A growing rat population and its impact
on the bird biodiversity.
Ilja van Vuuren, 12310123
Figure 1 (Flevo-park, a Hidden Gem in Amsterdam-Oost., 2020). The entrance to Flevo-park.
Table of Contents
Abstract ... 3 The research... 4 Method ... 5 Results ... 7 Discussion... 9 Improvement ... 9 Conclusion ... 11 References ... 12 Acknowledgements ... 14 Appendices ... 15Abstract
Rats have been increasing in the city of Amsterdam. Since the rats may eat the eggs of the birds (Le Corre, 2008), this could mean the rats have a negative impact on the bird biodiversity. This study has investigated the correlation between rat presence and bird biodiversity in Flevo-park. Flevo-park is in Amsterdam and the park has a wide variety of bird species, making it an ideal place for this study.
The research has also investigated the difference between bird biodiversity in a forested area and a field area. This was done by counting the birds and calculating the Shannon diversity index for 23 transects over a period of one month. In the end this study has shown that there was a significant difference between field and forested areas, whereas forested areas had a higher bird biodiversity index in general. The bird biodiversity compared between areas with rats and without rats showed no significant difference. This means that no negative correlation could be found between the presence of rats in an area and the bird biodiversity in the respective area. No action needs to be taken to preserve the Flevo-park bird biodiversity from the growing rat population. The people of Amsterdam will be able to enjoy the wide variety of birds in the Flevo-park.
The research
The city of Amsterdam is trying to control the rats living there (van Adrichem et al., 2013), since they spread diseases and are a pest to the inhabitants. A problem is that rodenticide resistance has been noted in rat populations through Europe (van Adrichem et al., 2013). The rat poison which was used is no longer sold by the municipality of Amsterdam (Huismuizen, 2020). They want to minimize the usage since rats are becoming more resistant to the poison and the poison gets into the environment which could kill other animals.(Bestrijden Met Rattengif). This will most likely lead to an increase of rat presence in Amsterdam. This increase is already in grey literature which say that there has been an increase of rats present in Amsterdam of 31 percent in one year (“Aantal Muizen Bij Amsterdammers Spectaculair Gestegen.,” 2016; Mice Are a Problem in Amsterdam: Here’s How to
Handle Them., 2018). This is an estimate based on rats’ sightings by people living in Amsterdam,
whereas there is no actual data on the number of rats. Especially with the decrease in pesticide usage (Huismuizen, 2020), the lack of data on the number of rats is a problem. This lack of data means that the rat population could be larger than expected. That is why the effects of rats need to be studied right away before the population of rats is so large it becomes problematic.
There is a lot of research on rats spreading diseases and parasites (Himsworth et al., 2013; Lee et al., 1982; Strand et al., 2015), including zoonotic bacteria (Leptospira interrogans, Yersina
pestis, Rickettsia typhi, Bartonella spp., Streptobacillus moniliformis), viruses (Seoul hantavirus), and
parasites (Angiostrongylus cantonensis). The problem is that there is still little known about their impact on the bird biodiversity. It is proven that rats have a negative impact on bird species on islands (Jones et al., 2008; Le Corre, 2008). They will eat the eggs of the birds making it more difficult for bird chicks to survive thus decreasing the bird population. Even though there exists research on the negative impact of rats on birds on island, there is still little research on the impact of city rats on urban birds. Urbanization already makes it difficult for a lot of bird species to survive in cities (Martin & Bonier, 2018), since dominant species have competitive interference with subordinate species. This leads to some species thriving in cities whereas others will struggle. This is a huge problem since the local diversity in cities is going down (Martin & Bonier, 2018). The biodiversity is important for humans since it can give physical as well as psychological benefits to people as well as shaping the character of a city(Sadler et al., 2010).
