Effects and consequences of natural and artificial light at night on small mammals in peri-urban Johannesburg, South Africa Tasha Oosthuizen Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy in the Faculty of Science at the University of Witwatersrand, Johannesburg May 2024 II Declaration I, Tasha Oosthuizen, declare that this thesis submitted for the degree of Doctor of Philosophy at the University of Witwatersrand, Johannesburg, is my own work. It has not been submitted before for any degree or examination at any other university. Tasha Oosthuizen 27/05/2024 III Abstract Studies investigating artificial light at night (ALAN) have increased over recent years. However, research examining the influence of ALAN on southern African small mammal species are lacking and even information on their basic biology is scarce. To close this knowledge gap, I investigated the effect of ALAN on different facets of animal behaviour in African small mammals. Firstly, I evaluated the impact of the natural (lunar cycle) and ALAN on the community composition and species abundance in two populations of small mammals. I chose two field sites: one facing Johannesburg (exposed to ALAN) and one facing away. I conducted mark-recapture trapping to ascertain the occurrence and abundance of small mammals. The Light site had both a higher species composition and a higher animal abundance when compared to the composition and abundance of the Dark site. The lunar cycle had an effect; on nights with a full moon, the species composition and animal abundance of both study sites declined, while on new moon nights, the opposite occurred, with an increase in both the species composition and abundance on the Light and Dark sites. The absence of a negative ALAN effect on the Light site can potentially be ascribed to the availability of microhabitats for small mammals to escape illumination, leaving them seemingly unaffected. Next, I assessed the locomotor activity of three species of commonly occurring rodents on the study area, one crepuscular (19 single-striped grass mice, Lemniscomys rosalia), one species with reportedly variable activity (19 angoni vlei rats, Otomys angoniensis) and one nocturnal (19 southern multimammate mice, Mastomys coucha). They were captured at a different location than the mark-recapture study sites and tested in captivity under natural (exposed to natural light and temperature changes), laboratory (standard laboratory conditions; 12h light:12h dark and constant temperature) and ALAN treatments. Lemniscomys rosalia exhibited crepuscular activity under all three experimental treatments, Otomys angoniensis was mostly nocturnal with some diurnal activity. The temporal activity profiles of the two species that showed some activity during the light hours were unaffected by ALAN. Mastomys coucha displayed strictly nocturnal activity during the natural and laboratory treatments, but during ALAN treatments the temporal activity profiles of some animals shifted so that they were active during the start of the day. Lemniscomys rosalia and O. angoniensis were more active under the natural treatment, whilst M. coucha was more active in the laboratory treatment. When exposed to 2 Lux ALAN presented remotely, there was no effect on the level of activity in O. angoniensis, L. rosalia showed a reduction of about 20% in its activity, whereas M. coucha reduced its activity by more than 50%. Finally, I studied how ALAN impacted the foraging behaviour of the three species under four treatments (during the day, at night, 2 Lux ALAN and 10 Lux ALAN). Foraging behaviour differed in the three species under different light conditions. Lemniscomys rosalia was risk-averse when feeding during the diurnal and nocturnal (no light at night) treatments. Otomys angoniensis showed irregular responses in their foraging behaviour under all foraging treatments. Mastomys coucha showed no differences when feeding under any of the nocturnal treatments, but it was inactive under the diurnal treatment. Overall, my study revealed that the effect of IV ALAN is not similar for all small mammalian species and appear to depend on both the spatial and temporal niches that the different species occupy. Strictly nocturnal animals seem to be affected the most, whereas animals that are active during the day showed lesser responses. Given the rapid increase in urbanisation and anthropogenic disturbances, more and more species are exposed to ALAN. Species that prefer darker, more secluded habitats appear to be more vulnerable and at higher risk of local extinctions as a result of disturbances, such as ALAN and habitat transformation. My study highlights that ALAN affects both nocturnal and diurnal rodents to the extent that it can have fitness consequences, including changed active times, foraging efficiency, movement patterns and susceptibility to predation. Finally, the disruption of rodent behaviour can have cascading effects for ecosystems and my study also emphasises the importance of safeguarding our night skies to protect biodiversity. V Acknowledgements I am grateful to my supervisors, Prof Neville Pillay and Dr Marietjie Oosthuizen, for guiding me through this journey, providing insight, proofreading multiple drafts and for answering the phone every time I called to confirm or discuss something for the tenth time. I am so thankful that I got to experience a helpful and beneficial relationship with my supervisors for this PhD, as it would have been so much more difficult if it weren’t for the open communication and support, I received from you both. I owe a huge thanks to Marietjie who encouraged me to continue with my academic journey after I completed my Masters, and to pursue my love for small mammals. I am also grateful to Marietjie and her husband Albert for opening their home to me as a second home, enduring weekly visits for fresh drinking water and clean laundry and some emotional support that was mostly absent in the field, except from the mice and monkeys of course. I am appreciative of Prof Lee Berger for supplying funding, fieldwork housing, laboratory space and a field vehicle. Additionally, I thank Cradle Nature Reserve, its staff and Tim Nash for allowing me to conduct my research on their property. To Oom Coen and Tannie Lorraine, I am grateful that you were there whenever I needed something and even to just see a friendly face when I have mostly only seen animals scared by my presence. My PhD journey would have been a lot harder if it wasn’t for all the volunteers who came out to help me and keep me company. There have been too many to name each one individually, but a lot of volunteers came out to help more than once and I would like to name them specifically. Thank you to Jolize Kruger, Arantxa Blecher, Cat Pinto, Helen van der Merwe, Thys Louw, Tumisho Phoshoko and Armand Engelbrecht. I specifically want to mention my gratitude to Caitlin van der Merwe and Arantxa Blecher for proofreading many of my drafts and giving helpful feedback on each one. I would also like to thank Eugene Oosthuizen, who came out on multiple days to help with some electrical work and made some equipment related to my laboratory experiments. Ultimately, this PhD journey would not have been possible without my Heavenly Father. I know this is a door You opened for a reason, and one of them was to test my faith throughout this challenging journey. I am eternally grateful for the love and support from my family and friends. I will never be able to accurately express how thankful I am for my parents, who were always just a phone call away, who had to provide emotional support and words of encouragement over the phone daily without complaint. My friends had to endure many long voice notes of me expressing frustration of a failed experiment or me being in the middle of a field with some or other issue. I am so grateful to you all who listened patiently and tried to give advice, but mostly for just being there when I needed someone to vent to. VI Lastly, I am indebted to the animals who formed part of my study. They were not only part of my study and the reason I am able to write this thesis, but when times were tough, they made my day and put a smile on my face and was literally the reason I woke up in the morning, when all I wanted to do was curl up in bed and forget about PhD for a day or two. VII Table of contents Declaration............................................................................................................................................ II Abstract ............................................................................................................................................... III Acknowledgements .............................................................................................................................. V List of Figures .................................................................................................................................... XII List of Tables ..................................................................................................................................... XVI Chapter 1: General Introduction ......................................................................................................... 1 1.1 Circadian biology ............................................................................................................................ 1 1.2 Lunar cycle and animal behaviour ................................................................................................ 2 1.3 Artificial light at night and animal behaviour .............................................................................. 3 1.4 Risk-sensitive foraging behaviour ................................................................................................. 5 1.5 General biology of my study species .............................................................................................. 6 1.6 Study area ........................................................................................................................................ 8 1.7 Motivation for this study ................................................................................................................ 9 1.8 Aim and Objectives ....................................................................................................................... 10 1.9 Layout of thesis ............................................................................................................................. 11 1.10 References .................................................................................................................................... 12 VIII Chapter 2: The impact of artificial light at night and the lunar cycle on small mammal trapping near a large metropolitan area in South Africa ................................................................................ 17 Abstract ................................................................................................................................................ 17 2.1 Introduction ................................................................................................................................... 17 2.2 Materials and Methods ................................................................................................................. 19 2.2.1 Trapping sites ........................................................................................................................... 19 2.2.2 Animal trapping and measurements ....................................................................................... 20 2.2.3 Data analyses ........................................................................................................................... 22 2.3 Results ............................................................................................................................................ 23 2.3.1 Trapping success ...................................................................................................................... 