Festivals versus birds

(1)

Festivals versus Birds

Do events affect bird diversity in Amsterdam city parks on the (middle) long-term?

BSc thesis by

Lisa de Groot (10790497)

Bachelor Future Planet Studies

Major Biology

(2)

Abstract

The pressure on urban ecosystems in Amsterdam city parks is increasing, every year a

growing amount of events take place at parks. Balance is needed between recreational

use of parks and retaining biodiversity in parks. The effects of events on biodiversity have

to be known in order to make better policy, especially due to the recent municipal

election earlier this year.

A data analyses is conducted, focussing on the bird diversity. Event data per park, visitor

data per park, bird diversity per park and the size (m2) of the parks is used in a regression

model.

The results showed no statistical significant effect of the number of events and the

number of visitors on bird diversity, almost all variance in bird diversity was explained by

the size of the parks. Although a significant result is shown, quality of the used data is low

due to a lack of standardization and monitoring protocol, therefore the results cannot be

used in policy making or to draw conclusions.

(5)

1. Introduction

Urban green is an important aspect of a healthy city, providing not only a green space for inhabitants to recreate and relax, but increasing their well being as well (Bolund & Hunhammar, 1999). Except for the benefits for inhabitants, urban green also has a great value to urban ecosystems and houses a great variety of flora and fauna (Elmqvist et al, 2015). Amsterdam alone counts 89 city parks, making urban green an important aspect of the structure of Amsterdam. The use of city parks in Amsterdam varies from a festival site to a picnic site. The increasing numbers of visitors and events in the city parks of Amsterdam of the past years are putting a rising pressure on the parks and its flora and fauna. A balance is needed between recreational use and retaining biodiversity in city parks. The actual effects of events in Amsterdam on the biodiversity in parks are not yet investigated, although a literature study is made on this subject. This study concentrates solely on the effects of events on birds. Amsterdam houses three types of events, identified by the municipality of Amsterdam as follows (Dirksen & de Fouw, 2016): Firstly, music or dance events: hosting a great amount of visitors on a small surface and using multiple podia and producing loud noise the entirely event. Secondly, other events: different types of events similar in appearances as dance/music events but without the production of loud noise. And the last types of events are firework shows, mostly taking place on New Years Eve or Koningsdag (Kingsday, a national holiday). This study only focuses on the events organized at city parks and their effects on the avian community. The known effects of events are noise pollution, light pollution and disturbance due to the presence of humans (Dirkens & de Fouw, 2016). When studying the effects of events on bird diversity these three effects and the factors influencing them have to be taken into account. To get an idea of the effects of events and urbanization on bird diversity, an introduction to noise pollution, light pollution and disturbance is made in the next paragraphs.

1.1 Background noise

With increasing global urban area, anthropogenic noise is increasing as well. Natural habitat is destroyed and replaced by new artificial habitat. Airports, busy roads, power plants and factories are all producers of anthropogenic noise, creating a new artificial acoustic background (Brumm & Slabbekoorn, 2005). Background noise is a natural phenomenon and includes the blowing wind through treetops, running water, raindrops falling on vegetation and noise produced by different animal species. Background noise is always present and affects the way animals communicate with each other. Natural background noise differs per habitat; most species settled in this particular habitat are well adapted to this specific noise. Because of the changing background noise by urbanization, these well-adapted species have difficulties to communicate due to the masking of

