Search

Prey switching is an important part of predator feeding and breeding ecology since prey populations and therefore prey availability is dynamic. Deliveries of prey made by the Eurasian kestrel (Falco tinnunculus) breeding in a nest box in southern Norway, were recorded with a camera. Temporal activity of ground dwelling prey was recorded using five Reconyx cameras mounted in tunnel shaped boxes. Both were analysed in terms of temporal overlap. Total observations made of nest deliveries and camera traps, respectively were: 383 and 80 of Shrews (Sorex spp.), 190 and 6 of lemmings (Lemmus lemmus), 102 and 53 of the bank vole (Myodes glareolus), 60 and 64 Microtus voles (Microtus spp.) and 20 deliveries of birds to the nest. Shrews, Microtus voles and the bank vole were defined as the main prey and birds as an alternative prey. The main prey species had similar, but varying temporal patterns of activity and deliveries. The kestrel increased deliveries of birds (their alternative prey) when deliveries of main prey were decreasing in the hours between 11:00 and 18:00, the decrease in main prey deliveries around this time was small, and bird deliveries in these hours were probably an additive energy source. The kestrels deliveries of alternative prey did therefore follow a consistent hourly trend. Our results are similar with other findings made of Norwegian breeding kestrels by Steen et al., (2011a), where the same patterns of alternative prey deliveries occurred, around the same timeframe as our results, but with another alternative prey, the viviparous lizards (Zootoca vivipara). This consistent temporal change in prey composition, where alternative prey was commonly delivered for a brief period, should increase the prey mass delivered to the nestlings, and create a more stable feeding pattern throughout the day. The switch to an increase of bird being delivered at the nest were a discreet, but rapid form of prey switching. This thesis is to my knowledge the only study using this methodology to determine hourly changes in prey composition for a predator. M-NF

Prey deliveries and feeding time was recorded and analyzed at two osprey (Pandion haliaetus) nests located in Østfold county, Norway during the breeding season of 2016. Feeding time analysis was only done when the nestlings were assisted by a parent. 201 prey deliveries were used for feeding time analysis. I found that carp species were the most important prey at the osprey nest in Isnes, while flounders were most important at the osprey nest in Skjeberg, and that this discrepancy was likely due to the difference in proximity to freshwater and saltwater fishing localities. 60% of prey items were delivered decapitated, and the frequency of decapitation was higher for carps and flounders, and low for perches and pikes. At Isnes, the probability of decapitation was affected by prey weight, while none of the explanatory variables affected probability of decapitation at the Skjeberg nest. Nestlings were just as likely to feed assisted as unassisted when they reached 50 days of age. The feeding parent spent an average of 1,678 s feeding a prey item to the nestlings, and there was a strong positive relationship between prey mass and feeding time. Decapitation reduced feeding time in all prey except for large carps at the Isnes nest, while the male at Skjeberg spent less time feeding a prey item to the nestlings than the female. The female at Isnes assisted the nestlings in 98% of feeding sessions, while the male delivered most of the prey. This supports the general conception that ospreys exhibit strongly divided parental care tasks. However, when the female at the Skjeberg nest disappeared prior to the nestlings being able to feed unassisted, the male continued to feed the nestlings, showing that these divided roles have some degree of plasticity. The female at Isnes continued feeding the nestlings although they appeared to be able to feed unassisted, suggesting a strategy of maximizing the control over allocation of resources delivered by the male between her and the nestlings. This suggests that a sexual conflict may be present in ospreys during the nesting period. Byttedyrleveringer og spisetid på to fiskeørnreir (Pandion haliaetus) i Østfold fylke, Norge, ble videoovervåket gjennom hekkesesongen i 2016. Analyse av spisetid ble kun gjort