Using agent-based models for management of marine populations
Marine species are exposed to anthropogenic disturbances that cause animals to change behaviour, reduce their access to food, cause them to die, and that may ultimately result in population declines. Here we describe how knowledge of behavioural responses to disturbances can be incorporated in highly realistic agent-based models to assess whether populations are able to survive the different disturbances that we expose them to. We use the harbour porpoise (Phocoena phocoena) as model organism and demonstrate how to predict the combined population impacts of bycatch and noise from wind farm construction and ships. Based on this, we discuss what is needed to extend the framework to other species. We argue that spatially explicit, process-based models are needed to fully understand how competition for food and behavioural reactions to disturbances will shape the dynamics of populations when we change the marine environment.
![Accounting forAnimal Movements
Dähne et al., MEPS (2017)
Probability
of
occurrence
[%PPM]
Hours after end of piling
Harbour porpoise response to
pile driving at DanTysk OWF
1.5–3 km
from pilings
6–9 km
from pilings
• Population densities can recover
within hours after piling](https://image.slidesharecdn.com/jnn221219abmsformarinemangbesedinburgh-221219155119-f8cc35c0/85/Using-agent-based-models-for-management-of-marine-populations-4-320.jpg)
Using agent-based models for management of marine populations
- 1. Using agent-based models for management of marine populations Jacob Nabe-Nielsen, Dept. of Ecoscience, Aarhus University BESAnnual Meeting, 19 Dec 2022, Edinburgh
- 2. Disturbances in Marine Environments Duarte et al., Science 371 (2021) DanTysk OWF
- 3. Agent-Based Models (ABMs) ABMs make it possible to – simulate animal movements and energetics using realistic landscapes – simulate changes in movements associated with disturbances – let population dynamics emerge from competition, bycatch etc.
- 4. Accounting forAnimal Movements Dähne et al., MEPS (2017) Probability of occurrence [%PPM] Hours after end of piling Harbour porpoise response to pile driving at DanTysk OWF 1.5–3 km from pilings 6–9 km from pilings • Population densities can recover within hours after piling
- 5. ABM for Porpoises: DEPONS Move, use energy Find food, increase energy Mate, give birth, lactate etc. Energy dependent mortality Reduce food in patches Emergent population size Time of the year Disturbance
- 6. Collection of Movement Data
- 7. GPS tracked porpoises Simulated animals Movements in DEPONS Random walk component Spatial memory component Simulated movements = +
- 8. FoodAvailability High-density area = high food levels North Sea Gilles et al., Ecosphere. (2016)
- 9. Population Effects of Pile-Driving Nabe-Nielsen et al., Cons. Lett. (2018)
- 10. Population Effects of Ship Noise • Vessels deter porpoises at distances >2 km • Affects porpoise metabolic rate and potentially their fitness
- 11. Changed FoodAvailability Changes in current velocity and mixing Christiansen et al. (2022); Front. Mar. Sci.,
- 12. Conclusions • It is important to consider animal movements when assessing population effects of disturbances. • Inclusion of realistic energetics makes it possible to assess impacts of changes in prey availability. • ABMs allow us to assess the combined impact of different disturbances (ships, pile-driving, fisheries, by- catch etc.). • ABMs are process-based; likely to produce realistic predictions under novel conditions.
Editor's Notes
- #2: Abstract: Marine species are exposed to anthropogenic disturbances that cause animals to change behaviour, reduce their access to food, cause them to die, and that may ultimately result in population declines. Here we describe how knowledge of behavioural responses to disturbances can be incorporated in highly realistic agent-based models to assess whether populations are able to survive the different disturbances that we expose them to. We use the harbour porpoise (Phocoena phocoena) as model organism and demonstrate how to predict the combined population impacts of bycatch and noise from wind farm construction and ships. Based on this, we discuss what is needed to extend the framework to other species. We argue that spatially explicit, process-based models are needed to fully understand how competition for food and behavioural reactions to disturbances will shape the dynamics of populations when we change the marine environment.
- #12: Mean changes in horizontal velocity (u) and mixing rate. Christiansen, N., Daewel, U., Djath, B. & Schrum, C. (2022). Emergence of large-scale hydrodynamic structures due to atmospheric offshore wind farm wakes. Front. Mar. Sci., 9, 1–17.