23 2.3.2 Diversity indices ........................................................................................................................ 25 2.3.3 Lunar cycle ............................................................................................................................... 25 2.3.4 Weather variables .................................................................................................................... 26 2.4 Discussion....................................................................................................................................... 27 2.5 Conclusions .................................................................................................................................... 31 2.6 References ...................................................................................................................................... 31 Chapter 3: Wild mice in an urbanized world: effects of light at night under natural and laboratory conditions in the single-striped grass mouse (Lemniscomys rosalia) ........................... 35 Abstract ................................................................................................................................................ 35 3.1 Introduction ................................................................................................................................... 35 3.2 Materials and Methods ................................................................................................................. 38 3.2.1 Animal capture ......................................................................................................................... 38 3.2.2 Animal maintenance ................................................................................................................ 38 3.2.3 Experimental design ................................................................................................................. 38 IX 3.2.4 Data analyses ........................................................................................................................... 39 3.3 Results ............................................................................................................................................ 40 3.4 Discussion....................................................................................................................................... 44 3.5 Conclusions .................................................................................................................................... 46 3.6 Acknowledgments ......................................................................................................................... 47 3.7 Declarations of interest ................................................................................................................. 47 3.8 Funding sources: ........................................................................................................................... 47 3.9 References ...................................................................................................................................... 47 Chapter 4: Temporal activity patterns under natural and laboratory conditions in the Angoni vlei rat, Otomys angoniensis ............................................................................................................... 51 Abstract ................................................................................................................................................ 51 4.1 Introduction ................................................................................................................................... 51 4.2 Materials and Methods ................................................................................................................. 54 4.2.1 Animal capture and maintenance ............................................................................................ 54 4.2.2 Experimental design ................................................................................................................. 54 4.2.3 Data analyses ........................................................................................................................... 55 4.3 Results ............................................................................................................................................ 56 4.4 Discussion....................................................................................................................................... 60 4.5 Conclusions .................................................................................................................................... 61 4.6 References ...................................................................................................................................... 62 X Chapter 5: A pioneer rodent species faces negative consequences under expanding urbanisation: insight into the activity profiles of the nocturnal multimammate mice (Mastomys coucha) ........ 65 Abstract ................................................................................................................................................ 65 5.1 Introduction ................................................................................................................................... 65 5.2 Materials and methods ................................................................................................................. 68 5.2.1 Study animals ........................................................................................................................... 68 5.2.2 Experimental design ................................................................................................................. 69 5.2.3 Data analyses ........................................................................................................................... 70 5.3 Results ............................................................................................................................................ 70 5.4 Discussion....................................................................................................................................... 76 5.5 Conclusions .................................................................................................................................... 78 5.6 References ...................................................................................................................................... 79 Chapter 6: Risk-sensitive foraging under different ALAN treatments in three African rodents 82 Abstract ................................................................................................................................................ 82 6.1 Introduction ................................................................................................................................... 82 6.2 Materials and methods ................................................................................................................. 88 6.2.1 Study site .................................................................................................................................. 88 6.2.2 Test species ............................................................................................................................... 89 6.2.3 Experimental design ................................................................................................................. 89 6.2.3.1 Seed preference testing for all three species ......................................................................... 89 6.2.3.2 Foraging experiment: L. rosalia and O. angoniensis ............................................................. 91 6.2.3.3 Preference and foraging experiment: M. coucha .................................................................. 92 6.2.4 Data analysis ............................................................................................................................ 93 XI 6.3 Results ............................................................................................................................................ 93 6.3.1 Latency to start moving............................................................................................................ 93 6.3.2 Latency to approach seeds ....................................................................................................... 96 6.3.3 Duration of feeding .................................................................................................................. 98 6.3.4 Frequency of feeding .............................................................................................................. 100 6.3.5 Risk-sensitive foraging ........................................................................................................... 102 6.3.6 Duration spent in the shelter.................................................................................................. 102 6.3.7 Mass of seeds consumed ........................................................................................................ 105 6.4 Discussion..................................................................................................................................... 107 6.4.1 Lemniscomys rosalia .............................................................................................................. 107 6.4.2 Otomys angoniensis ............................................................................................................... 108 6.4.3 Mastomys coucha .................................................................................................................. 109 6.4.4 Sex differences ........................................................................................................................ 110 6.5 Conclusions .................................................................................................................................. 110 6.6 References .................................................................................................................................... 110 Chapter 7: General Discussion ........................................................................................................ 114 7.1 Objectives and key findings of my study .................................................................................. 115 7.2 ALAN and African rodents ........................................................................................................ 118 7.3 Implications of my research and future study areas ................................................................ 120 7.4 Conclusions .................................................................................................................................. 121 7.5 References .................................................................................................................................... 122 Appendix ............................................................................................................................................ 126 XII List of Figures Chapter 1: Figure 1. The single-striped grass mouse. Photo credit: MK Oosthuizen ........................................... 6 Figure 2. The Angoni vlei rat. Photo credit: MK Oosthuizen .............................................................. 7 Figure 3. The southern multimammate mouse. Photo credit: MK Oosthuizen ................................... 8 Figure 4. A & B - The artificial sky brightness of Africa, with the area of my study site indicated in the light blue square. C - The reported skyglow for the Cradle Nature Reserve and Johannesburg, the nearest and largest metropolitan area to the study site. ..................................................................................... 9 Chapter 2: Figure 1. The locations of the Light and Dark sites on the Cradle Nature Reserve property. The Light site faced Johannesburg and the Dark site faced the opposite direction. .............................................. 