(6)

their vocals by background noise (Brumm & Slabbekoorn, 2005). Birds depend mainly on acoustic signals in order to attract mates or to defend their territory (Mockford & Marshall, 2009). Urban birds have to adapt their behaviour to urban noise in order to communicate efficiently. 1.1.1 Adaption to background noise In a changing acoustic environment some bird species will disappear and some will adapt to decrease the negative effects of sound pollution. The urban avian community is often similar in various urban areas and consists of persisting species and species arriving from other urban area, becoming a well adapt homogeneous bird community (Slabbekoorn, 2013). As results, the composition of the original avian community on a noisy urban site has changed. The persisting bird species all have something in common; they are adapted to urban noise. Urban background noise such as traffic noise consists mostly of low frequency. When the frequency of bird songs overlaps with this low frequency, their vocal signals are masked and communication becomes more difficult. Some of the persisting bird species can alter their vocal signals depending on the noise background; their frequency and loudness (intensity) is habitat dependent (Slabbekoorn 2013). This signal plasticity is shown in a study on great tits and their vocal plasticity in urban area. The minimum frequency of their vocal signals is significantly higher in urban area than in rural area, and positively correlated with background noise (Mockford & Marshall, 2009). In figure 1 (a) represents the frequency of vocal signals of great tits in rural area, (b) the frequency of urban background noise and (c) the frequency of signals of great tits in urban area. The minimum signal frequency (c) in urban area is slightly higher than the maximum frequency of urban noise (b), decreasing the masking effects of noise pollution. Unfortunately not every species is capable of adapting their frequency, making urban background noise reducing their habitat quality, resulting in noise avoidance and leaving the area (Francis et al, 2009). Consequentially, noise has a negative effect on nesting community richness and has a negative effect on nest predators as well. The noise avoidance of nest predators results in an increased nest success on noise adapted birds living in a noisy area (Francis et al, 2009). The nest density of the bird community has not affected by noise, but the nesting species richness is declining. The nesting community composition differs between noisy and rural area, due to the difference in richness between noisy and rural area and to the difference in indicator species for each site (Francis Figure 1: Frequency of vocal signals of the great tit in (a) rural area and (c) urban area. (b) is the frequency of urban background noise. (Mockford & Marshall, 2009)

(7)

et al.2009). These indicator species are often capable of changing their vocal frequency to prevent signal masking due to background noise and have a preference for noisy habitats such as city parks. An increase of nest success at noisy sites could be responsible for this preference. Nest predation is the number one cause of nest failure. The relationship between reduced predation and increased noise is shown in a study by Francis et al (2009), showing the reduction of the western scrub-jay at noisy sites. Scrub-jay residency levels were 32% higher at the control sites in comparison with the treatment sites, probably due to masking of their vocal frequency by the noise of the treatment site.

1.2 Light pollution

As discussed, noise pollution has a local impact on birds and causes decreased nesting species richness due to the masking of their vocal signals. Light pollution is an important factor negatively influencing birds as well. Most species on earth depend on day/night cycles. Nowadays (artificial) light is present during day and night time, especially in urban area. Especially birds migrating at night are affected, depending on visual indicators to navigate (Longcore & Rich 2004). Artificial light at night can cause disorientation and attraction to the artificial light sources. Migrating birds are using astronomical/visual cues and magnetic orientation to navigate, both interrupted by artificial light (Longcore & Rich 2004). Disorientation is causing migrating birds to collide with tall illuminated buildings, causing death or at least exhaustion (Van Doren et al 2017). While artificial lights attract migrating birds, some birds avoid the light sources (De Molenaar et al 2000). Some diurnal birds have increased orientation at night due to additional artificial light (Longcore & Rich 2004). Some of these diurnal birds are predators causing increased predation risk for birds migrating at night. It is clear that artificial light at night significantly influences the avian community. 1.2.1 Short-term effects of artificial lights An example of short-term effects of artificial light at night is the influence of New York light tribute in honour of the victims of 9/11 on birds (Van Doren et al 2017). This high intensity light beam installation is similar to the light installations used at sport events and festivals. During illumination of the tribute lights, clusters of birds were seen around and in the light beams. The bird density around the lights increased with a maximum of 150 times the baseline, defined as the mean bird density within 2 to 20 km of the tribute lights. The vocal activity of the birds increased within the bird clusters, their orientation decreased similar to the disorientation that occurs due to tall illuminated structures such as high buildings. Minutes after the tribute lights were turned off, the birds decreased their vocal intensity, increased their flight speed and moved away from the tribute site. The mortality rate of the birds affected by the tribute lights has not become clear, but previous