for byttedyr der ungene ble matet av en forelder. Totalt ble 379 byttedyr registrert, og spisetid ble analysert for 201 av disse. Karpefisk ble oftest levert på reiret i Isnes, mens flyndrefisk oftest ble levert på reiret i Skjeberg, og denne forskjellen i byttedyrvalg synes å ha sammenheng med reirets avstand til ferskvanns- og saltvannsfiskelokaliteter. Av byttedyrene som ble levert var 60% dekapiterte, og karpefisk og flyndrer var oftere dekapitert enn abbor og gjeddefisk. På Isnes økte sannsynligheten for dekapitering med fiskevekt, mens ingen av forklaringsvariablene påvirket sannsynlighet for dekapitering på Skjeberg. Det var like stor sannsynlighet for at ungene spiste selv som at de ble matet da de var 50 dager gamle. Forelderen som matet brukte i gjennomsnitt 1678 s på å mate ungene med ett byttedyr, og spisetiden økte med vekten på byttedyret. Dekapitering reduserte spisetiden for alle byttedyr med unntak av store karpefisk på reiret i Isnes, mens hannen på Skjeberg-reiret brukte kortere tid på å mate ungene med et bytte enn hunnen. Hunnen på Isnes-reiret matet ungene i 98% av måltidene, i mens hannen leverte mesteparten av byttedyrene. Dette støtter oppfatningen om at fiskeørn har klart adskilte foreldreoppgaver. På Skjeberg derimot, forsvant hunnen før ungene var i stand til å spise selv, noe som førte til at hannen begynte å mate ungene. Dette viser at de delte foreldrerollene hos fiskeørn til dels er plastiske. Hunnen på Isnes fortsatte å mate ungene selv om de var i stand til å spise selv, noe som antyder en strategi der hunnen maksimerer sin kontroll over fordelingen av mat levert av hannen mellom henne og ungene. Dette kan tyde på en seksuell konflikt mellom foreldrene hos fiskeørn i hekkeperioden. M-ECOL

A breeding population of marsh harriers (Circus aeruginosus) has established in SE Norway during the last decades, and the population is increasing. Yet, no study of the breeding ecology of this population has been conducted. I studied the diet, diel pattern of deliveries, and prey handling behavior of two nesting pairs of marsh harriers in SE Norway by video monitoring at the nests. A total of 1004 prey deliveries were recorded, and 948 of these were analyzed. All prey were vertebrates; 53.8 % were birds, 42.2 % were mammals, and the remaining 1.1 % of the identified prey were amphibians, fish and reptiles. The most numerous prey type was passerine birds (Passeriformes). Although these accounted for 51.3 % of the prey by number, they accounted for no more than 28.0 % of the mass delivered at the nests. The most important prey species, in terms of mass delivered at the nests, was the European water vole (Arvicola amphibius), which accounted for 24.5 % of the mass delivered. The probability of delivering a prey item in an hour block varied little during the daily activity period, but the probability of delivering an avian prey was highest in the afternoon, while the probability of delivering a mammalian prey was highest in the morning with an additional peak in the evening. Average prey body mass of the prey captured was 65.3 g, 87.0 g for mammalian prey and 48.0 g for avian prey. The male captured more passerines and smaller prey than the female, which could be linked to smaller size of the male than the female. The probability that a prey was decapitated prior to delivery at the nest increased with prey body mass and was higher for mammalian prey than for avian prey. For mammals, the probability that a prey was decapitated was affected by the age of the nestlings and which parent had captured the prey. For prey captured by the female, the probability that it was decapitated prior to delivery increased with nestling age, while opposite was the case for prey captured by the male. This is probably an effect of the changing hunting pressure when the female starts assisting the male in hunting in the latter part of the nestling period. The handling time increased with increasing prey body mass for both mammals and birds, but this increase was faster for mammalian prey, both when the female fed the nestlings and when the nestlings ingested prey unassisted. As the marsh harrier show functional response to prey, data from more nests and from more years, are needed to verify the trends found in my study. En bestand med hekkende sivhauk (Circus aeruginosus) har de siste tiårene