20 Figure 2. The number of small mammals captured by season on each trapping site at the Cradle Nature Reserve, Gauteng, South Africa. ........................................................................................................... 23 Figure 3. The total number of captures per day per season on the Dark and Light sites with the associated moon illumination percentage of each day. ......................................................................... 26 Chapter 3: Figure 1. The total activity counts of all Lemniscomys rosalia (mean ± SE) during the three environmental treatments (LAB – Laboratory treatment, ALAN – 2 Lux light at night, NAT – semi- natural environmental treatment). ......................................................................................................... 40 Figure 2. The activity of Lemniscomys rosalia over the 24h day for the three experimental treatments. A, C, E – Hourly activity counts (mean ± SE) of all individuals and the ambient temperatures during the three different treatments. B, D, F - Actograms of a single representative individual. The LAB (A and B) and ALAN (C and D) experimental treatments were on a square wave regime and a constant temperature, whereas NAT (E and F) mice were exposed to the natural dawn and dusk (approximately 05:00-07:00 and 17:00-19:00) and naturally fluctuating temperatures. ................................................ 41 Figure 3. A - The total activity counts (mean ± SE) of male and female Lemniscomys rosalia during the three different experimental treatments. B - The total activity counts (mean ± SE) of male and female Lemniscomys rosalia during the light phase and dark phase during the three experimental treatments. C XIII - The total activity counts (mean ± SE) of Lemniscomys rosalia during the dark and light phases of the three different experimental treatments. ............................................................................................... 43 Chapter 4: Figure 1. The hourly activity counts (mean ± SE) of O. angoniensis during each experimental treatment illustrated over the 24h day. .................................................................................................................. 56 Figure 2. The mean hourly activity counts (mean ± SE) recorded for O. angoniensis individuals during the dark and light hours across experimental treatments. ..................................................................... 57 Figure 3. The mean hourly activity counts (mean ± SE) of O. angoniensis individuals across the day per experimental treatment with females and males illustrated separately. .......................................... 57 Figure 4. A double-plotted actogram of a single O. angoniensis female to illustrate activity during the different experimental treatments. LAB – Laboratory 12L:12D, rLAN – 2 Lux remote LAN and NAT – Natural ambient conditions. ............................................................................................................... 59 Chapter 5: Figure 1. An experimental timeline representing each treatment: NAT – natural environmental treatment, LAB – laboratory conditions, rLAN – remote light at night and dLAN – direct light at night. .............................................................................................................................................................. 69 Figure 2. The mean hourly activity counts (mean ± SE) of all M. coucha individuals in four experimental treatments. NAT – natural environmental conditions, LAB – laboratory conditions, rLAN – light at night on the opposite side of the room and dLAN – light at night suspended directly above the cages. ..................................................................................................................................................... 71 Figure 3. The hourly activity counts (mean ± SE) for all individuals during the 24h of the day in the different experimental treatments. NAT – natural environmental conditions, LAB – laboratory conditions, rLAN – light at night on the opposite side of the room and dLAN – light at night suspended directly above the cages. ...................................................................................................................... 72 Figure 4. The mean hourly activity counts (mean ± SE) of all the female and male M. coucha in four different experimental treatments. NAT – natural environmental conditions, LAB – laboratory conditions, rLAN – light at night on the opposite side of the room and dLAN – light at night suspended directly above the cages. ...................................................................................................................... 73 Figure 5. The mean activity counts (mean ± SE) of all the female and male M. coucha in the dark and light phases of the day. NAT – natural environmental conditions, LAB – laboratory conditions, rLAN – XIV light at night on the opposite side of the room and dLAN – light at night suspended directly above the cages. .................................................................................................................................................... 74 Figure 6. A double-plotted actogram of a single M. coucha to illustrate the changes in onset (start of activity) and offset (end of activity) during the different experimental treatments. NAT – Natural ambient conditions, LAB – Laboratory 12L:12D, rLAN – 2 Lux remote LAN, dLAN – 2 Lux direct LAN. ..................................................................................................................................................... 75 Chapter 6: Figure 1. Top view of the seed preference test arena, with the shelter on one side and the seeds in the seed containers on the opposite side. A mesh lid was placed on top of the arena and the camera was suspended over the arena. ..................................................................................................................... 90 Figure 2. The difference in latency to move (s) between treatments for all species. ........................... 95 Figure 3. The latency to approach seeds (s) during each treatment for each species. .......................... 97 Figure 4. The latency to approach the seeds (mean ± SE) for O. angoniensis females and males for all treatments. ............................................................................................................................................. 98 Figure 5. The duration of feeding (s) in each treatment for all the species. ......................................... 99 Figure 6. The frequency feeding during different treatment for all species. ...................................... 101 Figure 7. The duration spent in the shelter (s) in each treatment for all species. ............................... 104 Figure 8. The duration spent in the shelter (s; mean ± SE) in the different treatments for O. angoniensis females and males. .............................................................................................................................. 105 Figure 9. The mass of seeds that were consumed after each treatment (g) for each species. ............. 106 Appendix: Appendix Figure 1. The Light site facing Johannesburg on the Cradle Nature Reserve. ................ 128 Appendix Figure 2. The Dark site facing the opposite direction from the Light site on the Cradle Nature Reserve. ............................................................................................................................................... 128 Appendix Figure 3. The field laboratory on the Cradle Nature Reserve property. ........................... 129 Appendix Figure 4. The standard laboratory room with no external windows in which the laboratory and ALAN treatments were conducted. .............................................................................................. 129 XV Appendix Figure 5. The area where the study animals were kept during the natural environmental treatment. ............................................................................................................................................ 130 XVI List of Tables Chapter 2: Table 1. The seasonal sampling period dates for the Dark and Light sites on the Cradle Nature Reserve. .............................................................................................................................................................. 21 Table 2. The abundance of small mammals and species richness by season in Dark and Light trapping sites at the Cradle Nature Reserve. M. coucha, O. angoniensis and R. d. chakae were identified genetically. ............................................................................................................................................ 24 Table 3. Bray-Curtis dissimilarity and diversity indices by season and site for small mammal trapping done on Cradle Nature Reserve, Gauteng, South Africa. ...................................................................... 25 Table 4. The number of captures based on whether it rained and minimum temperature during the night on both sites on the Cradle Nature Reserve. ........................................................................................ 27 Chapter 6: Table 1. Predictions for the influence of treatment on six response variables on the foraging behaviour of L. rosalia (L. r.), O. angoniensis (O. a.) and M. coucha (M. c.). ...................................................... 85 Table 2. The number of individuals of L. rosalia, M. coucha and O. angoniensis that preferred different seed types. Otomys angoniensis was tested with oats instead of millet (see text).................. 90 Table 3. The foraging variables recorded from the videos of each treatment with associated descriptions. .............................................................................................................................................................. 92 Appendix: Appendix Table 1. The colour and its associated artificial brightness scores measured in μcd/m2 which corresponds to the world maps. This table was adapted from Table 1 in Falchi et al. (2016) and the sky brightness measures were accessed at https://lightpollutionmap.info/................................................ 126 Appendix Table 2. The models considered for the number of animals and species caught. ............. 127 Appendix Table 3. The contributions of each weather variable to five separate principal components (PCs). .................................................................................................................................................. 