Project Topics
Get Complete Project Material File(s) Now! »
Land management tools
The most common conservation tool in response to land-use changes is the creation of protected areas to preserve species and habitats. More than 15% of the world’s terrestrial areas are covered by protected areas (Juffe-Bignoli et al., 2014). The main aim of protected areas is to buffer biodiversity from diverse and often intense anthropogenic pressures (Margules & Pressey, 2000). Yet light pollution is rarely considered as a criterion in the designation and management of protected areas. They tend to be darker than non-protected areas even in intensely lit regions such as Europe (Gaston, Duffy, et al., 2015; Guetté et al., 2018) and although they suffered an increase in light pollution level in recent years, it is limited compared to surrounding areas (Guetté et al., 2018). Nevertheless, the proportion of protected areas’ surface considered dark is decreasing (Fig. 14 ;Gaston, Duffy, et al., 2015). In addition, when considering a 500 km buffer around protected areas, the peak in mean ALAN occurs in the first 25 km (Guetté et al., 2018). Protected areas designation is constraint by the spread of human settlements and as the distance between protected areas and urban areas shrinks (McDonald, Erickson, & McDonald, 2000), they may be increasingly impacted by diffuse pollutions.
Limit the duration of lighting
Part-night lighting schemes are increasingly being used and likely to become widespread in regions with developed lighting infrastructures due to energy price and concerns about carbon dioxide emissions (Gaston et al., 2012). Few studies investigated the potential effect of such measures but they do not prove conclusive in reducing the impact of ALAN on bat species (Azam et al., 2015; Day et al., 2015). Indeed, street lighting is mostly important for humans activities after dusk and before dawn which coincide with the most critical hours of activity of crepuscular and nocturnal species (Gaston et al., 2012). Intelligent lighting schemes, using motion-sensors to detect users, may have more ecological benefits as they could increase the length of non-lit periods (Rowse et al., 2016) however they haven’t been investigated yet.
Reduce the quantity of light
At the local scale, the quantity of light can be measured through the energy input of a lamp (power, intensity) or the light emitted (luminance, illuminance). At large scales, the quantity of light is often evaluated through remote sensing as a measure of radiance. The illuminance, measured in lux, is the most commonly employed metric as it can be easily measured in the field with low-cost equipment whereas luminance measures require expensive equipment and power and intensity can’t be measured in the field, these information are held by private companies or municipalities in charge of the outdoor public lighting. Illuminance is a photometric measure which means that it is relative to the human perception of light. Although this can be criticized as each species perceives light in a different way, this metric has the advantage of being used by lighting engineers and thus allows for direct knowledge transfer for possible lighting planning adaptations. Dose-dependent responses to illuminance have been shown in bird activity onset and human and bird melatonin secretion suppression (West et al., 2011; de Jong, Jeninga, et al., 2016). However, some species seem to be impacted by light irrespective of the illuminance level (Stone et al., 2012; Azam et al., 2018). Nonetheless, diming schemes have the potential to improve light pollution levels at the landscape level. A study using remote sensing data of both light emissions and vegetation (as a proxy of suitable habitats) showed that a 20% reduction in light emissions mostly concentrated in rural areas (as opposed to city centers and already dark areas) in two natural reserves and their surroundings (5 km buffer) could increase the surface of dark suitable habitats by up to 46% (Fig. 16 ;Marcantonio et al., 2015).
Prevent areas from being artificially lit
The simplest approach to reduce light pollution is to restrict nighttime lighting to the minimum necessary for human use and remove installation in already light saturated areas and from areas where it is not indispensable. A good example of the harm induced by the installation of unneeded lighting is the decrease in bat colonies presence in churches due to the implementation of aesthetic lighting (Rydell, Eklöf, & Sánchez-Navarro, 2017). Unfortunately, light illuminance classes. ‘*’ indicates that light illuminance classes were significantly different from control unlit treatment (P < 0.01) (extracted from Azam et al., 2018). the global trend is not toward street light removal but toward the installation of more public lighting and the increase of private lighting due to the low cost of LEDs.
Knowledge gaps and thesis plan
There is a large body of literature demonstrating the numerous and dramatic impacts of artificial light at night on organisms’ physiology and behavior. They suggest that a wide variety of taxa are influenced by the presence of lighting with different responses depending on the spectrum, the timing and the quantity of light. Studies’ findings indicate potential major repercussions on individual’s fitness and reproductive success with possible cascading effects on population dynamics, community composition and ecosystems functioning. As urban areas keep on extending, understanding, assessing and enhancing urban biodiversity is of major importance from both a conservation and a social perspective (Kowarik, 2011). Therefore, it appears important to better understand the influence of anthropogenic pressures on species inhabiting cities such as bats which are, in addition, protected at the European level (Council Directive 92/43/EEC, 1992). Therefore, during my PhD thesis, I investigated the impact of light pollution on bats (Box 1) at the local scale and at the city scale. In the first chapter, I present the biological and light data I used for my research. At the two scale studied, both ground-based and remote sensing data can be used to evaluate the level of light pollution. These sources of information present different advantages and drawbacks which are important to consider for their use in analyses and to interpret results. All the studies presented in this PhD are based on acoustic recordings of bat ultrasounds that can be semi-automatically identified to the species or species-group level with specific software. However the outputs of such software are not straightforward to exploit as they may include errors and false identifications. In the first study presented, I participated to the development of a methodology aiming at properly selecting bat identifications to then carry out analysis on bats response to their environment. This study intended to answer the following question.