etablert seg i Sørøst-Norge, og populasjonen øker stadig. Likevel har ingen studier blitt gjennomført på hekkebiologien til denne populasjonen. Jeg har analysert dietten, døgnmønster for byttedyrleveringer og behandling av byttedyr for to hekkende sivhaukpar i Sørøst-Norge gjennom videoovervåking av reirene. Av 1004 byttedyr som ble levert på reirene ble 948 analysert. Alle byttedyrene var virveldyr; 53.8 % var fugler, 42.2 % var pattedyr, mens de resterende 1.1 % av de identifiserte byttedyrene var amfibier, fisk og reptiler. Spurvefugler (Passeriformes) var den mest tallrike byttedyrgruppen. Selv om disse utgjorde 51.3 % av antallet byttedyr, utgjorde de ikke mer enn 28.0 % av den totale biomassen levert på reirene. Det viktigste byttedyret, i form av biomasse, var vånd (Arvicola amphibius), som utgjorde 24.5 % av all biomasse levert på reirene. Sannsynligheten for å levere et byttedyr til reiret varierte lite gjennom den aktive perioden av døgnet, men sannsynligheten for å levere en fugl var størst om kvelden, mens sannsynligheten for å levere et pattedyr var høyest på formiddagen og på kvelden. Gjennomsnittlig byttedyrvekt var 65.3 g, 87.0 g for pattedyr og 48.0 g for fugler. Hannen fanget mer spurvefugl og mindre bytter enn hunnen, som kan linkes opp mot at hannen er mindre enn hunnen. Sannsynligheten for at et bytte ble dekapitert før levering på reiret økte med byttedyrvekt og var høyrere for pattedyr enn for fugler. For pattedyr var sannsynligheten for at et bytte var dekapitert avhengig av alderen på ungene og hvilken av foreldrene som leverte byttet. For pattedyr fanget av hunnen økte sannsynligheten for at byttet var dekapitert med ungenes alder, mens det motsatte var tilfellet for pattedyr fanget av hannen. Dette er trolig en effekt av et endret jaktpress når hunnen begynner å assistere hannen i jakten mot slutten av hekkeperioden. Håndteringstiden økte med byttedyrvekt for både pattedyr og fugler, men økte raskere for pattedyr, både når hunnen matet ungene og når ungene spiste på egenhånd. Siden sivhauken viser funksjonell respons til endret byttedyrtetthet trengs det data fra flere reir og over flere år for å kunne verifisere funnene i mitt studie. M-ECOL

Through continuous video monitoring of three tawny owls (Strix aluco) nests in Akershus county in southeast Norway during April and May 2016. I recorded prey deliveries and prey handling. A total of 284 prey items were registered, adding up to a total of 6795.7 g of prey mass delivered with a mean prey mass of 24.4 g. The prey mass ranged from 0.2-300 g. All prey items were identified to species or higher taxa. Vertebrates accounted for 89.4 % of all prey items by number and 99.2 % by mass, invertebrates accounted for 9.9 % by number and 0.1 % by mass, while unidentified prey accounted for 0.7 % by number and 0.7 % by mass. Mammals were the largest prey group accounting for 69.4 % by number and 78.0 % by mass, and within mammals, wood mouse (Apodemus sylvaticus) contributed the most with 31.0 % by number and 32.4 % by mass. Birds were the other large group accounting for 16.5 % by number and 17.2 % by mass. Other prey deliveries were amphibians (2.8 % by number and 3.0 % by mass) and reptiles (0.7 % by number and 1.0 % by mass). Most of the prey deliveries occurred during the night, and the highest probability (0.38) for a prey delivery happened during the hour block lasting from 22.00 to 22.59. The nestling’s need for assistant feeding increased with prey mass and decreased as the nestlings grew older. The probability that mammals were ingested unassisted at least 50 % of the time occurred after 21 days and after 25 days for birds. The probability that any prey item was ingested unassisted at least 50 % of the time occurred after 20 days. The prey mass consumed per nestling increased with nestling age in a non-linear model and peaked at 24 days, while the prey mass consumed per nestling increased with the season (days after monitoring started at the first nest) in a linear model. Ved kontinuerlig videoobservasjon på tre reir av kattugle (Strix aluco) i Akershus fylke i Sørøst-Norge i april og mai 2016, registrerte jeg byttedyrleveringer og byttedyrhåndtering. Totalt ble 284 bytter registrert, noe som gav 6795.7 g levert byttedyrmasse med en gjennomsnittsvekt på 24.4 g. Vekten på byttedyrene varierte fra 0.2-300 g. Alle byttedyr