Verhaltensforschung Uni Bielefeld

 Department of Animal Behaviour, University of Bielefeld, Germany


Deutsche Seiten

 Dirk Petzold, Diplombiologe

Water Rats     


Verhaltensökologie der Australischen Schwimmratte
Eco-ethology of the Australian Water Rat

Water Rats
IEC 97
Field Study

 This is the short summary of my master thesis, 1995

The Diving Behaviour of Australian Water Rats:

 Optimal Diving and Diving Limits

 Submerged foraging requires special behavioural adaptations for air-breathing animals according to limited time available for finding food. This study shows the strategies Australian water rats (Hydromys chrysogaster, Muridae) use under diverse experimental settings. Water rats are carnivorous and streamlined, they have waterproof fur and webbed hindfeet. Two females were offered pieces of fish or mealworms in water depths varying from 25 to 485 cm at Bielefeld University and Konrad Lorenz Institute, Vienna. Main topics were the description of diving behaviour, foraging strategies, optimal diving, and indications of physiological dive limits.

 No differences can be shown between the animals, and they show little reaction to disturbances. There is no influence of time, learning or satiation during and between the experiments. Water rats propel themselves through the water using their hind legs, they have little buoyancy and search for food while eyes and nostrils are closed by using their vibrissae only. Food is taken to the surface for eating. When food items become very small, animals may also feed while submerged.

 Diving should be optimized with respect to time limits and energetic costs. The 'optimal diving model' by Houston & Carbone (1992) is tested and data fit well. Descent and ascent takes equal time and are highly depth dependent. Speed is constant at 0,72 m/s for all depths, most likely being the minimum cost of transport speed suggesting efficiency maximiziation. Median search time for unsuccessful dives increases up to 11 s at 3,5 m water depth. This means that the animals search for a shorter time than possible at shallow depths. There is no influence of depth to searching success. Total diving time followed potential functions (R² = 0,987 fish, R² = 0,925 mealworm), at first increasing fast, then more slowly before reaching maximum diving duration. Median surface time is 12 s, starting to increase when previous diving exceeds 15 s. It is not affected by food density.

 Maximum diving duration is 36 s which implies 13,14 m possible maximum depth. This is supported by the extrapolation of traveltime and total diving time functions which intersect at 44,5 s / 16,2 m (mealworm) and 37,2 s / 13,95 m (fish). After 30% of unsuccessful dives to more than 4 m depth extremely long surface intervals are observed. This is interpreted as a consequence of anaerobic diving. Aerobic Dive Limit (ADL) is defined as maximum time spend submerged without increased lactate load in the blood due to anaerobic respiration. Lactate breakdown requires an elongated stay at the surface, therefore sudden increase of the minimal surface interval may indicate ADL. Up to 13 s diving time the minimum surface interval is 1 s, limiting the next dive to 7 s, likely because this lasts only for a single breath. This is supported by allometric calculations for a single breath suggesting an energy consumption of 4,2 W using 1,79 ml O2 for an 8,3 s dive. At ~ 27 s diving time another increase in minimum surface intervals occurs to several minutes, most likely because of partial anaerobic respiration. Allometry results in a total of 9,9 ml O2 store and 46 s maximum aerobical diving. This coincides with the measured and calculated maximum diving durations. ADL must be lower because some organs have to remain aerobic. Plotting a moving sum of four subsequent dive cycles results in the best correlation of surface intervals with diving durations, implying an oxygen deficiency occuring during four dives and cleared off by a single long surface interval afterwards. The increase of surface time happens at 80 s total diving time, suggesting an ADL of 20 s and mixed metabolism.

These results indicate no specific adaptations to anaerobic respiration, particularly enhanced O2 stores, high levels of lactate dehydrogenase etc. Additional diving time gained through anaerobic respiration can only amount to seconds requiring minutes of surface stay thereafter and hence cannot be used for optimal diving. The switch model for mixed metabolism is unlikely for water rats.

The work done by the water rats, measured as the sum of vertical dive distances, increases, while effectivity (prey per time unit) and food gain per experiment decreases with depth. The percentage of successful dives increases, stagnating at 100 cm at 50% (mealworm) resp. 300 cm / 80% (fish). Maximum energy output (work per time unit) is reached for mealworms at shallower depth than for fish, which are found quicker and achieve higher effectivity. In deeper water the rat stops foraging at higher remaining food densities. Thus, they use a stopping criterion depending on depth and density. Searching time increases only a little when food density declines, showing exponential increase only for fish at 1 m depth when duration of previous unsuccessful dives is included. Computer simulations give similar results suggesting random search. Orientation by vibrissae results in local and successive food information which requires time dependent foraging stategies.

The diving behaviour of Australian water rats corresponds to optimal diving and allows inferences about diving limits and physiological parameters. Apart from morphology, the most important adaptations to semiaquatic life are optimal diving and flexible searching behaviour with density dependent food exploitation.

Back to the Water Rat Homepage

 Zurück zur Schwimmratten-Homepage

 I´m looking forward to questions and comments! Contact

University Webpages: Uni Bielefeld, Faculty of Biology, Dept. of Animal Behaviour

Dirk Petzold
Haberstraße 14
D-33613 Bielefeld, Germany

Note: this are not official webpages of the university

© Copyright
& Disclaimer