Error risk modelling from the semi-automated identification
Successes and fails in the automatic identification noted by manual checks were modelled in relation to the confidence indexes provided by the software, allowing to predict the needed confidence index to ensure a given maximum error rate tolerance (Fig. 2). Confidence indexes corresponding to the maximum error rate rates (i.e. 0.5, 0.4, 0.3, 0.2 and 0.1) did not vary much for some species such as B. barbastellus (0.12-0.20), Eptesicus serotinus (0.18-0.29) and Rhinolophus hipposideros (0.39-0.45), and more for others, e.g. Nyctalus leislerii (0.29- 0.59), P. kuhlii (0.16-0.44) and Plecotus ssp. (0.18-0.36) (Table 2). In addition, the needed confidence indexes to limit error risks were overall low for B. barbastellus (0.12-20), E. serotinus (0.18-0.29), P. kuhlii (0.16-0.44), Plecotus ssp. (0.18-0.36), Myotis spp. (0.21-0.42), and higher for P. nathusii (0.67-0.77) and N. noctula (0.51-0.61) (Table 2). A low maximum error rate tolerance allows to highly reduce the false positive rate (e.g. from 0.20 in raw data to 0.01 in data with a 0.1 maximum error rate for Myotis spp.), and even to fully remove false positives (e.g. for E. serotinus and N. leislerii starting from the 0.1 and 0.2 rates of maximum error rate, respectively) (Table 2; Fig. 2). However, reduce the maximum error rate tolerance can also generate moderate to high false negative rates, such as for E. serotinus (0.15), Myotis spp. (0.28), N. leislerii (0.79), N. noctula (0.51), P. kuhlii (0.38) or Plecotus spp. (0.27) (Table 2; Fig. 2).
Table of contents :
GENERAL INTRODUCTION
1. Global land-use change
2. Light pollution
3. Light pollution’s impacts on biodiversity
3.1. Diversity of perception of a fundamental cue
3.2. Effects of light on daily biological events
3.3. Effects of light on seasonal biological events
3.4. Effects of light on movements and spatial distribution
3.5. Effects of light on interactions and community composition
4. The conservation challenge posed by light pollution
4.1. Why should light pollution be a focus of environmental research in the 21st century?
4.2. Current and future lighting
5. Management and technical levers of action
5.1. Policies gaps
5.2. Land management tools
5.3. Technical tools
6. Knowledge gaps and thesis plan
CHAPTER 1: BIOLOGICAL AND LIGHT DATA
1. Light pollution data
2. Bat activity data
2.1. Bat acoustic data
2.2. Data sampling
2.3. Bat calls identification
Article 1. Barré K., Pauwels J., Le Viol I., Claireau F., Julien J.-F., Julliard R., Kerbiriou C.,
Bas Y. Robustness of using a semi-automated method to account for identification
errors in bat acoustic surveys
Appendices
CHAPTER 2: EVALUATING THE INFLUENCE OF LIGHT POLLUTION ON BAT ACTIVITY AT THE CITY-SCALE
Introduction
Aims of the chapter
Principal results & discussion
Perspectives
Article 2. Pauwels J., Le Viol I., Azam C., Valet N., Julien J.-F., Bas Y., Lemarchand C.,
Sànchez De Miguel A., Kerbiriou C. Accounting for artificial light impact on bat
activity for a biodiversity-friendly urban planning
Appendices
CHAPTER 3: IMPACT OF ARTIFICIAL LIGHT ON LANDSCAPE CONNECTIVITY: CURRENT STATE & SCENARIOS OF OUTDOOR LIGHTING
Introduction
Aims of the chapter
Principal results & discussion
Perspectives
Article 3. Laforge A., Pauwels J., Faure B., Bas Y., Kerbiriou C., Fonderflick J., Besnard A.
Light reduction improves connectivity for bats in an urban landscape
Appendices
Article 4. Pauwels J., Laforge A., Bas Y., Fonderflick J., Besnard A., Valet N., Le Viol I.,
Kerbiriou C. Flying through the city: new lighting technologies alter landscape
connectivity for bats in urban areas
Appendices
CHAPTER 4: LOCAL SCALE EFFECTS OF LIGHT CHARACTERISTICS
Introduction
Aims of the chapter
Principal results & discussion
Perspectives
Article 5. Pauwels J., Kerbiriou C., Bas Y., Valet N., Le Viol I. Adapting street lighting to limit
light pollution impacts on bats in protected areas
Appendices
GENERAL DISCUSSION
1. Principal results
2. Technical challenges of the study of bats
3. Importance of scale in the evaluation of light pollution’s effects
4. The future of lighting
REFERENCES