127 https://lightpollutionmap.info/ 1 Chapter 1: General Introduction 1.1 Circadian biology Most living organisms have evolved endogenous rhythms that coincide with external environmental cycles (Buijs et al., 2003; Dominoni et al., 2016). Although endogenous rhythms are self-sustaining, they are not exactly 24 hours long and must be synchronised or entrained to rhythmic external cycles daily (Buijs et al., 2003). The daily solar cycle is the most predictable environmental cycle and therefore, light is the primary environmental cue to which animals entrain their rhythms (Benstaali et al., 2001; Tapia-Osorio et al., 2013). Other environmental factors such as temperature, feeding times and in some cases locomotor activity can also affect the endogenous rhythms of animals. In the absence of light, these factors are to some extent able to entrain circadian rhythms, but when light is present, it overrides these effects and the secondary external cycles re-enforce light entrainment (Benstaali et al., 2001; Ikeno et al., 2014). By synchronising behavioural and physiological processes to cyclic environmental factors, animals gain extrinsic adaptive fitness, since it enables an organism to predict environmental conditions (Sharma, 2003). Animals can also obtain intrinsic fitness by synchronising their internal processes such that the entire organism is in harmony with its environment (Sharma, 2003). Light influences the internal biological clock that is responsible for the generation of innate biological rhythms (Buijs et al., 2003; Dominoni et al., 2016). In mammals, the master biological clock is called the suprachiasmatic nucleus (SCN) and is located above the optic chiasm in the hypothalamus of the brain (Buijs et al., 2003; Tapia-Osorio et al., 2013). The SCN receives photic information from a subset of light-sensitive retinal ganglion cells (RGCs) located within the retina of the eye. This system enables animals to entrain to the environmental light/dark cycle, identify environmental changes, and elicit an appropriate response (Benstaali et al., 2001; Tapia-Osorio et al., 2013). Most animals have a specific temporal niche, such that approximately 26% of extant mammals are predominantly diurnal, while about 44% of mammals are nocturnal (Benstaali et al., 2001; Jones et al., 2009; Prugh and Golden, 2014). The remaining proportion of mammals are either crepuscular, showing activity bouts during dawn and dusk, or cathemeral, i.e., species that display irregular activity periods during both day and night (Benstaali et al., 2001; Jones et al., 2009; Prugh and Golden, 2014). Diurnality in mammals is thought to have evolved from nocturnality (Roll et al., 2006) and diurnal animals have preserved some of the nocturnal morphological features, such as retinae that are rod-dominated (Peichl, 2005). However, the majority of diurnal species have higher proportions of cones than nocturnal species (Peichl, 2005). Cones are important for colour detection; they are highly acute but not very sensitive to light (Peichl, 2005). Rods have lower acuity but are very sensitive to light and enhance vision in low 2 light conditions (Peichl, 2005). In the presence of light, diurnal species become more active, which is termed positive masking (Ikeno et al., 2014). Negative masking is the presence of light that causes a decrease in activity in nocturnal species (Ikeno et al., 2014). Thus, the presence of light awakens diurnal animals and promotes sleep in nocturnal animals (Yan et al., 2020). Temporal niche partitioning in the timing of activity occurs in sympatric species that have specialised morphological adaptations of the eyes to be active within certain periods (Bennie et al., 2014; Lear et al., 2021). However, morphological adaptations alone do not determine the activity period, since factors, such as predation pressure, competition, food resources, weather conditions and human disturbances all play a role in defining the temporal niche of a particular species (Bennie et al., 2014; Lear et al., 2021). 1.2 Lunar cycle and animal behaviour Rodents are important prey species and often have to weigh the costs and benefits of foraging against potential predation risks (Lima and Dill, 1990; Mandelik et al., 2003). Most rodents are nocturnal (Hawkins and Golledge, 2018) and can use the moonlight to navigate under low light levels. The lunar cycle creates different intensities of natural light depending on the phase of the moon (Weaver, 2011). A full lunar cycle is 29.5 days and the light intensity of a new moon is around 0.0001 Lux, whereas a full moon can reach intensities of up to 2 Lux (Weaver, 2011). Visibility to predators can lead to several behavioural changes in prey species, such as increased vigilance during full moon nights (Russart and Nelson, 2018). During moonlit nights, prey animals tend to avoid foraging in open areas because increased illumination creates greater predation risk (Mandelik et al., 2003). There are several notable examples. During full moon nights, fewer wood mice (Apodemus sylvaticus) were trapped compared with new moon nights (Perea et al., 2011). The higher illumination from a full moon resulted in increased foraging efficiency of short-eared owls (Asio flammeus), but their prey species, deer mice (Peromyscus maniculatus), decreased their activity and feeding (Clarke, 1983). Nocturnal common spiny mice (Acomys cahirinus) that foraged in an open habitat, were significantly influenced by moonlight, with mice visiting fewer artificial food trays to limit detection under moonlit nights (Mandelik et al., 2003). Allenby’s gerbils (Gerbillus andersoni allenbyi) were more vigilant during the brightest full moon phase followed by the waning moon, waxing moon and lastly the new moon and they ceased foraging sooner during the waxing moon, followed by the full, new, and waning moon (Kotler et al., 2010). During the brighter part of the lunar cycle (full and waxing moon), these gerbils also increased vigilance and reduced foraging, resulting in a poorer body condition, but as the cycle progressed and starvation increased, the gerbils spent more time foraging despite exposure and risk 3 (Kotler et al., 2010). The response of animals to varying levels of light intensity over the lunar cycle can facilitate predictions of behavioural alterations during artificial light at night. 1.3 Artificial light at night and animal behaviour Artificial light close to urbanised areas can be brighter than moonlight and is more constant, both throughout the night and over consecutive days. As a result, artificial light at night (ALAN) has the potential to affect the natural nocturnal behaviour of animals more severely than moonlight (Falchi et al., 2016; Russart and Nelson, 2018). The rapid growth rate of the human population increases the need for more structures and developments, and it encroaches on natural habitats, impacting wildlife. Most roads in urbanised areas are lit by streetlights and depending on the light bulbs used, the light intensities of streetlights range between 1 and 10 Lux at ground level (Preto and Gomes, 2019), which far exceeds the natural light intensities reflected by the moon. Light pollution is a broad term, defined as the change in the natural low light levels at night as a result of the increased artificial light at night (Raap et al., 2015). The two main categories of light pollution are point source and skyglow. Point source is artificial light that is concentrated within a specific area and can be further separated into light trespass, glare, over-illumination, and clutter (Rajkhowa, 2012). Light trespass is the presence of unwanted light found in a person’s property, which can result in insufficient sleep (Gaston et al., 2012; Rajkhowa, 2012). Glare, as a result of street and vehicle lights, is a common safety issue, since the eye is not always able to process the high light intensity (termed photostress) and the after-effects can persist for up to an hour after exposure (Gaston et al., 2012; Rajkhowa, 2012). Over-illumination is the excessive use of lights resulting from improper positioning of lights that provide light beyond the desired area (Gaston et al., 2012; Rajkhowa, 2012). The inappropriate arrangement of lights is termed light clutter and can lead to disorientation, especially along roads (Gaston et al., 2012; Rajkhowa, 2012). Sky glow, in contrast, is the product of all the extra light that is reflected into the sky and then reflected back to earth by the atmosphere, increasing the brightness of the sky (Gaston et al., 2012; Rajkhowa, 2012). Skyglow is measured in magnitude per square arc-second (mag/arcsec2) and can range from 22.0 mag/arcsec2 for the darkest areas and less than 17.5 mag/arcsec2 in the brightest areas (https://lightpollutionmap.info/, Appendix Table 1). The artificial brightness is measured in μcd/m2, and ALAN in most laboratory studies is measured in Lux. The different ways to measure light depend on the scientific field of study (Hänel et al., 2018), I used a handheld light meter, measuring in Lux, which is in accordance with past laboratory studies. Artificial light at night can have several negative effects on people, including disrupted sleep patterns, melatonin suppression, depressive symptoms, and fatigue (reviewed in Cho et al., 2015). Several studies have shown negative impacts of ALAN on wildlife (Gaynor et al., 2018; Łopucki et al., https://lightpollutionmap.info/ 4 2021; Sanders et al., 2021; Willems et al., 2021). Under ALAN, diurnal and crepuscular animals can extend their active hours, whereas nocturnal species often decrease their activity throughout the night (Russart and Nelson, 2018). Animals which experience ALAN can also alter their general behaviour in terms of their active period, home range size, and their interactions with conspecifics (Hoffmann et al., 2019). Species that show some diurnal activity, such as striped field mice (Apodemus agrarius) and bank voles (Myodes glareolus), decreased their activity throughout the daylight hours under ALAN (Hoffmann et al., 2019). In addition, these species showed no difference in diurnal and nocturnal home range sizes under ALAN, in contrast to the distinctly larger home range sizes during the day under natural conditions (Hoffmann et al., 2019). This was possibly because of activity asynchronisation under ALAN, since individual interactions decreased and conspecific home ranges did not overlap and as a result could lead to missed mating opportunities (Hoffmann et al., 2019). This can also have an indirect effect on predation risk, since animals can no longer rely on group safety. Likewise, by being active throughout the entire day, animals can now be predated on by a wider variety of predators (Hoffmann et al., 2019). Common spiny mice (A. cahirinus) reduced their general activity, as well as foraging behaviour (i.e., the number of visits and movements between food patches decreased significantly) during ALAN (Rotics et al., 2011). This change in risk perception reduces movement between patches and increases within-patch use (Rotics et al., 2011), which then decreases predation risk by birds of prey, such as owls, that hunt moving prey (Mandelik et al., 2003; Rotics et al., 2011). During continuous ALAN, animals will alter their spatial movement, impacting landscape connectivity by avoiding more exposed foraging patches (Bird et al., 2004). Pinyon mice (Peromyscus truei) were trapped less frequently under ALAN, possibly because they were avoiding illuminated areas (Willems et al., 2021). ALAN can also be beneficial to some species if they increase their activity during the night (Dominoni et al., 2016), allowing for better detection of predators and food (Prugh and Golden, 2014). Six wader bird species increased their nocturnal food intake by 78% when they foraged in areas with ALAN (Santos et al., 2010). Even though the increased nocturnal activity of diurnal species under ALAN is widely mentioned anecdotally, there are a limited number of studies with empirical evidence to test its effects. Thus, there is a need for more studies focussing on how diurnal species experience ALAN and their responses to light at night. Living in urban environments could provide benefits that are absent in non-urban, more natural areas (Łopucki et al., 2021). Urban animals have greater and more constant access to food resources. Due to the limited suitable habitat and increased availability of food in an urban environment, one could expect increased aggressive interactions between individuals (Łopucki et al., 2021). However, the opposite was observed in the striped field mouse (A. agrarius), where individuals were more tolerant of each other when food was present, apparently to prevent physical injury and excess stress, or to avoid wasting time and energy on competition rather than foraging (Łopucki et al., 2021). An alternative explanation can be that individuals cannot defend large quantities of food, leading to a greater tolerance 5 of conspecifics (Łopucki et al., 2021; Thomas et al., 2018). Yet, this tolerance was not observed in common spiny mice (A. cahirinus) in a more natural setting with ALAN; at higher ambient light intensities, there were greater levels of intraspecific competition since most mice were competing to forage in the shaded areas, which offered safety from predators (Rotics et al., 2011). Competition was greater because the mice lowered their overall activity in the light, and focussed their foraging to a very limited period after the nocturnal illumination (Rotics et al., 2011). 