(8)

research states that birds can get exhausted because of disorientation, leading to death (Van Doren et al 2017). 1.2.2 Middle long-term effects of artificial lights An example of middle long-term effects of artificial light on birds is the influence of light sources on choice of nesting sites, breeding period, breeding success and predation of godwits in the province of North Holland, the Netherlands. Artificial light at highways has a significant negative effect on the choice of nesting place of godwits, especially in the first 250 meters a decrease of nests has been observed (De Molenaar et al 2000). However, this research focuses only on the godwit population in 1998 and 1999. Some explanations given are rejection by light due to predation prevention because of increased visibility of prey, in this case the godwit. The distance to the artificial light sources and the egg volume, whereby a greater egg volume represents an increased condition of the parental godwits and breeding success, indicates the relation between breeding success and artificial light. Closer to the light sources an increased egg volume was observed; however relation was too weak to draw a conclusion.

1.3 Tolerance to disturbance in city parks

City parks are the green islands inside a crowded urban area, constantly disturbed by visitors. At events such as festivals at parks this disturbance is even greater. The avian community reacts on disturbances by fleeing the area, although individuals increase their tolerance when exposed often to disturbance (Samia et al 2015). Small birds are often more tolerance than larger birds and migrating species are less tolerant than native species because of the lack of human contact when migrating (Samia et al 2015). The distance when the bird flees between the visitor and bird is called the flush distance (Fernández-Juricic 2001). Fernandez-Juricic et al (2001) investigated the bird tolerance to human disturbance in Madrid city parks in relationship with the habitat types within the parks. They chose common urban bird species that differ in body size: The Woodpigeon, Blackbird, Magpie and House Sparrow, all four represented in western Europe urban avian communities such as in Amsterdam city parks. They found that habitat structure is correlated with flush distance. When shrub cover is increased, the flush distance decreased. Shrub cover was mostly responsible for low flush distance, especially when shrub height increased. The influence of coniferous cover on flush distance depends on the species, mainly woodpigeons and magpies used coniferous canopy as escape cover. When the tree height increased, the flush distance increased as well. The canopy cover is further away from the disturbed birds making them fled earlier when approach. Furthermore, when parks were visited less, the flush

(9)

distance increased for each species, indicating an increasing tolerance for human disturbance when birds are exposed to increasing visitor numbers. Concluding, the tolerance for human disturbance depends on habitat structure in parks and the number of visitors daily visiting the park.

1.4 Previous research and aim

The short-term effects on bird diversity before, during and directly after events is already shown in several studies (Krijgsveld et al 2012, Dirksen & de Fouw, 2016). A study has been conducted in an Amsterdam city park, during the festival Amsterdam Open Air (Krijgsveld et al 2012). The study concluded that no serious disturbances on breeding birds have been observed before, during and after the festival. While the study of disturbance due to Amsterdam Open Air only focuses on short-term effects of just one event, this study will focus on the middle-long term effects of all events at the 89 city parks and recreational green Amsterdam contains in a period of 2 years (2016 and 2017). The aim of this study is to compare different parks with a difference in recreational pressure to investigate whether a variance is observed in bird diversity. Event data and park data is used to analyse the bird diversity data. No significant difference in bird diversity among parks is expected, although some species sensitive for disturbance could be absent in parks with the most recreational pressure.

(10)

2. Method

In order to investigate the diversity differences between parks, multiple factors have to be taken into account. Amsterdam counts 89 city parks, each different in size, amount of visitors and events, biodiversity and management. In some parks almost no events take place and in other parks every week an events is organized. A multiple regression analysis is conducted to find the drivers that have an effect on bird diversity. Although the effects of events on bird diversity is the main subject, other variables such as the area of the parks have to be taken into account in order to find all drivers that could have influence on the bird diversity. In the next paragraph the used data is discussed.

