underground power lines (Spearman rank correlation: r s = -0.84, P = 0.005) and with increasing length of marked power lines (r s = -0.76, P = 0.017). The number of bustards colliding with power lines was significantly lower at marked than at unmarked power line sections (paired Wilcoxon test: Z = 2.07, P = 0.038; Fig. 4).
Figure 4. Difference of mean annual number of bustard collisions (± SD) per km marked and unmarked power lines (n = 9 years).
Underground cabling had a prominent effect on Bustard mortality caused by power line collisions as indicated by the model selection. The GLM including the variable underground cabling only achieved the lowest AICc value (26.21) and a much higher AICc weight (0.743) than models containing both variables (AICc = 29.43, AICc weight = 0.148) or only the variable power line marking (AICc = 30.06, AICcweight = 0.108).
Discussion
Although Great Bustards tend to avoid areas near power lines when selecting suitable habitats (Lane et al. Reference Lane, Alonso and Martín2001) and are able to adapt their flight behaviour (Raab et al. Reference Raab, Spakovszky, Julius, Schütz and Schulze2011), collisions with overhead power lines still represented an important mortality factor in our study area during recent decades (Faragó Reference Faragó1981, Reiter Reference Reiter2000, own unpubl. data). The mean annual collision rate of 1.6% documented by our study for the time period 2002–2011 is similar to the collision rate of 2.2% recorded for the national population of Portugal (Infante et al. Reference Infante, Neves, Ministro and Brandão2005).
Our results clearly indicate that Great Bustard mortality rate declined significantly in the two years following the first underground cabling and power line marking measures. Although power line marking appeared to reduce the collision risk, underground cabling explained most of the reduced mortality after implementation of these two conservation measures. The effect of the latter was so prominent that in our modelling approach, the GLM containing only the variable underground cabling performed best of all models. Although the general importance of power line marking as a conservation measure to reduce collision risk should not be underestimated (e.g. Alonso et al. Reference Alonso, Alonso and Muñoz-Pulido1994), its efficacy varies with species and can be particularly low for birds such as bustards (Jenkins et al. Reference Jenkins, Smallie and Diamond2010). Furthermore, the effectiveness of marking also depends on the design of markers (Jenkins et al. Reference Jenkins, Smallie and Diamond2010, Janss and Ferrer Reference Janss and Ferrer1998). Unfortunately, our dataset is too small to allow for more detailed analyses quantifying effects of marker design on collision risk of Great Bustards in our study area.
The conservation measures implemented to reduce collision casualties in our study area most likely contributed to the continuous increase of the West-Pannonian Great Bustard population (Raab et al. Reference Raab, Kollar, Winkler, Faragó, Spakovszky, Chavko, Maderič, Škorpíková, Patak, Wurm, Julius, Raab and Schütz2010). Indeed, the relatively small number of bustards found to have collided with power lines, might give a rather incomplete account. The mortality caused by power line collision is supposedly much higher than suggested by the casualties discovered in the field. An unknown proportion of carcasses may not have been found in higher vegetation or may have been carried away by carrion-feeding mammals. However, it is unlikely that the completeness of annually reported casualties changed during the period of our study due to a more or less unchanged number of recorders (predominantly farmers and hunters). Therefore, the declining mortality rate of power line victims should reflect a real decline in the risk of power line collisions. Besides a reduced collision risk, underground cabling may additionally have contributed to the recovery of the West-Pannonian Great Bustard population by providing access to potentially suitable habitats that were not used before, because Great Bustards appear to avoid the vicinity of power lines (Lane et al. Reference Lane, Alonso and Martín2001).
The two conservation measures, power line marking and underground cabling, were implemented in the most important West-Pannonian Great Bustard habitats and across highly frequented flight routes. Our results demonstrate that these measures successfully decreased the mortality rate of Great Bustards within a short time period. The extensive underground cabling representing 43.1 km of medium-voltage power lines and the marking of 89.7 km high-voltage power lines were realised at costs of €3.02 million and €1.01 million respectively. The average costs of underground cabling of medium-voltage power lines and marking of high-voltage power lines were €70,000 (min-max: €50,000-100,000) and €11,700 (€9,200–16,700) per km power line. Unfortunately, the underground cabling of high-voltage power lines appears to be not practicable due to tremendously high costs (more than €0.7–1.0 million per km) and technical difficulties. Therefore, we recommend underground cabling of medium-voltage power lines as one priority conservation measure in Great Bustard areas, while high-voltage power lines should be equipped with markers. Not only bustards but also other larger bird species may significantly benefit from such conservation measures.
Acknowledgements
This paper is a result of data collected in the framework of the LIFE Projects ‘‘Cross-border Protection of the Great Bustard in Austria” (LIFE05 NAT/A/000077, LIFE09 NAT/AT/000225; www.grosstrappe.at) and ‘‘Conservation of Otis tarda in Hungary” (LIFE04 NAT/HU/000109, www.tuzok.hu). These projects were supported by the EU, many project partners and co-financiers. Additional EU co-financed projects (INTERREG-III: HUBP5M2_0015; LPF: 5-N-A1025/148-2009 and LE: RU5-LE-040/000, RU5-LE-152/000 and RU5-S-428/001-2005) were supported by the Austrian Federal Ministry for Agriculture, Forestry, Environment and Water management (BMLFUW) and the regional governments of Lower Austria and Burgenland. We are grateful to more than 700 people involved in conservation projects for the West-Pannonian Great Bustard population and who reported the majority of Great Bustard casualties found in the field.