1.4 Risk-sensitive foraging behaviour All animals make foraging decisions daily, while taking different risks, such as ALAN, into account. These decisions include what to eat, where to eat, how long to spend eating, and how to obtain the food with the lowest estimated risk possible. The Optimal Foraging Theory (OFT) was developed in an attempt to understand and predict these behaviours (Pyke et al., 1977). The OFT assumes that the consumer will always forage optimally, with regard to the diet choice, choice of feeding patch, choice of when to leave a patch and the movement decisions between patches (Bartumeus and Catalan, 2009; Pyke et al., 1977). The Marginal Value Theorem and the OFT together maintain that the animal will constantly maximise its fitness by gaining the most energy while foraging in a patchily distributed resource environment (Bartumeus and Catalan, 2009; Charnov, 1976), but many scientists have challenged the OFT. One of the questions is whether there is optimal behaviour in nature since an animal would have to know and learn everything from their environment to make optimal choices, but this would require long periods of time, which is not always available to a foraging animal (Craft, 2016; Pierce and Ollason, 1987). There are also various risks that each consumer faces when foraging, which are not taken into account (Craft, 2016; Pierce and Ollason, 1987). Yet, OFT helped develop other foraging theories, for example, the Risk-Sensitive Foraging Theory (RSFT) (Barnard et al., 1985). Foraging choices under risky situations are explained by the RSFT (Barnard et al., 1985; Craft, 2016), maintaining that a risk-sensitive or risk-averse individual will opt for the food choice with a fixed return, whereas a risk-prone individual will show bias towards the food choice with a variable return (Barnard et al., 1985; Craft, 2016). Bank voles (M. glareolus) showed risk-sensitive foraging in terms of microhabitat use, since individuals foraged under vegetation of a certain height that they perceived as safe, and each individual had its own range of preferred vegetation heights (Dammhahn et al., 2022). Australian rodents showed a similar response, where more food was consumed in microhabitats of dense and unburnt vegetation, compared with burnt and exposed microhabitats (Doherty et al., 2015). The RSFT is important for understanding the influence of ALAN on different populations of small mammals and their perception of risk and their responses to foraging decisions. The Mongolian five- toed jerboa (Allactaga sibirica) decreased its food searching efforts under ALAN, yet increased its 6 vigilance and spent less time in each foraging patch, leaving more food uneaten (Zhang et al., 2020). Santa Rosa beach mice (Peromyscus polionotus leucocephalus) limited their activity in illuminated patches and also consumed fewer seeds closer to a light source (Bird et al., 2004). Dwarf striped hamsters (Cricetulus barabensis) foraged faster in a food patch under ALAN, and reduced their active times and thus body mass in patches with both low vegetation and the presence of ALAN (Shuai et al., 2023). Overall, many small mammals perceive ALAN as high risk and adjust their foraging behaviour accordingly. 1.5 General biology of my study species As the literature suggests, both the lunar cycle and ALAN influence the behaviour of multiple species with different temporal niches. Thus, I studied a variety of species, but three were the focus of multiple chapters in this thesis and they were selected based on their abundance and their temporal niche to represent a range of temporal preferences. These were the single-striped grass mouse (Lemniscomys rosalia), the Angoni vlei rat (Otomys angoniensis), and the southern multimammate mouse (Mastomys coucha). The single-striped grass mouse (Family: Muridae, Figure 1) is terrestrial and has been described as diurnal with crepuscular activity, or crepuscular with diurnal bouts (Kingdon, 2013; Skinner and Chimimba, 2005). This species occurs within tall and dense vegetation and can be found singly, in pairs, or in small groups, and is thus tolerant of conspecifics (Skinner and Chimimba, 2005). However, some laboratory studies suggest aggression between conspecifics (Kingdon, 2013). This species is granivorous and breeds during the summer months (Monadjem et al., 2015; Skinner and Chimimba, 2005). Figure 1. The single-striped grass mouse. Photo credit: MK Oosthuizen 7 The Angoni vlei rat (Family: Muridae, Figure 2) is terrestrial and has been described as diurnal, crepuscular or nocturnal (Kingdon, 2013; Skinner and Chimimba, 2005). It is found in grasslands and woodlands, often close to water sources (Kingdon, 2013; Skinner and Chimimba, 2005). This species has been observed singly, in pairs or in small groups and breeding can occur throughout the year, but peaks in the summer months (Skinner and Chimimba, 2005). It is strictly herbivorous (Skinner and Chimimba, 2005). Figure 2. The Angoni vlei rat. Photo credit: MK Oosthuizen The southern multimammate mouse (Family: Muridae, Figure 3) is terrestrial and strictly nocturnal (Kingdon, 2013; Skinner and Chimimba, 2005). This species can occur in a wide range of habitats, including human-dense areas (Kingdon, 2013; Skinner and Chimimba, 2005). It is omnivorous, but relies mostly on grass seeds and will opportunistically eat arthropods (Skinner and Chimimba, 2005). In resource-scarce times, it can be cannibalistic (Skinner and Chimimba, 2005). It breeds aseasonally, and under favourable conditions, it can experience population eruptions, with the potential to produce a maximum of 24 pups per litter (Monadjem et al., 2015; Skinner and Chimimba, 2005). Since individual home ranges show a high degree of overlap, it is thought to be tolerant towards conspecifics. It is also a post-burn pioneer that colonise previously disrupted (i.e., burnt) areas after which specialist species become the dominating species (Perrin et al., 2001; Skinner and Chimimba, 2005). 8 Figure 3. The southern multimammate mouse. Photo credit: MK Oosthuizen 1.6 Study area I conducted my research at the Cradle Nature Reserve (-25.9214, 27.8503), Gauteng, South Africa, an approximately 9000-hectare property within the Magaliesberg Biosphere Reserve (MBR). The MBR consists of a grassland plateau and sub-Saharan savanna. There are also some Afromontane Forest fragments still present in the area (Mucina and Rutherford, 2006). This site is part of a world heritage site, namely the Cradle of Humankind World Heritage site, which aids in maintaining the integrity of the area and the biodiversity that falls within it (https://magaliesbergbiosphere.org.za/). This site includes cultural heritage along with many archaeological sites. This reserve is uniquely suited for my research objectives, since it is about 38km from the centre of Johannesburg, the largest metropolitan area in southern Africa. Johannesburg is rapidly expanding and as a result has a high concentration of ALAN. The Cradle Nature Reserve has a skyglow measurement of between 20.4 and 19.4 magnitude/arc second2, whereas urbanised areas of Johannesburg measure at approximately 18.5 magnitude/arc second2 (Figure 4, https://lightpollutionmap.info/, Appendix Table 1). The area surrounding Johannesburg measures the highest artificial sky brightness compared with the rest of southern Africa (Figure 4). Within my study area, I chose three separate study sites. One site faced Johannesburg with a high concentration of ALAN reflected from the city (i.e., Light site) and one site faced the opposite direction and received less ALAN (i.e., Dark site) in comparison to the Light site. Lastly, I chose a site separate from the first two with a high abundance of the three study species to catch the laboratory animals without impacting the mark-recapture study. https://magaliesbergbiosphere.org.za/ https://lightpollutionmap.info/ 9 Figure 4. A & B - The artificial sky brightness of Africa, with the area of my study site indicated in the light blue square. Black indicates the lowest artificial brightness, measured at < 1.74 μcd/m2, blue measures between 13.9 and 27.8 μcd/m2, yellow between 223 and 445 μcd/m2, red between 890 and 1780 μcd/m2 and pink between 3560 and 7130 μcd/m2. Image extracted from Falchi et al. 2016. C - The reported skyglow for the Cradle Nature Reserve and Johannesburg, the nearest and largest metropolitan area to the study site. This is measured using magnitude per arc second2, areas with a measurement closer to black, or 22 magnitude per arc second2, have very little exposure to skyglow and areas closer to white, or less than17.5 magnitude per arc second2, have a relatively high exposure to skyglow. https://lightpollutionmap.info (Falchi et al., 2016). 1.7 Motivation for this study As the human population grows over time, we encroach on the natural world at an ever-increasing rate. It is thus important for us to understand how this will influence the natural world and what we could potentially do to mitigate the negative consequences. With the human population increasing by approximately 1% per year (Roser et al., 2019), infrastructure development for human habitation also increases (Rotics et al., 2011). Aside from important consequences, such as habitat fragmentation and water pollution, a large concern of human expansion is ALAN (Gaynor et al., 2018; Rotics et al., 2011). A study investigating skyglow across the world in 2016, found that approximately 23% of the earth’s land surface was exposed to ALAN (Falchi et al., 2016). Another study examining ALAN over four years found a 2.2% increase in the earth’s area that was exposed to light (Kyba et al., 2017). Furthermore, some locations, including Africa, showed significant growth in artificial lighting over a limited study period (Kyba et al., 2017). These studies are some of the few available that considered all A B C https://lightpollutionmap.info/ 10 countries, but they were conducted more than five years ago, making their findings important, but possibly underestimating the severity of ALAN in the present day. Laboratory studies, which are the standard protocol for ALAN studies on rodents, frequently generate results that differ from experiments under natural conditions. Therefore, it is important to understand how animal responses to light compare in laboratory and field settings. Although more studies are being done on the effects and implications of ALAN on animal behaviour, very little has been done on wild animals in their natural habitat (Raap et al., 2015). The limited number of studies that are available are biased toward wild populations of birds. For example, great tits (Parus major) showed increased parental care when a light source was placed at the entrance of nesting boxes (Titulaer et al., 2012) and female blue tits (Cyanistes caeruleus) emerged earlier in the morning when they experienced artificial lighting (Schlicht et al., 2014). During the first study year of ALAN exposure, the great tit (P. major) and pied flycatcher (Ficedula hypoleuca) laid their eggs earlier (de Jong et al., 2015). One study on rodents showed that the nocturnal Cairo spiny mice (A. cahirinus) decreased their overall activity and foraging under ALAN in field conditions, whereas diurnal golden spiny mice (A. russatus) did not (Rotics et al., 2011). Very few studies have been conducted on the impact of ALAN on rodents in southern Africa, a taxonomic group that comprises the majority of mammals in most ecosystems in the subregion. Presently, studies on urbanisation and its impact on animal behaviour are taxonomically biased towards species that can easily disperse once they experience unsuitable conditions, such as avian species (Mazza et al., 2020). Rodents are excellent models to study the effect of urbanisation and ALAN on animal behaviour since they are easy to track, capture, house and maintain. Rodents have limited dispersal abilities that force them to remain in an altered area (Mazza et al., 2020). Furthermore, rodents are key biological constituents of ecosystems, since they play a role in distributing plant seeds and they serve as prey for multiple predators, both aerial and terrestrial (Viljoen and Oosthuizen, 2023). Drastic changes in rodent behaviour will have knock-on effects in the wider ecosystem. By understanding how urbanisation and ALAN alter rodent behaviour, we gain insight into the responses of other animals and how functional trophic levels can be modified. 