2.1 Used data

City parks and recreational green Information is received from the data portal of the municipality of Amsterdam: Amsterdam.data.nl. An ESRI shapefile (a geospatial vector data format used in GIS (geographic information system) software) containing the names, locations and the area in square meter of every park is used. Bird diversity data Bird diversity data is received from the NDFF (national database of flora and fauna) database. The NDFF database consists of more than hundred databases combined and validated. The data consists of single observations from ‘Telmee’, ‘Waarnemingen.nl’, NDFF subscribers and inventories as specified by a protocol or project. Every observation is validated automatically or if necessary manually. Only bird richness is used since the abundance is depending on the number of observers: for example, less common species such as a long-eared owl could have been observed several times by different observers while only one owl was actual present in that area. Between 2013 and 2017 23,667 individual observations of birds are made in Amsterdam city parks. Event data Event data is received from the municipality of Amsterdam. This study focuses on the number of events, their location, the number of visitors per event and the year. The number of visitors and the number of total events per park contains information regarding the disturbance and recreational pressure that birds have to endure. Light pollution data due to events is not used, since all events are during daytime. The guidelines of the municipality of Amsterdam state all events must stop at 11 PM (Dirksen & de Fouw, 2016). Furthermore, the guideline states that if high intensity lights are present, they cannot be pointed at canopy (Dirksen & de Fouw, 2016). These measures ensure minimal light pollution. While noise pollution is an important factor influencing the behaviour of birds, the

(11)

municipality do not keep track on the sound intensity (dB) used at events, consequentially no noise observations were used although it could affects the avian community as shown in the literature study.

2.2 Preparation of suitable data

The received data was not standardized. Event data sets containing all events organized in Amsterdam were used, containing no clear location profiles of the parks but only the addresses adjacent to the parks. The addresses of the event data set are transformed to x-y coordinates using a geocoder, after which the coordinates were loaded into arcGIS: a Geographical Information System used to generate, edit, analyze and visualize geographical data. A buffer of 100 m was made around Amsterdam city parks using the Buffer command in arcGis, containing all x-y coordinates of addresses adjacent to parks. The addresses belonging to the coordinates were renamed according to the closest parks, if they were within the 100 m buffer, and a new location profile is created for each event. Now each individual event contains the park name as location profile instead of an address. The same was done with bird diversity data, adding a location profile to every observed bird species. Since all events and bird data contains a clear location profile after the performed GIS alteration, the total number of events and bird species could be calculated per park. This was done for each year: The total number of events and their visitors per park in 2016, 2017 and 2018 and the total number of bird species observed per park in 2013 till 2017. Total visitors and events over 2016-2018 and over 2016-2017 were calculated as well using R.

2.3 Used models

First partial correlations were calculated between bird diversity and number of events, number of visitors and the size in m2 per park. Secondly a multiple linear regression was calculated between bird diversity and event, visitor and park size variables. The linear regression accounts for all predictor variables. The output coefficients summarize the p-values per predictor and are used to evaluate the hypotheses to identify variables that may not have a relationship with bird diversity. H0: ßi = 0 when other independent variables are included Figure 2: Input: park polygons and output: park polygons with buffer, covering a larger area.

(12)

HA: ßi ≠ 0 when other independent variables are included Thus, the null hypothesis is rejected if the ßi parameter estimate is not equal to zero, suggesting strong evidence that the independent variable has a relationship with the dependent variable.