ble identifisert ned til art eller høyere klassifisering. Vertebrater utgjorde 89.4 % av alle byttedyrene i antall og 99.2 % i vekt, evertebrater utgjorde 9.9 % av alle byttedyrene i antall og 0.1 % i vekt, mens uidentifiserte byttedyr utgjorde 0.7 % av alle byttedyrene i antall og 0.7 % i vekt. Pattedyr var den den største gruppen av byttedyr med 69.4 % av alle byttedyrene i antall og 78.0 % i vekt, og av pattedyr så var den vanligste arten skogmus (Apodemus sylvaticus) som stod for 31.0 % av antall byttedyr og utgjorde 32.4 % av total byttedyrmasse. Fugler var den andre store gruppa, og denne utgjorde 16.5 % av antall byttedyr og 17.2 % av total byttedyrmasse. Andre byttedyr som ble levert var amfibier (2.8 % i antall og 3.0 % i masse) og reptiler (0.7 % i antall og 1.0 % i masse). De fleste byttedyrleveringene forekom i løpet av natten, og den høyeste sannsynligheten (0.38) for at det forekom en byttedyrlevering var i timesblokken fra 22.00 til 22.59. Ungenes behov for fôringshjelp økte med byttedyrmasse og minket med ungens alder. Sannsynligheten for at et pattedyr ble spist uten hjelp i minst 50 % av gangene forekom etter 21 dager og for fugler etter 25 dager. Sannsynligheten for at et hvilket som helst byttedyr ble spist uten hjelp i minst 50 % av gangene forkom etter 20 dager. Byttedyrmasse konsumert per unge økte med ungenes alder i den ikke-lineære modellen og hadde en maksimumsverdi ved 24 dager, mens byttedyrmasse konsumert per unge økte med sesong (dvs. antall dager som hadde gått etter at videoobservasjonsperioden hadde startet) i den lineære modellen. M-ECOL

The diet, diel pattern of prey deliveries, and prey handling behaviour of a nesting pair of the peregrine falcon (Falco peregrinus) in southeast Norway was studied with video recording equipment at the nest and direct observations of parental behaviour and interactions in the vicinity. The study was conducted from the single nestling was 10 days old until it fledged. A total of 88 prey items were recorded delivered, all of which were birds. The most important prey type were thrushes (Turdus sp.), by both numbers (47.7%) and net body mass delivered (57.1%). Small and larger passerines pooled was the next most important prey type by number (39.8%), but contributed little (9.8%) by net body mass. Woodpigeons (Columba palumbus) made up only 6.8% by numbers and was the next most important prey type by net body mass delivered (29.2%). The probability of delivering a prey item at the nest was highest during morning and evening, and lowest around midday. The probability of delivering a thrush rather than a small passerine was low in the morning and the evening, and high near midday. The female was recorded to deliver the majority of prey (95%) at the nest, whereas the male started to deliver prey items directly at the nest from when the nestling was 30 days old. Outside the nest, I observed the male was delivering some prey items to the female, whereas the female dismembered the prey and fed the nestling. The probability that a prey item had been plucked prior to delivery at the nest decreased with increasing prey body mass when the nestling fed unassisted, but not when the female fed the nestling. The probability that a prey item was decapitated before delivery was lower if the prey item was a thrush than when it was another kind of bird, and decreased with nestling age. The nestling required maternal assistance in feeding until it was 29 days old. Handling time increased with body mass of the prey, both when the female assisted in feeding the nestling and when the nestling fed unassisted. I was unable to decide whether some of the prey items delivered at the nest had been retrieved from caches, so the time of delivery was not necessarily shortly after capture for all prey. Future studies should include this aspect by the use of a thermal-imagery camera in addition to regular video camera. Dietten, døgnmønster for byttedyrleveringer, og behandling av byttedyr for et hekkende par av vandrefalk (Falco peregrinus) ble studert med videoopptak på reiret og direkteobservasjoner av atferden til og interaksjonene mellom foreldrene i nærheten av reirområdet. Studiet ble gjennomført i sørøst-Norge fra da den eneste ungen i reiret var 10 dager gammel til den ble flygedyktig. Totalt 88 byttedyr ble filmet levert, hvorav alle var fugler. Den