If the rat population is growing, they will most likely eat more bird eggs. The bird biodiversity which is already suffering under the urbanization might struggle even more. Therefore, it is important to research the effect of city rats on birds. To see whether the birds will avoid rats, since these rats eat the eggs of birds (Le Corre, 2008), which would mean an increase in rats could result in a decrease in bird biodiversity. Therefore, this research is trying to find out the extent of the rat impact before it’s too late for the birds in Amsterdam. This research can be used to preserve the biodiversity of birds in Amsterdam and other cities with similar characteristics.
The area of study is the Flevo-park which is a park located in Amsterdam east. It is the ideal park for this study because of its favourable location. Since it is located in the edge of Amsterdam near water you will find plants and animals not found in the centre of Amsterdam(van der Sijde, 2015). There will also be birds migrating here as well as birds from Ijsselmeer that hibernate here(van der Sijde, 2015). For this research the focus will be on birds that stay in Flevo-park during spring since the research time is limited to the month of April.
Another aspect is the Jewish cemetery located in Flevo-park, which acts as barrier of nature where humans cannot frequently go. This makes it ideal for animals since they will not be disturbed there. It even has a natural value of five which is the highest possible score (the rest of the park has a score of three)(van der Sijde, 2015). Overall, this makes the park ideal for the research into the impact of rats on the biodiversity of birds.
The aim of this research is to examine the influence of rats on bird biodiversity in the Flevo-park in Amsterdam. The main research question is “can a correlation be found between rat presence and bird biodiversity?”. The hypothesis is that there is a negative correlation between rat presence and bird biodiversity since rats have a negative impact on bird species on islands (Jones et al., 2008). It is already proven that birds react to visual and auditory cues of predators like rats (Stanbury & Briskie, 2015). On top of that the rats eat bird eggs (Møller, 1983), thus it is expected to have a
negative correlation.
The sub question is, “Is there a significant difference between forested areas and field areas?”. The biodiversity of birds in the Flevo-park is expected to be higher in forested areas than the field areas. The hypothesis is that the bird biodiversity in forested area is higher since the birds are nesting in forested areas and are thus more present there.
Method
Cameras were placed as well as food, chewing cards and audio recorders. These were placed to study and record the rats in Flevo-park. There were seven Spypoint Force-Dark trail cameras which were placed in a Spypoint security boxes SB-200. They were locked with Masterlock python cables, and they were placed somewhere between 10 and 30 cm’s above the ground. The food that was placed as a lure at 3 sites was either peanut butter, hazelnuts, and raisins. It is important to keep in mind that this attracts rats and thus might have increased the rat sightings. The number of rats was determined by counting the number of rats sighted by the cameras as well as the count of rats chewing on the chewing cards. The areas are throughout the Flevo-park such that the data is representative of the bird and rat population in the Flevo-park. They were areas with different characteristics such that the different aspects of Flevo-park need to be considered.
The different types of areas in Flevo-park that were considered were a field area and a forested area. The birds reside more in forested areas since they can sit in the trees as well as make their nests, since the research was done in April when they are nesting. The fields have less trees and more people walking there. This means that the birds are less present in these areas since they can’t sit in the trees and they get scared away by people as well as dogs brought to the park.
Secondly the areas in which the bird biodiversity was recorded needed to be determined. To make sure that the sample data was representative of the whole population it was needed that the sample is random (Hill et al., 2005), for this the two-stage sampling was used. As mentioned before there are different types of areas, two-stage sampling makes sure that these areas can be considered separately, such that it is still possible to compare them. Using a random number
generator to select the areas has ensured that the sample was random and thus representative of the population.
There are transects which separate the study area into smaller areas. These smaller areas were all given a type depending on what was present. So, if there are mostly trees, they will fall into the forest categories. The transect was divided into their respective category as can be seen in Figure 3. As seen in Figure 4 these areas are once again to be divided, now into 16 sub-areas.
To choose an area a random number generator was used, for each type there are two areas per day. After these areas are chosen, the sub areas need to be determined. There will be four sub areas chosen from each area, once again this was done with a random number generator to make sure the sample is randomized.