3. Results

3.1 Partial correlations

First the linear correlation of the variables bird diversity and number of events, bird diversity and the total park area and bird diversity and number of visitors was calculated. The Pearson correlation coefficient is computed and to interpreting these correlation coefficients (r) the guidelines composed by Evans (1996) are used, suggesting the interpretation of r as follows: 0.0 to 0.19 indicates a very weak linear correlation, 0.2 to 0.39 a weak correlation, 0.4 to 0.59 a moderate correlation, 0.6 to 0.79 a strong correlation and 0.8 to 1.0 a very strong correlation. In order to reject the null hypothesis (no correlation is found) the significance of the results, the p value of the correlation has to be under 0.05. The correlation coefficient (r) of bird diversity and park surface area (m2) is 0.768, indicating a strong correlation between the two variables. The used bird diversity data consists of the mean diversity per park in a span of 5 years and the park surface area is log-normal distributed. The p value of the correlation is p = 2.2e-16, rejecting the null hypothesis. Thus, significance strong positive correlation is measured between bird diversity and park surface area; when the surface area of a park increases, the bird diversity is increasing as well. The correlation coefficient of bird diversity and number of visitors (see figure 3) is 0.255. The bird diversity data and visitor of 2016 and 2017 is used, because only event data from 2016 till 2018 is available. The p value is 0.0151: a significance linear correlation has been found, although the correlation coefficient is between a 0.2-0.39 interval: resulting in a weak positive correlation. Not only the correlation between bird diversity and visitors is Figure 3: Bird diversity plotted against number of visitors per park and a regression line. Correlation coefficient is 0.255 and is statistical significant.

(13)

calculated, the correlation between bird diversity and number of events per park is calculated as well (see figure 4). The impact of the number of events and their visitors on bird diversity could differ. The correlation coefficient of the number of events per park and bird diversity per park is 0.260, with a p-value of 0.0133. The correlation is significant, but because of the low correlation coefficient the correlation is weak.

3.2 Multiple regression

A linear model is used to calculate the multiple linear regression. As response (dependent) variable (y) mean bird diversity per park over 2016 and 2017 is used and as predictor (independent) variables (x) mean events per park, mean visitors per park and the total area surface (m2) per park is used. The regression analysis is used to determine the strength of the effect of the independent variables on the bird diversity; thus the strength of the relationship between bird diversity and park area, events and visitors. The analysis will answer the question “what is a better predictor of the bird diversity in Amsterdam city parks”. Because the model consists of multiple independent variables, the regression line cannot be visualized in a normal 2D plot. The output of the model can be found in table 1. The adjusted R-squared for the model is 0.5857, meaning 58,6% of the original variability is explained by the model, the other 41,4% is explained by residual variability. The p-values of the predictors are indicators of the relationship of the predictors with the response; a low p-value

Table 1: Number of bird species per park

1

Independent variables

ß coefficient

SE

t-value

p-value

(Intercept)

- 5.146e+01

5.611 -9.171

2.21e-14 ***

Mean Event 2016-2017

+ 1.555e-03

2.746e-01

0.006

0.995 Mean visitors 2016-2017

+ 6.267e-05

1.721e-04

0.364

0.717 Park size (m2)

2

+ 5.452

5.116e-01

10.656 < 2e-16***

R

squared

(adjusted) = 0.5857, p-value: < 2.2e-16

Note: 1 _{Dependent variable,}2_{Park size (m2) is log transformed} p < 0.001 ‘***”, p <0.01 ‘**’, p<0.05 ‘*’

Figure 4: Bird diversity plotted against number of events per park and a regression line. Correlation coefficient is 0.260 and is statistical significant.

(14)

indicates the rejection of the null hypothesis meaning the independent variable has no correlation with the dependent variable. If the p-value is under 0.05, the predictor variations are related to variations in bird diversity. The p-values of both events and visitors are high, respectively 0.995 and 0.717, and therefore those predictors have no significance effect on bird diversity: the null hypothesis has not been rejected. The p-value of the total surface area (m2) per park is < 2e-16, the null hypothesis is rejected thus correlation has been found between the surface area of parks and their bird diversity. Since the total surface area is the only independent variable that is statistically significant a smaller model is preferred. The number of events and visitors can be excluded of the final model, resulting in a simple regression model of bird diversity and total surface area. A regression model with park size log transformed is used, because the relationship between bird diversity and park size is exponential. The output of the model can be found in table 2. The adjusted R-squared of the polynomial regression model is 0.5927, 59,27% of the variability of the original model is explained. The p-value of the model and of the ß coefficient of the independent variable are both under the significant level of 0.001, and thereby rejecting the null hypothesis.