viktigste byttetypen var troster (Turdus sp.), både i antall (47.7%) og i nettovekt levert (57.1%). Småfugl var den nest viktigste byttetypen i antall (39.8%), men bidro lite (9.8%) i nettovekt. Ringdue (Columba palumbus) utgjorde kun 6.8% i antall og var den nest viktigste byttetypen i nettovekt (29.2%). Sannsynligheten for å levere et byttedyr til reiret var høyest på morgenen og kvelden, og lavest midt på dagen. Sannsynligheten for å levere en trost i stedet for en småfugl var lav på morgenen og kvelden, og høy midt på dagen. Hunnen leverte, ut fra opptakene, flesteparten av byttedyrene (95%) til reiret, mens hannen begynte å levere byttedyr direkte til reiret fra da ungen var 30 dager gammel. Utenfor reiret observerte jeg at hannen allokerte byttedyr til hunnen, mens hunnen preparerte byttedyret og matet ungen. Sannsynligheten for at et byttedyr var ribbet før levering til reiret minket med økende byttevekt når ungen spiste selv, men ikke når hunnen matet ungen. Sannsynligheten for at et byttedyr ble dekapitert før levering var lavere for troster enn for andre typer fugl, og minket med ungens alder. Ungen behøvde å bli matet av hunnen frem til den var 29 dager gammel. Håndteringstid økte med byttevekt, både når hunnen matet ungen og når ungen spiste selv. Jeg kunne ikke finne ut om noen av de leverte byttedyrene hadde blitt hentet fra et lagringssted, og derfor var ikke leveringstidspunktet nødvendigvis alltid like etter at byttedyret ble fanget. Fremtidige studier burde inkludere dette aspektet ved å bruke et termisk kamera som supplement til videokamera. M-ECOL

To record prey delivery and handling, and nest of the eagle owl (Bubo bubo), a breeding pair was video monitored during 6 June – 25 July in Meland municipality in Hordaland County, Norway. A total of 49 prey items were recorded delivered. Of those, mammals amounted 53%, birds 36%, og amphibians 2% by frequency, while 8% were not possible to identify to any taxa. Hedgehogs (Erinaceus europaeus) were the most frequent prey delivered, both by frequency (39) and by biomass (68%), indicating that the pair had specialized in hunting hedgehogs. Tufted duck (Aythya fuligula) was the most numerous avian species, with 3 deliveries, while brown rat (Rattus norvegicus) was delivered twice. All prey items were delivered within 10 hours from solar midnight, defined as the darkest point throughout the night, and the majority within 3 hours from solar midnight. This strengthens the perception of the eagle owl as a nocturnal predator. Hedgehogs were to a greater extent than other prey delivered during dusk and dawn, whereas birds and rodents were delivered mainly at night. The probability that a delivered prey item was a hedgehog decreased as precipitation increased. The probability that a delivered prey item was a bird increased as nestling age increased. The probability that a delivered prey item was a duck increased as deviation from solar midnight increased. The probability that the eagle owl delivered an item of the same prey type as the previous one delivered increased with increasing deviation from solar midnight. The probability that the eagle owl delivered an item of the same species as the previous one delivered decreased as the time since previous delivery increased. Birds were the only prey that was decapitated prior to delivery, and 73% of avian prey was delivered decapitated. The probability of a bird being decapitated prior to delivery increased as its body mass increased. The probability that the nestlings fed unassisted increased as their age increased, and decreased as prey mass increased. As more traditional diet analyses based on pellets and prey remains has proved to be biased, video monitoring proves to be a useful tool when mapping diet and breeding biology in raptors. This in turn can be implemented in the preserving of birds of prey, because identification of their food sources is essential from conservation considerations. However, only two eagle owl nests has previously been video monitored for this purpose, thus a currently small basis of data makes conclusions daring for now. For å undersøke byttedyrsseleksjon og byttedyrshåndtering ble et hekkende par hubro (Bubo bubo) videofilmet 6 juni – 25 juli i Meland kommune i Hordaland. Totalt ble det levert 49 byttedyr på reiret. Av disse utgjorde pattedyr 53%, fugl 36%, og amfibier 2%. 