Finally, in advance the bird species in the park were
researched. This made sure that when the research started the birds would be more easily recognizable. This was done in one week; the preparation made sure that the birds are counted quicker as well as making sure to distinguish between different bird species. For birds
Figure 2 (Hill et al., 2005). Two stage sampling.
Figure 3: Map of the research area showing which parts are forest and which are fields (dark green is forest and light green is field).
that were still unknown the app ‘INaturalist’ was used. Photographs of the bird can be uploaded along with a description and then the community of iNaturalist can determine the species. This was once again checked by the researcher to make sure it was the correct species.
To be able to determine the bird biodiversity the Shannon index was used. The Shannon index formula, as can be seen in Figure 5 (The Shannon Index of
Species Diversity Is Calculated Using This Equation), calculates the
Shannon index value which can be used as an indicator of bird biodiversity. It calculates this from the amount of bird species, as is measured in the preparatory counting of bird species, and the number of individuals per species that is counted. To calculate the Shannon index, one takes the number of individuals of one species and divides it by the total number of individuals, giving the proportion. This proportion can then be raised to the power of two, followed by taking the natural logarithm. The result of this must then be multiplied with the proportion calculated before. This is then repeated for each species, adding the final product of each iteration together to give the
Shannon index. The Shannon index was measured and calculated for each sub area which is then added and divided so that the average for that area is calculated.
After all the preparation is done the main research had started. Every week there were three days where the bird biodiversity was measured. For each day there were four hours of research time, which means that for each area one hour was assigned, and that for each sub-area there are 15 minutes to count the birds. The research was continued for one month, which means that in the end there are 46 measures of biodiversity per type of area. So, each area was surveyed twice, to make this sufficient data to make an analysis of the impact of rats on bird biodiversity.
For the data analysis the bird biodiversity was compared to the number of rats present in the area. This was done with a univariate linear regression analysis, which can determine a relationship between an independent and a dependant variable. For this research the independent variable is the number of rats and the dependant variable is the bird biodiversity. The expected outcome was that there was a negative relationship, meaning that when more rats are present there will be less bird biodiversity.
Figure 5 (Chegg study). Shannon index which calculates bird biodiversity with the amount of species and their respective amount.
Results
The bird species with the highest count were the magpies. The magpies had the highest count from all species with an observation of 15 magpies in one subarea. This was in area 9, subarea 11 which is a field area. The second highest count which is not a magpie are the crows, doves, and herons with a count of 7 which is significantly lower.
Despite the high number of magpies in field areas on average the field areas scored a lower Shannon diversity index than forested areas since there was more diversity in the forested areas. This is reinforced by the data as seen in Figure
6, which shows the Shannon biodiversity index for each transect. This was expected since the Shannon biodiversity index looks mostly at the diversity. The field areas have in general a lower Shannon index than the forested areas. Thus, it must be taken into account that the forested areas have a higher biodiversity index since there is a wider spread of species.
For now only a visual analysis of the data is done, to test whether there is a significant difference a Kruskal Wallis analysis
was conducted. The sample size was too small since there were only 11 field areas and 12 forested areas, thus the Kruskal Wallis test was used instead of a paired t-test. In the end the Kruskal Wallis test gave a p-value of 0.00004915679 which means there is a significant difference between a forested area and a field area. So, there is a
statistically significant difference between forested areas and field areas, meaning it needs to be taken into account when looking at the connection between bird biodiversity and the rat presence.
Figure 6. The Shannon biodiversity index for each transect.
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 1
Shannon value in field and forest area
Shannon value
Figure 7. The average Shannon value for a field area vs a forest area. A field area is a 0 whereas a forested area is a 1.