Table 2: Number of bird species per park

1

Independent variables

ß coefficient

SE

t-value

p-value

(Intercept)

- 52.3208

5.4260

-9.643

1.95e-15***

Park size (m2)

2

+ 5.5512

0.4859

11.424 < 2e-16***

R

squared

(adjusted) = 0.5927, p-value: < 2.2e-16

Note: 1_{Dependent variable,}2 _{Park size (m2) is log transformed} p < 0.001 ‘***”, p <0.01 ‘**’, p<0.05 ‘*’

4. Discussion

Previous research (Krijgsveld et al 2012, Dirksen & de Fouw, 2016) made the effects of events on the avian population clear. Sound pollution, light pollution and human disturbance due to urbanization are the main causes of increasing pressure on birds. Events have the same effects on birds but are incidental and the avian community is already adjusted to urban environment and additional effects of urbanization. Therefore one might expect little to no response on bird diversity due to events, especially due to the event guidelines composed by the municipality of Amsterdam, composed to minimize harmful effects on flora and fauna. The analysed data revealed, as expected, no effects of events on bird diversity. The number of visitors and events per park had no significance effect on the bird diversity with a p-value far greater than 0.05, while surface area showed much more importance

(15)

in the linear regression with a p-value of < 2e-16. Concluding almost all variation in bird diversity between parks is explained by the size of the parks. Recently, a study conducted on effects of Amsterdam Open Air (Krijgsveld et al, 2012) and a literature study conducted on the effects of events on flora and fauna in city parks shows no effects on the avian community as well (Dirksen & de Fouw 2016). There was no significant decline in birds before during and after the festival Amsterdam Open air, hold at Gaasperpark. The literature study showed no effects as well, because commercial events are discrete (short-term) and hold at venues with already a large amount of human disturbance. In addition, city parks are not classified as vulnerable nature. This study is an addition to this knowledge. Our results support the two studies: no effects are shown. A comparative study, such as this study, between different parks with their own bird diversity and number of events, including parks with no events, has never been done before in Amsterdam. Instead of observing a single park or a single event, when analysing multiple data points, the most important drivers and vulnerable areas become clear. Nevertheless, to scientifically approve the results, the used data has to meet some criteria.

4.1 Data quality

Although some clear significant results are shown, the credibility of the results depends on the used data. The data is obtained from the municipality of Amsterdam and from the NDFF database. Firstly, the used data consist only of the years 2016 and 2017. Bird diversity data goes back up to five years, but event data does not. The municipality of Amsterdam started monitoring permit requests for all events hold at Amsterdam since 2015, but the data of 2015 does not include visitor numbers. In order to conduct a good comparison study of different parks, more data of multiple years is needed. 4.1.1 The importance of a monitoring protocol The quality of the bird diversity data is questionable. The main reason is that the results cannot be scientifically approved because no protocol was followed during bird monitoring. Volunteers can share data at any point of the day or year, with no assurance that a change in measurements is caused by natural change or different monitoring methods between individual volunteers. For example, in 33 parks of the total 89, less than one bird species was observed according to the NDFF data. To prevent biased bird data, a monitoring protocol has to be composed and implemented as part of the bird diversity research by the municipality of Amsterdam and has to concentrate on an inventory of bird richness and abundance. Monitoring protocols ensure quality and a known confidence level of data, are necessary to detect increase or decrease of bird species and bird populations not affected by changes in volunteers and are necessary when compared and shared with other institutes.