8.2% av leverte byttedyr lot seg ikke identifisere til noe taksonomisk nivå. Piggsvin (Erinaceus europaeus) var det hyppigst leverte byttedyret både av frekvens (39%) og i biomasse (68%). Uten sammeligning var piggsvinet det viktigste pattedyret, og det er naturlig å anta at paret hadde spesialisert seg i jakt på piggsvin. Toppand (Aythya fuligula), levert 3 ganger var den mest hyppige arten av fugl levert på reiret, mens brunrotte (Rattus norvegicus) ble levert 2 ganger. Alle byttedyr ble levert innen et tidsom på 10 timer fra solar midnatt, definert som døgnets mørkeste tidspunkt, majoriteten av disse innen 3 timer. Dette styrker oppfatningen av hubro som et nattaktivt rovdyr. Piggsvin ble i større grad enn andre byttedyr levert i skumringen, mens fugl og gnagere i hovedsak ble levert om natten. Sannsynligheten for at et levert byttedyr var et pinnsvin minket med økende nedbør. Sannsynligheten for at et levert byttedyr var en fugl steg ettersom ungenes alder økte. Sannsynligheten for at et levert byttedyr var en and økte med økt avstand fra solar midnatt. Videre økte sannsynligheten for at et levert byttedyr var av samme dyreklasse som forrige leverte bytte med økt avstand fra solar midnatt, mens sannsynligheten for at samme art ble levert minket ettersom tiden siden forrige levering økte. Fugler var den eneste dyreklassen som var dekapitert før levering, og 73% av leverte fugler var dekapitert. Sannsynligheten for at en fugl var dekapitert før levering økte ettersom vekten hos fuglen økte. Sannsynligheten for at ungene spiste uassistert av byttedyrene økte ettersom de ble eldre men minket ettesom størrelsen på byttedyret økte. Mer tradisjonelle diettanalyser, f.eks gulpebolleanalyser, har vist seg å kunne være feilaktig og er høyst avhengig av kvaliteten hos observatører. Derimot har videoovervåkning vist seg som et godt verktøy for nøye kartlegging av rovfuglers diett og atferd. I sin tur kan dette implimenteres i forvaltningen av artene, ettersom identifisering av fuglenes byttedyr er essensielt i bevaringshensyn. For hubro er dette gjort bare to ganger tidligere, og følgelig bør resultatene foreløbig behandles med forsiktighet. M-NF

I studied home range and area use of the female Eurasian kestrel (Falco tinnunculus), in the boreal forested area of Trysil municipality (61°00' - 61°41'N, 11°44' - 12°52'E) in the breeding season of 2011, a peak vole year. Six females were tagged and radio-tracked using VHF telemetry in the pre-fledging period, and two of these females were additionally followed in the early post-fledge period. Females were located between 49 and 141 times in the pre-fledging period ( χ = 98 ± 13 SE), totaling 589 observations, which allowed the investigation of area use questions using the Minimum Convex Polygon (MCP) and fixed kernel href methods. I found a large variation in home range size between females as measured by the overall 100% MCP (range 0.69 - 6.16 km2, median = 0.94 km2, χ = 1.84 ± 0.87 SE km2), and in the levels to which each home range was centered on the nest. Home range size was significantly positively correlated with the proportion of bog with forest (overall 95% kernel) and the proportion of forest of developmental stage 3 present (overall 95% kernel, and overall 100% MCP), and significantly negatively correlated with proportion of pooled developmental stage 4 and 5 (overall 95% kernel). A trend approaching significance was found for negative correlation with Microtus vole abundance (overall 95% kernel). A higher probability of being in the core foraging area (foraging 50% kernel) was associated with decreased distance to nest and decreased distance to bog, highlighting the importance of maintaining proximity to the nest site for the kestrel as a central-place forager, and the importance of proximity to open habitat. In addition, post-fledge tracking highlighted the importance of the nest in the early fledging period. My results indicate the importance of taking into account vegetation structure and how it relates to prey availability, when looking at the effect on home range size. I further suggest that the role of modern forestry practices should be investigated, with regards to the effects habitat fragmentation and patch size and shape may play on spatial and temporal area use by the female kestrel in boreal forest. 2018-08-15 M-NF