1.8 Aim and Objectives The main aim of my study was to investigate the influence of artificial light at night on southern African small mammals, specifically rodents. 11 Study objectives, hypotheses, predictions: • To assess whether and how the lunar cycle and ALAN influenced small mammal community assemblages by investigating the abundance and composition of two populations at a peri-urban field site, one facing the largest metropolitan area in southern Africa (Johannesburg) and one facing away. I predicted that the lunar cycle and ALAN would negatively affect species abundance and composition, specifically that high light levels would relate to a lower abundance and composition of species. • To evaluate how the activities of a diurnal/crepuscular species, a species with variable activity and a nocturnal species differed between natural, laboratory and ALAN treatments. I expected all three rodent species selected for study would show higher overall activity during a natural treatment, which comprised of animals being confined to cages but kept in an outside enclosure where they experienced natural light and temperature fluctuations. I anticipated activity to be lower in a standard laboratory treatment (12L:12D at a constant 24°C). I predicted the lowest activity to occur under the ALAN treatment in the laboratory (12L:12D, constant 24°C and 2 Lux light during the D phase). • To investigate how the foraging behaviour of three rodent species with different activity profiles, was influenced by different environmental risk levels using ALAN. I hypothesised that the diurnal/crepuscular rodent would experience the control treatments as high risk and the ALAN treatments as normal risk. I expected the two remaining rodent species would experience a “high risk environment” under ALAN treatments and subsequently alter their foraging behaviour and reduce movement compared with the control treatments. 1.9 Layout of thesis My thesis consists of seven chapters: a general introduction (Chapter 1) followed by five data chapters (Chapters 2 – 6) and concluding with the general discussion (Chapter 7). Chapters 2 to 6 have been written as individual manuscripts intended for submission to different academic journals. Chapter 3 has been published in the Journal of Chronobiology International. There is some overlap in information presented throughout the thesis, but the overall editing of the thesis was maintained in each data chapter. The pages have been numbered sequentially across the thesis, but the figures and tables are numbered successively within each chapter and each chapter has its own reference list. All the data chapters were co-authored by both my supervisors who oversaw all aspects of the different chapters. However, I was the principal investigator who collected the data, conducted experiments, ran formal analyses, wrote up each chapter, and compiled the thesis. 12 1.10 References Barnard, C.J., Brown, C.A.J., Houston, A.I., McNamara, J.M., 1985. Risk-sensitive foraging in common shrews: an interruption model and the effects of mean and variance in reward rate. Behav. Ecol. Sociobiol. 18, 139–146. https://doi.org/10.1007/BF00299042 Bartumeus, F., Catalan, J., 2009. Optimal search behavior and classic foraging theory. J. Phys. A Math. Theor. 42, 1–12. https://doi.org/10.1088/1751-8113/42/43/434002 Bennie, J.J., Duffy, J.P., Inger, R., Gaston, K.J., 2014. Biogeography of time partitioning in mammals. Proc. Natl. Acad. Sci. 111, 13727–13732. https://doi.org/10.1073/pnas.1216063110 Benstaali, C., Mailloux, A., Bogdan, A., Auzéby, A., Touitou, Y., 2001. Circadian rhythms of body temperature and motor activity in rodents. Life Sci. 68, 2645–2656. https://doi.org/10.1016/S0024-3205(01)01081-5 Bird, B.L., Branch, L.C., Miller, D.L., 2004. Effects of coastal lighting on foraging behavior of beach mice. Conserv. Biol. 18, 1435–1439. https://doi.org/10.1111/j.1523-1739.2004.00349.x Buijs, R., van Eden, C., Goncharuk, V., Kalsbeek, A., 2003. The biological clock tunes the organs of the body: timing by hormones and the autonomic nervous system. J. Endocrinol. 177, 17–26. https://doi.org/10.1677/joe.0.1770017 Charnov, E.L., 1976. Optimal foraging, the marginal value theorem. Theor. Popul. Biol. 9, 129–136. https://doi.org/10.1016/0040-5809(76)90040-X Cho, Y., Ryu, S.-H., Lee, B.R., Kim, K.H., Lee, E., Choi, J., 2015. Effects of artificial light at night on human health: a literature review of observational and experimental studies applied to exposure assessment. Chronobiol. Int. 32, 1294–1310. https://doi.org/10.3109/07420528.2015.1073158 Clarke, J.A., 1983. Moonlight’s influence on predator/prey interactions between short-eared owls (Asio flammeus) and deermice (Peromyscus maniculatus). Behav. Ecol. Sociobiol. 13, 205– 209. https://doi.org/10.1007/BF00299924 Craft, B.B., 2016. Risk-sensitive foraging: changes in choice due to reward quality and delay. Anim. Behav. 111, 41–47. https://doi.org/10.1016/j.anbehav.2015.09.030 Dammhahn, M., Lange, P., Eccard, J.A., 2022. The landscape of fear has individual layers: an experimental test of among‐individual differences in perceived predation risk during foraging. Oikos, 1–17. https://doi.org/10.1111/oik.09124 de Jong, M., Ouyang, J.Q., Da Silva, A., van Grunsven, R.H.A., Kempenaers, B., Visser, M.E., Spoelstra, K., 2015. Effects of nocturnal illumination on life-history decisions and fitness in two wild songbird species. Philos. Trans. R. Soc. B Biol. Sci. 370, 20140128. https://doi.org/10.1098/rstb.2014.0128 Doherty, T.S., Davis, R.A., van Etten, E.J.B., 2015. A game of cat-and-mouse: microhabitat influences rodent foraging in recently burnt but not long unburnt shrublands. J. Mammal. 96, 324–331. https://doi.org/10.1093/jmammal/gyv034 Dominoni, D.M., Borniger, J.C., Nelson, R.J., 2016. Light at night, clocks and health: from humans to wild organisms. Biol. Lett. 12, 20160015. https://doi.org/http://dx.doi.org/10.1098/rsbl.2016.0015 https://doi.org/10.1007/BF00299042 https://doi.org/10.1088/1751-8113/42/43/434002 https://doi.org/10.1073/pnas.1216063110 https://doi.org/10.1016/S0024-3205(01)01081-5 https://doi.org/10.1111/j.1523-1739.2004.00349.x https://doi.org/10.1677/joe.0.1770017 https://doi.org/10.1016/0040-5809(76)90040-X https://doi.org/10.3109/07420528.2015.1073158 https://doi.org/10.1007/BF00299924 https://doi.org/10.1016/j.anbehav.2015.09.030 https://doi.org/10.1111/oik.09124 https://doi.org/10.1098/rstb.2014.0128 https://doi.org/10.1093/jmammal/gyv034 https://doi.org/http:/dx.doi.org/10.1098/rsbl.2016.0015 13 Earnest, D.J., Turek, F.W., 1982. Splitting of the circadian rhythm of activity in hamsters: effects of exposure to constant darkness and subsequent re-exposure to constant light. J. Comp. Physiol. A 145, 405–411. https://doi.org/10.1007/BF00619345 Falchi, F., Cinzano, P., Duriscoe, D., Kyba, C.C.M., Elvidge, C.D., Baugh, K., Portnov, B.A., Rybnikova, N.A., Furgoni, R., 2016. The new world atlas of artificial night sky brightness. Sci. Adv. 2, e1600377. https://doi.org/10.1126/sciadv.1600377 Fonken, L.K., Workman, J.L., Walton, J.C., Weil, Z.M., Morris, J.S., Haim, A., Nelson, R.J., 2010. Light at night increases body mass by shifting the time of food intake. Proc. Natl. Acad. Sci. 107, 18664–18669. https://doi.org/10.1073/pnas.1008734107 Gaston, K.J., Davies, T.W., Bennie, J., Hopkins, J., 2012. Reducing the ecological consequences of night-time light pollution: options and developments. J. Appl. Ecol. 49, 1256–1266. https://doi.org/10.1111/j.1365-2664.2012.02212.x Gaynor, K.M., Hojnowski, C.E., Carter, N.H., Brashares, J.S., 2018. The influence of human disturbance on wildlife nocturnality. Science. 360, 1232–1235. https://doi.org/10.1126/science.aar7121 Hänel, A., Posch, T., Ribas, S.J., Aubé, M., Duriscoe, D., Jechow, A., Kollath, Z., Lolkema, D.E., Moore, C., Schmidt, N., Spoelstra, H., Wuchterl, G., Kyba, C.C.M., 2018. Measuring night sky brightness: methods and challenges. J. Quant. Spectrosc. Radiat. Transf. 205, 278–290. https://doi.org/10.1016/j.jqsrt.2017.09.008 Hawkins, P., Golledge, H.D.R., 2018. The 9 to 5 Rodent – Time for Change? Scientific and animal welfare implications of circadian and light effects on laboratory mice and rats. J. Neurosci. Methods 300, 20–25. https://doi.org/10.1016/j.jneumeth.2017.05.014 Hoffmann, J., Schirmer, A., Eccard, J.A., 2019. Light pollution affects space use and interaction of two small mammal species irrespective of personality. BMC Ecol. 19, 1–11. https://doi.org/10.1186/s12898-019-0241-0 Ikeno, T., Weil, Z.M., Nelson, R.J., 2014. Dim light at night disrupts the short-day response in Siberian hamsters. Gen. Comp. Endocrinol. 197, 56–64. https://doi.org/10.1016/j.ygcen.2013.12.005 Jones, K.E., Bielby, J., Cardillo, M., Fritz, S.A., O’Dell, J., Orme, C.D.L., Safi, K., Sechrest, W., Boakes, E.H., Carbone, C., Connolly, C., Cutts, M.J., Foster, J.K., Grenyer, R., Habib, M., Plaster, C.A., Price, S.A., Rigby, E.A., Rist, J., Teacher, A., Bininda-Emonds, O.R.P., Gittleman, J.L., Mace, G.M., Purvis, A., 2009. PanTHERIA: a species-level database of life history, ecology, and geography of extant and recently extinct mammals. Ecology 90, 2648– 2648. https://doi.org/10.1890/08-1494.1 Kingdon, J., 2013. Mammals of Africa: Rodents, Hares and Rabbits. Bloomsbury Publishing, London. Kotler, B.P., Brown, J., Mukherjee, S., Berger-Tal, O., Bouskila, A., 2010. Moonlight avoidance in gerbils reveals a sophisticated interplay among time allocation, vigilance and state-dependent foraging. Proc. R. Soc. B Biol. Sci. 277, 1469–1474. https://doi.org/10.1098/rspb.2009.2036 Kyba, C.C.M., Kuester, T., Sánchez de Miguel, A., Baugh, K., Jechow, A., Hölker, F., Bennie, J., Elvidge, C.D., Gaston, K.J., Guanter, L., 2017. Artificially lit surface of Earth at night increasing in radiance and extent. Sci. Adv. 3, e1701528. https://doi.org/10.1126/sciadv.1701528 https://doi.org/10.1007/BF00619345 https://doi.org/10.1126/sciadv.1600377 https://doi.org/10.1073/pnas.1008734107 https://doi.org/10.1111/j.1365-2664.2012.02212.x https://doi.org/10.1126/science.aar7121 https://doi.org/10.1016/j.jqsrt.2017.09.008 https://doi.org/10.1016/j.jneumeth.2017.05.014 https://doi.org/10.1186/s12898-019-0241-0 https://doi.org/10.1016/j.ygcen.2013.12.005 https://doi.org/10.1890/08-1494.1 https://doi.org/10.1098/rspb.2009.2036 https://doi.org/10.1126/sciadv.1701528 14 Lear, K.O., Whitney, N.M., Morris, J.J., Gleiss, A.C., 2021. Temporal niche partitioning as a novel mechanism promoting co-existence of sympatric predators in marine systems. Proc. R. Soc. B Biol. Sci. 288, 20210816. https://doi.org/10.1098/rspb.2021.0816 Lima, S.L., Dill, L.M., 1990. Behavioral decisions made under the risk of predation: a review and prospectus. Can. J. Zool. 68, 619–640. https://doi.org/10.1139/z90-092 Łopucki, R., Klich, D., Kiersztyn, A., 2021. Changes in the social behavior of urban animals: more aggression or tolerance? Mamm. Biol. 101, 1–10. https://doi.org/10.1007/s42991-020-00075- 1 Mandelik, Y., Jones, M., Dayan, T., 2003. Structurally complex habitat and sensory adaptations mediate the behavioural responses of a desert rodent to an indirect cue for increased predation risk. Evol. Ecol. Res. 5, 501–515. Mazza, V., Dammhahn, M., Lösche, E., Eccard, J.A., 2020. Small mammals in the big city: behavioural adjustments of non‐commensal rodents to urban environments. Glob. Chang. Biol. 26, 6326–6337. https://doi.org/10.1111/gcb.15304 Monadjem, A., Taylor, P.J., Denys, C., Cotterill, F.P.D., 2015. Rodents of Sub-Saharan Africa: a biogeographic and taxonomic synthesis. De Gruyter, Berlin. Mucina, L., Rutherford, M.C., 2006. The vegetation of South Africa, Lesotho and Swaziland. South African National Biodiversity Institute, Pretoria. Mulder CK, Gerkema MP, Van der Zee EA (2013) Circadian clocks and memory: time-place learning. Front. Mol. Neurosci. 