(16)

Oackley, Thomas and Fancy (2003), suggest a monitoring protocol containing three sections: Narrative, standard operating procedures and supplementary materials. The narrative section provides an explanation of the reason a resource is selected, background information concerning the resource, how results helps policymaking or management decisions and an overview of the different components of the protocol. The standard operating procedure presents details of how the protocol components will be executed and contains instructions. Supplementary materials include example datasets, supporting data and reports and data management. A suggestion of a monitoring protocol in Amsterdam city parks, in general similar composed as the proposal of Welsh (1995) for the forest bird-monitoring program in Ontario, Canada. 1) First sample locations have to be selected: Parks in Amsterdam of different sizes, with and without events. For example blablabla, depending on the amount of available volunteers for monitoring. 2) Volunteers, randomly distributed across the park, record all birds seen and heard in a time frame of 10-minute period. All volunteers are familiar with bird identification. 3) Recording is repeated three times throughout the day: once at 9 AM, once at 2 PM and once at 9 PM. Bird recordings take place two times a year at the end of May or the beginning of June and at the end of June, every year around the same days. 4) Shortly after monitoring, the bird data is entered in a data frame and verified after which it is imported into an excel sheet for further data analyses. 4.1.2 The importance of standardized data The event data received from the municipality of Amsterdam differs per year. Before 2018 no clear location profiles were composed, instead of the park name the street adjacent to the park was submitted as location, often with spelling mistakes. The event data set of 2015 did not contain visitor numbers per event, making the data unusable for our analysis. District offices saved all event data individually before 2015; consequently no overview data set of all districts together exists before 2015. Furthermore, noise data was missing; only an estimate was made of loud music present on the event site without actual noise (dB) data measured. To ensure event and park datasets are suitable for further research, a template is suggested. The suggested template and metadata can be found in Appendix A.

(17)

5. Conclusion

No significance correlation or regression was found between events, visitors and bird diversity because p-values for both events and visitor were much greater than the significance level of 0.05. Park size, on the other hand, is significantly correlated to bird diversity; almost all variety is explained by park size in the regression model of bird diversity and event, visitors and park size. Although statistically significant results are measured, it is not statistically responsible to use these results in policymaking or comparison studies. The diversity data quality is to low due to the absence of a monitoring protocol and event data is not complete. Although the municipality monitors event data, the dataset organisation is not suitable for analyses and standardization of the data is suggested. In order to answer the research question first better data has to be collected by composing a monitoring protocol and standardize the event data.

(18)

6. References

Bolund, P., & Hunhammar, S. (1999). Ecosystem services in urban areas. Ecological economics, 29(2), 293-301. Brumm, H., & Slabbekoorn, H. (2005). Acoustic communication in noise. Advances in the Study of Behavior, 35, 151-209. De Molenaar, J. G., Jonkers, D. A., & Sanders, M. E. (2000). Wegverlichting en natuur: III: lokale invloed van wegverlichting op een gruttopopulatie (No. 64). Alterra. Dirksen, S., de Fouw, J. (2016). Effecten van grootschalige commerciele evenementen en muziekfestivals in parken en groengebieden op natuurwaarden – review van de literatuur. Elmqvist, T., Setälä, H., Handel, S. N., Van Der Ploeg, S., Aronson, J., Blignaut, J. N., ... & De Groot, R. (2015). Benefits of restoring ecosystem services in urban areas. Current Opinion in Environmental Sustainability, 14, 101-108. Fernández-Juricic, E., Jimenez, M. D., & Lucas, E. (2001). Bird tolerance to human disturbance in urban parks of Madrid (Spain): Management implications. In Avian ecology and conservation in an urbanizing world (pp. 259-273). Springer, Boston, MA. Francis, C. D., Ortega, C. P., & Cruz, A. (2009). Noise pollution changes avian communities and species interactions. Current biology, 19(16), 1415-1419. Krijgsveld, K. L., Jonkvorst, R. J., & van der Vliet, F. (2012). Effecten van dancefestival Amsterdam Open Air op broedvogels. Longcore, T., & Rich, C. (2004). Ecological light pollution. Frontiers in Ecology and the Environment, 2(4), 191-198. Mockford, E. J., & Marshall, R. C. (2009). Effects of urban noise on song and response behaviour in great tits. Proceedings of the Royal Society of London B: Biological Sciences, rspb20090586. Samia, D. S., Nakagawa, S., Nomura, F., Rangel, T. F., & Blumstein, D. T. (2015). Increased tolerance to humans among disturbed wildlife. Nature communications, 6, 8877. Slabbekoorn, H. (2013). Songs of the city: noise-dependent spectral plasticity in the acoustic phenotype of urban birds. Animal Behaviour, 85(5), 1089-1099. Slabbekoorn, H., & den Boer-Visser, A. (2006). Cities change the songs of birds. Current biology, 16(23), 2326-2331.