The data on the rats was very limited with only 4 areas in which they were observed so binomial data was used. This means that for each area that had recorded rats in it was labelled with a 1 and the rest with a 0. So, the data from the cameras and the chewing cards was taken, if a rat was present that area was given a 1. The Shannon index for areas with no rats versus the areas with rats was then compared as can be seen in Figure 8. From figure 8 it is already clear that the Shannon diversity index is higher for areas with rats’ present. To mathematically analyse this the Kruskal Wallis was once again used since the sample size was too small. There were 19 areas without rats and only four areas with rats thus making the sample size too small for an ANOVA test. The test gave a p-value of 0.2181818 which means there is no
statistical significance and the null hypothesis is not rejected. In figure 9 the amount of
bird that were observed are shown, whereas birds are counted double in different areas and thus it is not an indication of the amount of birds present in the Flevo-park. It does show the proportion of birds compared to each other, whereas the Pica Pica also known as a magpie is observed the most and thus is the most present in this area with a total
count of 721. The second highest observed is the carrion crow with a count of 250.The third highest count is the Heron with a count of 176.
An overview of the most import statistics has been added to the appendices. These show the descriptive statistics including the ones mentioned before like count of magpies and p-values of the analysis.
Figure 9. The species of bird and the amount of times they were observed. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0 1 Sh an n o n v alu e
No rats present(0), rats present(1)
Shannon value for no rats and rats
Figure 8. The Shannon value for no rat’s vs rats with the standard deviation.
Discussion
The results on the difference between forested areas and field areas show that in general forested areas have higher biodiversity than the field areas in the Flevo-park. This was expected since birds can nest in the trees and the study was conducted during breeding season. It also gives them some cover from predators or other disturbances like humans walking by. An interesting mention is that the fields have a very high number of magpies and crows. So, these birds are very resistant to disturbances from the humans. This could mean that if a park is getting overcrowded or a lot of the trees are removed to make a field, a lot of the smaller birds as well as several larger birds could suffer from this. The magpies and crows on the other hand can still thrive in an environment like that.
The rat presence seemed to have no correlation to the bird biodiversity. This could be because all the rats were observed in the forested area and together with forested areas having a higher bird biodiversity this leads to a skewed result. The areas where the rats are in are compared to other forested areas as well as the field area whereas the field areas lower the score for no rat areas. It can also mean that the rats have little to no effect on the birds. This could be since there are too little rats in the Flevo-park, there were only four out of 23 areas that contained rats. It could also mean that the rats have no impact on the birds since they feed on other things than bird eggs.
In the end it is probably a combination of the things mentioned above, there are still too few rats combined with their little impact this shows no real correlation. To further investigate this, it would be recommended to take a longer research time.
Improvement
The research was done by a single person with limited resources and time. Therefore, it was not performed optimally, so aspects that could have been improved are discussed here.
Since the fieldwork time was limited to one month it was best to study each transect one hour per day and then repeating it a second time. For a follow-up study it would be recommended to be in the transect for a longer time and at the same time of the day for each transect. At the start of the day there would be a lot of birds but then at the end of the day there would be almost no birds. So, to make sure the time of day has no influence the transects could be studied at the same time each day.
There were multiple studies done parallel to each other, whereas one study placed food throughout the park to see how much would be eaten. It needs to be considered that this could have attracted rats and thus skewed the data. For a follow up study, it would be advised to make sure there are as little outside factors that influence the rat behaviour.
The data from this research had all the rat sightings in the forested area which could be a consequence of the cameras and the chewing cards being mostly in the forest. To make sure that it is also possible to spot rats in the field areas cameras as well as chewing cards should be placed in the field. This will make the rat presence data more representative of the real situation.
Another important aspect is that the researcher for this study was a beginner bird spotter. He had almost no knowledge on birds when this research started, nevertheless a lot of bigger birds where easy to identify. The biggest difficulty was with smaller birds since they look more like each other as well as being quite fast. It is recommended to have a professional bird spotter to identify the birds as well as someone to write it down. This way the bird spotter can keep focusing on the
birds as to make sure he doesn’t miss one and the helper can write them down. In such a way the birds will be more accurately counted.