6, 1–10. https://doi.org/10.3389/fnmol.2013.00008 Peichl, L., 2005. Diversity of mammalian photoreceptor properties: adaptations to habitat and lifestyle? Anat. Rec. Part A Discov. Mol. Cell. Evol. Biol. 287A, 1001–1012. https://doi.org/10.1002/ar.a.20262 Perea, R., González, R., San Miguel, A., Gil, L., 2011. Moonlight and shelter cause differential seed selection and removal by rodents. Anim. Behav. 82, 717–723. https://doi.org/10.1016/j.anbehav.2011.07.001 Perrin, M.R., Ercoli, C., Dempster, E.R., 2001. The role of agonistic behaviour in the population regulation of two syntopic African grassland rodents, the striped mouse Rhabdomys pumilio (Sparrman 1784) and the multimammate mouse Mastomys natalensis (A. Smith 1834) (Mammalia Rodentia). Trop. Zool. 14, 7–29. https://doi.org/10.1080/03946975.2001.10531141 Pierce, G.J., Ollason, J.G., 1987. Eight reasons why optimal foraging theory is a complete waste of time. Oikos 49, 111-117. https://doi.org/10.2307/3565560 Preto, S., Gomes, C.C., 2019. Lighting in the workplace: recommended Illuminance (lux) at workplace environs, in: Advances in intelligent systems and computing. Springer, Verlag, pp. 180–191. https://doi.org/10.1007/978-3-319-94622-1_18 Prugh, L.R., Golden, C.D., 2014. Does moonlight increase predation risk? Meta-analysis reveals divergent responses of nocturnal mammals to lunar cycles. J. Anim. Ecol. 83, 504–514. https://doi.org/10.1111/1365-2656.12148 Pyke, G.H., Pulliam, H.R., Charnov, E.L., 1977. Optimal foraging: a selective review of theory and tests. Q. Rev. Biol. 52, 137–154. https://doi.org/10.1086/409852 https://doi.org/10.1098/rspb.2021.0816 https://doi.org/10.1139/z90-092 https://doi.org/10.1007/s42991-020-00075-1 https://doi.org/10.1007/s42991-020-00075-1 https://doi.org/10.1111/gcb.15304 https://doi.org/10.3389/fnmol.2013.00008 https://doi.org/10.1002/ar.a.20262 https://doi.org/10.1016/j.anbehav.2011.07.001 https://doi.org/10.1080/03946975.2001.10531141 https://doi.org/10.2307/3565560 https://doi.org/10.1007/978-3-319-94622-1_18 https://doi.org/10.1111/1365-2656.12148 https://doi.org/10.1086/409852 15 Raap, T., Pinxten, R., Eens, M., 2015. Light pollution disrupts sleep in free-living animals. Sci. Rep. 5, 13557. https://doi.org/10.1038/srep13557 Rajkhowa, R., 2012. Light pollution and impact of light pollution. Int. J. Sci. Res. 3, 861–867. Roll, U., Dayan, T., Kronfeld-Schor, N., 2006. On the role of phylogeny in determining activity patterns of rodents. Evol. Ecol. 20, 479–490. https://doi.org/10.1007/s10682-006-0015-y Roser, M., Ritchie, H., Ortiz-Ospina, E., 2019. World Population Growth – Our World in Data. Rotics, S., Dayan, T., Kronfeld-Schor, N., 2011. Effect of artificial night lighting on temporally partitioned spiny mice. J. Mammal. 92, 159–168. https://doi.org/10.1644/10-MAMM-A-112.1 Russart, K.L.G., Nelson, R.J., 2018. Artificial light at night alters behavior in laboratory and wild animals. J. Exp. Zool. Part A Ecol. Integr. Physiol. 329, 401–408. https://doi.org/10.1002/jez.2173 Salgado-Delgado, R., Ángeles-Castellanos, M., Buijs, M.R., Escobar, C., 2008. Internal desynchronization in a model of night-work by forced activity in rats. Neuroscience 154, 922–931. https://doi.org/10.1016/j.neuroscience.2008.03.066 Sanders, D., Frago, E., Kehoe, R., Patterson, C., Gaston, K.J., 2021. A meta-analysis of biological impacts of artificial light at night. Nat. Ecol. Evol. 5, 74–81. https://doi.org/10.1038/s41559- 020-01322-x Santos, C.D., Miranda, A.C., Granadeiro, J.P., Lourenço, P.M., Saraiva, S., Palmeirim, J.M., 2010. Effects of artificial illumination on the nocturnal foraging of waders. Acta Oecologica 36, 166–172. https://doi.org/10.1016/j.actao.2009.11.008 Schlicht, L., Valcu, M., Loës, P., Girg, A., Kempenaers, B., 2014. No relationship between female emergence time from the roosting place and extrapair paternity. Behav. Ecol. 25, 650–659. https://doi.org/10.1093/beheco/aru035 Sharma, V.K., 2003. Adaptive significance of circadian clocks. Chronobiol. Int. 20, 901–919. https://doi.org/10.1081/CBI-120026099 Shuai, L.-Y., Wang, L.-Q., Xia, Y., Xia, J.-Y., Hong, K., Wu, Y.-N., Tian, X.-Y., Zhang, F.-S., 2023. Combined effects of light pollution and vegetation height on behavior and body weight in a nocturnal rodent. Environ. Pollut. 329, 121676. https://doi.org/10.1016/j.envpol.2023.121676 Skinner, J.D., Chimimba, C.T., 2005. The Mammals of the Southern African Sub-region, third. ed. Cambridge University Press, Cape Town. Tapia-Osorio, A., Salgado-Delgado, R., Angeles-Castellanos, M., Escobar, C., 2013. Disruption of circadian rhythms due to chronic constant light leads to depressive and anxiety-like behaviors in the rat. Behav. Brain Res. 252, 1–9. https://doi.org/10.1016/j.bbr.2013.05.028 Thomas, L.S., Teich, E., Dausmann, K.H., Reher, S., Turner, J.M., 2018. Degree of urbanisation affects Eurasian red squirrel activity patterns. Hystrix, Ital. J. Mammal. 29, 175–180. https://doi.org/10.4404/HYSTRIX-00065-2018 Titulaer, M., Spoelstra, K., Lange, C.Y.M.J.G., Visser, M.E., 2012. Activity patterns during food provisioning are affected by artificial light in free living Great Tits (Parus major). PLoS One 7, e37377. https://doi.org/10.1371/journal.pone.0037377 Viljoen, A., Oosthuizen, M.K., 2023. Dim light at night affects the locomotor activity of nocturnal African pygmy mice (Mus minutoides) in an intensity-dependent manner. Proc. R. Soc. B Biol. Sci. 290, 20230526. https://doi.org/10.1098/rspb.2023.0526 https://doi.org/10.1038/srep13557 https://doi.org/10.1007/s10682-006-0015-y https://doi.org/10.1644/10-MAMM-A-112.1 https://doi.org/10.1002/jez.2173 https://doi.org/10.1016/j.neuroscience.2008.03.066 https://doi.org/10.1038/s41559-020-01322-x https://doi.org/10.1038/s41559-020-01322-x https://doi.org/10.1016/j.actao.2009.11.008 https://doi.org/10.1093/beheco/aru035 https://doi.org/10.1081/CBI-120026099 https://doi.org/10.1016/j.envpol.2023.121676 https://doi.org/10.1016/j.bbr.2013.05.028 https://doi.org/10.4404/HYSTRIX-00065-2018 https://doi.org/10.1371/journal.pone.0037377 https://doi.org/10.1098/rspb.2023.0526 16 Weaver, R.E., 2011. Effects of simulated moonlight on activity in the desert nightsnake (Hypsiglena chlorophaea). Northwest Sci. 85, 497–500. https://doi.org/10.3955/046.085.0308 Willems, J.S., Phillips, J.N., Vosbigian, R.A., Villablanca, F.X., Francis, C.D., 2021. Night lighting and anthropogenic noise alter the activity and body condition of pinyon mice (Peromyscus truei). Ecosphere 12, 1–18. https://doi.org/10.1002/ecs2.3388 Yan L, Smale L, Nunez AA (2020) Circadian and photic modulation of daily rhythms in diurnal mammals. Eur. J. Neurosci. 51, 551–566. https://doi.org/10.1111/ejn.14172 Zhang, F.-S., Wang, Y., Wu, K., Xu, W.-Y., Wu, J., Liu, J.-Y., Wang, X.-Y., Shuai, L.-Y., 2020. Effects of artificial light at night on foraging behavior and vigilance in a nocturnal rodent. Sci. Total Environ. 724, 138271. https://doi.org/10.1016/j.scitotenv.2020.138271 https://doi.org/10.3955/046.085.0308 https://doi.org/10.1002/ecs2.3388 https://doi.org/10.1111/ejn.14172 https://doi.org/10.1016/j.scitotenv.2020.138271 17 Chapter 2: The impact of artificial light at night and the lunar cycle on small mammal trapping near a large metropolitan area in South Africa Abstract Nocturnal light levels affect animal activity, for example, rodents lower their activity during full moon nights because of increased visibility and predation risk. The illumination of a full moon pales in comparison to the illumination provided by artificial light at night (ALAN) close to human habitation, which is rapidly increasing with the growing human population. I investigated how the abundance and composition of a small mammal community changes over the lunar cycle during different seasons and how this differs between a site (Light site) facing Johannesburg, the largest metropolitan area in southern Africa, and a control site (Dark site). Using baited live traps, I caught more animals on the Light site compared with the Dark site, contrary to expectations. I trapped the lowest abundance and composition during the full moon on both sites and this increased closer to the new moon when visibility decreased. Shannon and Simpson diversity indices and Pieloe’s evenness score indicated a diverse community on both sites, which differed across seasons. The Light site had a higher small mammal diversity during winter and the Dark site during autumn. There is a paucity of African-based studies that investigate the influence of the lunar cycle and ALAN on the abundance and diversity of small mammal communities. My study shows trends opposite to those expected in communities exposed to ALAN and this could be because of the comparatively low light levels or the use of microhabitats that ameliorate the effects of ALAN. Keywords: abundance, ALAN, anthropogenic disturbance, lunar cycle, small mammals, South Africa 2.1 Introduction As the human population continues to grow exponentially, the demand for housing and development increases at a rapid pace. This causes the transformation and depletion of natural spaces that force wild animals to seek other suitable habitats (Sol et al., 2013). However, the remaining natural areas are impacted by another growing challenge, i.e., artificial light at night (ALAN) (Finch et al., 2020). Falchi et al. (2016) estimated that 23% of the earth’s surface was already exposed to ALAN and this percentage has definitely increased since their study was published. Animals use light as a source of temporal information (Benstaali et al., 2001) and the increased levels of ALAN results in changes that were not 18 anticipated or cannot be mitigated currently. These changes include shifts in spatial use patterns of the affected individuals, altered interspecific interactions and foraging behaviours etc. We can investigate the responses of prey animals to ALAN by comparing changes in their behaviour during dark and full moon. A full moon provides the highest illumination in the lunar cycle, which is approximately 2 Lux (Penteriani et al., 2013). In the few days around full moon, prey species decrease their activity to avoid predator detection because of the greater visibility (Kotler et al., 2010; Kronfeld-Schor et al., 2013; Prugh and Golden, 2014). For example, rabbits (Oryctolagus cuniculus) travelled longer distances and used simpler movement patterns as their main predator avoidance strategy during new moon and became less active and employed more complex moves to avoid predator detection during full moon nights (Penteriani et al., 2013). Prey animals thus modify behaviour to minimise the risk of exposure and detection by predators during nights with high visibility. Animal activity patterns are regulated by circadian rhythms, which are primarily influenced by light (Benstaali et al., 2001). Artificial and unnatural light patterns can result in the desynchronisation of activities, meaning that animals could mistime activities due to altered light cues or miss potential mating encounters due to different individual temporal patterns (Gaston et al., 2012). Moreover, desynchronisation could alter the behaviour of competing species that utilise separate temporal niches and impact species interactions (Hoffmann et al., 2018), and ultimately coexistence. Bank voles (Myodes glareolus), which altered their activity and feeding times because of ALAN, showed temporal overlap in activity times with wood mice (Apodemus sylvaticus), increasing competition because of interspecific encounters (Hoffmann et al., 2018). Competition is not the only interspecific interaction that changes under ALAN. Since rodents are a prey source for both terrestrial and aerial predators (Gutman et al., 2011), ALAN can cause a permanent state of high predation risk, thus impacting their foraging and other activities (Kotler et al., 2010). For example, the Mongolian five-toed jerboa (Allactaga sibirica) spent less time in artificially illuminated patches and had an overall reduced food intake despite being more efficient at finding food under