(19)

Van Doren, B. M., Horton, K. G., Dokter, A. M., Klinck, H., Elbin, S. B., & Farnsworth, A. (2017). High-intensity urban light installation dramatically alters nocturnal bird migration. Proceedings of the National Academy of Sciences, 114(42), 11175-11180. Van Kempen, H., Smeets, H. (2013). Het Grote Groenonderzoek 2013. Welsh, D. A. (1995). An overview of the Ontario forest bird monitoring program in Canada. In: Ralph, C. John; Sauer, John R.; Droege, Sam, technical editors. 1995. Monitoring bird populations by point counts. Gen. Tech. Rep. PSW-GTR-149. Albany, CA: US Department of Agriculture, Forest Service, Pacific Southwest Research Station: p. 93-98, 149.

7. Appendices

Appendix A

Template standardized table

Template for a standardized table Object ID Number of object Park Name of one out 89 Amsterdam city parks Size Surface area of park in m2 Bird species per year Total number of bird species per park, in a separate column for each year. Bird observations per year Total number of birds observed during monitoring protocol per park, in a separate column for each year. Number of events per year Total number of events per year per park, in a separate column for each year. Number of visitors of events per year Total number of visitors of events per year per park, in a separate column for each year. Number of total visitors per year Total number of visitors of parks per year, in a separate column for each year.

(20)

Appendix B

Data Management Plan 1. Types of data 1. What types of data will you be creating or capturing? All data is existing data retrieved from the municipality of Amsterdam. 2. How will you capture, create, and/or process the data? All data is analyzed with R studio, after it was standardized using arcGIS for spatial data. 2. Contextual details (Metadata) needed to make data meaningful to others 1. What will be the naming convention of your files? All files are stored in folders: Analysis_script folder contains all used analysis scripts for R Derived_data folder contains all result data: graphs, output and used standardized tables. Used_data folder contains all used data retrieved from the municipality of Amsterdam. All naming of the content is self explanatory, in the name the function of the file is explained. 2. What file formats will you be using? In this repository the following formats are used: .txt, = text file .mxd = map file, can be opened with arcMaps .dbf = attribute format, containing attributes in columns for each shape. .prj = projection format, containing the coordinate system and projection. .shp = shape format, the feature geometry of bird observations. .shx = shape index format, containing a positional index for the feature geometry. .jpeg and .png = images of graphs .xls and .xlsx = excel files to store tables .csv= csv files to store tables used in R studio 3. Quality control 1. What will you do to ensure that the data will not be erroneous? I had to work with existing data, although some of the data is erroneous. Yet I used all data, and discussed the errors in the discussion of my thesis. 2. What are the roles in quality control? I asked the opinion of my supervisor. 4. Storage, backup and security 1. What will be the URL where your data will be available?

The data is available at https://figshare.com/s/d1ce30217afdee465f42, and is private because the data can not be shared without permission of the municipality of Amsterdam.

(21)

2. What is your backup plan for the data? The data is stored on an USB as well. 5. Copyright and reuse 1. Who will own the copyright or intellectual property rights to the data? The municipality of Amsterdam and the NDFF database. 2. How and among who will the data be shared during the project and when the project is finished? The data can only be shared among the municipality of Amsterdam, my supervisor and two master students who did this project as well. The data is property of the municipality of Amsterdam, with their permission the data can be shared with others.

Festivals versus birds