Conclusion
The results showed that there was a significant difference between bird biodiversity in field areas and in the forested areas. When new parks are being build this needs to be considered, whereas every park needs a lot of forested areas to maintain a high bird biodiversity. Which in turn is good for the mental and physical health of the people visiting. This could also be considered outside of parks, where a lot of trees will be a positive influence on bird biodiversity.
No correlation could be found between rat presence and bird biodiversity. This means that there is no urgent action that needs to be taken. At least for now the rat population is no danger to the birds, but it is advised to keep track of the rat population. If it grows to a larger size their impact might grow and thus become a danger to the bird biodiversity. For now, the discontinuing of rat poison has no negative impact for the birds and can be continued without short term consequences.
References
Aantal muizen bij Amsterdammers spectaculair gestegen. (2016). AT5.
https://www.at5.nl/artikelen/152501/aantal_muizen_bij_amsterdammers_spectaculair_gestegen
Bestrijden met rattengif. (n.d.). Meldpunt Ongedierte.
https://www.meldpuntongedierte.nl/ratten/rattenbestrijding/bestrijden-met-rattengif#:~:text=Waarom is rattengif verboden%3F,zoals vogels%2C katten en honden.
Flevopark, a hidden gem in Amsterdam-Oost. (2020). A Dutch Experience.
https://adutchexperience.com/flevopark-a-hidden-gem-in-amsterdam-oost/
Hill, D., Fasham, M., Tucker, G., Shewry, M., & Shaw, P. (2005). Handbook of Biodiversity Methods. Survey, Evaluation and Monitoring. Austral Ecology, 32(8), 956–956.
https://doi.org/10.1111/j.1442-9993.2007.01799.x
Himsworth, C. G., Parsons, K. L., Jardine, C., & Patrick, D. M. (2013). Rats, cities, people, and pathogens: A systematic review and narrative synthesis of literature regarding the ecology of rat-associated zoonoses in urban centers. Vector-Borne and Zoonotic Diseases, 13(6), 349–359. https://doi.org/10.1089/vbz.2012.1195
Huismuizen. (2020). GGD. https://www.ggd.amsterdam.nl/dierplagen/muizen/
Jones, H. P., Tershy, B. R., Zavaleta, E. S., Croll, D. A., Keitt, B. S., Finkelstein, M. E., & Howald, G. R. (2008). Severity of the effects of invasive rats on seabirds: A global review. Conservation
Biology, 22(1), 16–26. https://doi.org/10.1111/j.1523-1739.2007.00859.x
Le Corre, M. (2008). Conservation biology: Cats, rats and seabirds. Nature, 451(7175), 134–135. https://doi.org/10.1038/451134a
Lee, H. W., Luck Ju Baek, & Karl M . Johnson. (1982). Isolation of Hantaan Virus , the Etiologic Agent of Korean Hemorrhagic Fever , from Wild Urban Rats. The Journal of Infectious Diseases,
146(5), 638–644. https://www.jstor.org/stable/30112439
Martin, P. R., & Bonier, F. (2018). Species interactions limit the occurrence of urban-adapted birds in cities. Proceedings of the National Academy of Sciences of the United States of America,
115(49), E11495–E11504. https://doi.org/10.1073/pnas.1809317115
Mice Are a Problem in Amsterdam: Here’s How to Handle Them. (2018). Expatrepublic.
https://www.expatrepublic.com/mice-problem-amsterdam-heres-handle/#:~:text=Nobody knows exactly why mice,of warm winters as factors.&text=Last year%2C reports of rats,31 percent%2C according to AT5.
Møller, A. P. (1983). Damage by rats Rattus norvegicus to breeding birds on Danish islands.