light (Zhang et al., 2020). Animals can therefore experience reduced fitness under ALAN, if they have to employ the behavioural adaptations they use under high intensity light in the lunar cycle (Zhang et al., 2020). Within a small mammal community, there are a multitude of natural factors influencing the population dynamics of individual species. Biotic factors, such as intra- and interspecific competitors and predators, and abiotic factors, such as climatic factors and the lunar cycle, act in tandem to create the spatial and temporal environment of an animal (Pratas‐Santiago et al., 2017). The combination of these factors creates a complex system and teasing apart the components of these systems is necessary to understand the larger processes and dependencies in an ecosystem (Radchuk et al., 2016). For example, the amount of available food is dependent on rainfall, and during periods of low rainfall, there could be low food availability, resulting in competition for resources (Shilereyo et al., 2023). Moreover, the 19 amount of rainfall influences the thickness of vegetation cover, which could impact how small mammals move on the ground. Foraging success is also affected by the lunar cycle. Predation risk is higher during full moon nights (Kotler et al., 2010). However, depending on levels of satiety, prey species could risk being exposed if they need to forage to meet their energy needs or reduce foraging if they are not energetically compromised (Bedoya-Perez et al., 2013). My study aimed to investigate how the abundance and composition of small mammals on two field sites differed over the lunar cycle, seasonally and under the influence of ALAN. This study is important because little is known about how African small mammals respond to changes in moon illumination and to artificial light at night. I conducted small mammal trapping surveys in two areas at a peri-urban field site, outside Johannesburg, South Africa. One site faced the city of Johannesburg (Light site), and the other faced away from the city (Dark site). I had three predictions. 1) Since the Light site received a greater level of ALAN, I predicted that I would trap fewer species and lower abundance of animals on this site, as the predation risk is presumably higher because of the increased visibility. I did not necessarily expect that trapping fewer animals would be related to more animals being preyed upon, but that the animals would be more cautious and would not enter traps as readily as in areas with lower visibility. I expected to trap more animals and species on the Dark site, since there were fewer days with high illumination. 2) Population fluctuations differ seasonally based on the number of available resources. Thus, I established how the abundance and composition of small mammals varied seasonally. I predicted the highest small mammal abundance in spring and summer during the breeding season for most small mammals, and lower numbers during autumn and winter because of the declining number of resources. 3) Since the moon phase is associated with levels of predation risk, I also studied how the abundance and composition changed across the lunar cycle. I predicted that greater moon illumination (days around a full moon) would result in lower trapping success, whereas reduced moon illumination (days around a new moon) would show a higher trapping success. I expected this trend to be present on both trapping sites. 2.2 Materials and Methods 2.2.1 Trapping sites I conducted my research on the Cradle Nature Reserve (-25.9214, 27.8503) located within the Magaliesberg Biosphere in South Africa. This area consists of both savanna and grassland biomes with some woody vegetation and herbaceous grasses (Mucina and Rutherford, 2006; Ramahlo et al., 2022). Two trapping sites were identified. The sites were visually inspected, and I established that they were of similar slope and elevation (Appendix Figures 1 and 2). The first site (Light site) faced Johannesburg, which is the largest metropolitan area in southern Africa with high levels of ALAN and the second site 20 (Dark site) faced away from Johannesburg (Figure 1). These two sites were approximately 1.7km apart and very few ALAN sources were present on the property itself with the majority of the sources being located at quite a distance and buffered with densely wooded areas. Figure 1. The locations of the Light and Dark sites on the Cradle Nature Reserve property. The Light site faced Johannesburg and the Dark site faced the opposite direction. 2.2.2 Animal trapping and measurements I trapped small mammals in four seasons within one year (Table 1), using 75 PVC live animal traps per site (7.5 x 7.5 x 30cm). A permanent grid was set up on each site to ensure traps were placed in approximately the same position during each trapping session and the traps were covered with loose vegetation to aid in insulation and mimic a more natural look. All traps per site were separated by 10m and placed in five rows of 15 traps each. The traps were baited with a mixture of sunflower seeds, sunflower oil, rolled oats, granola and salt. Traps were opened at approximately 17h00 every day and checked the following morning at sunrise (between 05h00 and 06h30 depending on the season). Traps were closed during the day since I was interested in the activity of nocturnal rodents to assess the impact of light at night on this guild. 21 Table 1. The seasonal sampling period dates for the Dark and Light sites on the Cradle Nature Reserve. Seasons Start date End date Autumm 10/05/2022 10/06/2022 Winter 03/08/2022 03/09/2022 Spring 11/11/2022 12/12/2022 Summer 30/01/2023 02/03/2023 Trapping lasted for a total of 32 continuous days per season (Table 1) to account for the varying illumination levels throughout a full lunar cycle. During winter, cotton wool was placed in the traps to provide some warmth for the occupants during cold nights. When I confirmed that the trap contained a small mammal, I emptied the contents of the trap into a transparent plastic Ziploc® freezer bag to identify the animal to species level where possible. Three cryptic species that were not morphologically identifiable were later identified through mtDNA cyt b sequencing using tissue from trapped individuals, obtained by cutting a small piece of the animal’s external ear. I weighed the animal using a hanging scale (Pescola®, Switzerland, 1g precision). I sexed each animal using their anogenital distance (longer in males than females). Thereafter, the reproductive status of each animal was recorded; males were classified as either scrotal or non-scrotal and females were classified as pregnant when the abdomen was swollen, lactating when milk could be expressed from the nipples and if neither was observed I considered them non-reproductive (White and Geluso, 2012). Each animal was fitted with a pair of unique ear tags (National Band & Tag Company, USA), which allowed me to identify re-captured individuals. The animal was then released at the site of capture. A trapping permit was approved by the Gauteng Department of Agriculture and Rural Development (CPF6-0231) and the University of Witwatersrand Animal Research Ethics committee gave ethical clearance for this study (2021/08/09B). To assess the influence of the lunar cycle on the trapping, I downloaded the moon illumination data from the Time and Date website (https://www.timeanddate.com/moon/south- africa/johannesburg?month=3&year=2023, accessed April 2023). The moon illumination percentages (moon phase per day) used in the analyses were retrieved from this website, it was calculated at lunar noon and took refraction into account. In addition, weather data from Lanseria, Johannesburg was accessed through the VisualCrossing website (https://www.visualcrossing.com/weather/weather-data- services#, accessed April 2023). The downloaded weather variables included the minimum temperature, wind speed, cloud cover, humidity and precipitation per trapping day. https://www.timeanddate.com/moon/south-africa/johannesburg?month=3&year=2023 https://www.timeanddate.com/moon/south-africa/johannesburg?month=3&year=2023 https://www.visualcrossing.com/weather/weather-data-services https://www.visualcrossing.com/weather/weather-data-services 22 2.2.3 Data analyses In winter, the Dark site was burnt down completely after only 10 days of trapping because of a runaway fire, thus I did not use winter data for the Dark site. Species richness was calculated within site and season, using the count of species per site and per season (Table 2). All further data analyses were done using the R software (R v4.2.1, Boston, United States). I calculated the diversity of the small mammals by site and season using the Shannon and Simpson diversity indices and Pielou’s evenness index in the “vegan” package (Oksanen et al., 2022). To analyse whether the indices differed between sites and seasons, I ran a Kruskal-Wallis test for the Simpson index, since it was not normally distributed (Shapiro-Wilk test: P < 0.05) and t-tests for the Shannon and Pielou indices, as they were normally distributed (Shapiro-Wilk test: P > 0.05). In order to assess the homogeneity of populations across sites and seasons, I used the betadisper function to determine the Bray-Curtis dissimilarity score, using means per site per season (Oksanen et al., 2022). I used linear models to analyse which factors influenced the abundance and composition of the populations. Both response variables (abundance of animals caught - including new and recaptured animals, and the composition - the number of different species caught per day) were tested for normality using the Shapiro-Wilk test and were non-parametric (P < 0.05). The abundance of animals and composition of species were considered daily to coincide with the changes in moon illumination. I analysed the data using generalized linear models (GLMs), with a Poisson distribution and log link function. Predictor variables included the site, season and moon illumination as percentage illumination. To obtain the most parsimonious model per response variable, I used the drop1 function to remove non-significant variables in a stepwise manner. All model versions were then compared using the “MuMIn” package (Barton, 2023), and the model with the highest weight and lowest AICc was used for all reported results per response variable (Appendix Table 2). Post-hoc comparisons were completed for all significant categorical variables using the “emmeans” package (Lenth et al., 2020). If moon illumination significantly influenced the response variables, it was further analysed using Spearman correlations. These correlations were run using the moon illumination percentage with the response variables, namely abundance of animals caught and composition of species. All tests were two-tailed, and model significance set at 0.05. Weather variables including minimum temperature, wind speed, cloud cover, humidity and precipitation, were included in the preliminary analysis using the residuals obtained from a principal component analysis (PCA). The PCA was completed using the “FactoMineR” (Husson et al., 2023) and “factoextra” (Kassambara and Mundt, 2022) packages in R. Five different principal components (PCs) were extracted (Appendix Table 3) and the PC explaining more than 50% of the variance was considered in the GLMs. However, the drop1 function excluded PC1 from all GLMs, and thus the PC values were excluded from all further analysis. 23 2.3 Results The models for the statistical analyses considered all predictor variables and their interactions. The final models used depended on whether some variables were retained in the analyses (Appendix Table 2). For ease of explaining the statistical analyses, I presented the outcomes for predictor variables separately below. 2.3.1 Trapping success During the 128 trapping days (38 400 trap nights), a total of 396 small mammals were caught of which 72.2% were recaptured individuals (Figure 2). The highest number of new captures was during autumn on the Light site and during spring and summer on the Dark site (Table 2). The highest number of rec