Biological Conservation, 25(1), 5–18. https://doi.org/10.1016/0006-3207(83)90027-7
Sadler, J., Bates, A., Hale, J., & James, P. (2010). Bringing cities alive: the importance of green spaces for people and biodiversity. Urban Ecology.
https://books.google.nl/books?hl=nl&lr=&id=Vp6_VAZaXwAC&oi=fnd&pg=PA230&dq=bringing+ cities+alive&ots=afq_BaKGgG&sig=WHeCdimW8ktlQGrB13AeUAgcyfg&redir_esc=y#v=onepag e&q=bringing cities alive&f=false
Stanbury, M., & Briskie, J. V. (2015). I smell a rat: Can New Zealand birds recognize the odor of an invasive mammalian predator? Current Zoology, 61(1), 34–41.
https://doi.org/10.1093/czoolo/61.1.34
Strand, T. M., Löhmus, M., Persson Vinnersten, T., Rasbäck, T., Sundström, K., Bergström, T., & Lundkvist, A. (2015). Highly pathogenic leptospira found in urban brown rats (Rattus norvegicus) in the largest Cities of Sweden. Vector-Borne and Zoonotic Diseases, 15(12), 779–781.
https://doi.org/10.1089/vbz.2015.1800
The Shannon index of species diversity is calculated using this equation. (n.d.). Chegg Study.
https://www.chegg.com/homework-help/questions-and-answers/shannon-index-species-diversity-calculated-using-equation-laboratory-community-contains-fo-q42262930
van Adrichem, M. H., Buijs, J. A., Goedhart, P. W., & Verboom, J. (2013). Factors influencing the density of the brown rat (Rattus norvegicus) in and around houses in Amsterdam. Lutra, 56(2), 77–91. https://zoogdierwinkel.nl/sites/default/files/imce/nieuwesite/Winkel/pdf
download/Lutra_56-2_Website.pdf#page=6
van der Sijde, G. (2015). Stadspark & natuurgebied. Natura, 4, 6–8. file:///C:/Users/iljav/Downloads/Natura2015112004001.pdf
Acknowledgements
Throughout the making of this dissertation I received a great deal of support and constructive criticism. I would like to thank them all.
First of all I would like to thank Caitlin Emily Black for her help in the forming of the fieldwork as well as giving me constructive criticism on the writing. This has helped me take my thesis to a new and higher level. I would also like to thank Renske Hoondert for her help on the theoretical part of the thesis. Renske Hoondert has helped with guiding me on the right path with the data analysis as well as giving constructive criticism on my writing as well.
On top of that I would like to thank Lune Walder, Sascha Rem, Amee van Boheemen, Antonia van der Grinten and Mathijs Blom. They were the fellow students that helped me by providing me with the data on rats so I could investigate the correlation. They also made for a great group to ask for advice or to discuss certain problems.
Appendices
Figure 10. Shannon biodiversity index form.
Species identified Number of individuals Total number of individuals
Stage1 Shannon Rat presence Field(0)/Forest(1) 1 1.350649 0 1 2 1.422346 1 1 3 1.543496 1 1 4 1.161905 0 1 5 1.103558 0 1 6 1.051429 0 0 7 1.302726 0 1 8 1.511077 0 1 9 0.708489 0 0 10 0.854767 0 0 11 0.978668 0 0 12 0.679467 0 0 13 0.940485 0 0 14 0.822249 0 0 15 0.829031 0 0 16 1.13595 0 1 17 1.056618 0 0 18 0.946736 1 1 19 1.089608 0 1 20 1.232386 0 1 21 0.761306 0 0 22 1.153032 0 0 23 1.195787 0 1 Description Statistic
Most common bird species Pica pica (721 total) Figure 12. The species of bird and the amount of times they were observed.
Second most common bird species
Corvus corone (250 total) Third most common bird species Ardea cinerea (176 total) Highest Shannon diversity index 1.543495682
Lowest Shannon diversity index 0.679466546 P-value Rat vs No rat 0.1660079 P-value Forest vs Field 4.91568E-05 Table 2. This table shows the descriptive statistics.
