Apply now: 3-year postdoctoral fellow on optimizing adaptive management!

I’m delighted to announce that our 3-year postdoctoral fellow position on optimizing adaptive management is now open for application!

We are seeking a highly motivated and dynamic postdoctoral research fellow to join CSIRO Ecosystem Sciences’ conservation decisions team led by Dr Iadine Chades, to undertake research on optimizing adaptive management decisions under imperfect detection. The postdoctoral research fellow will be supervised by Iadine Chades, Andy Sheppard (CSIRO) and Pr Tom Dietterich (Oregon State University).

Resources to halt global biodiversity decline are still inadequate.  Managers of threatened species therefore need guidance on how to best invest their scarce resources to maximise the chance of saving species in the long term.  Decision theory is now helping decision-makers prioritise biodiversity threat management across time and space but a major drawback with current decision approaches is their need for “data-hungry” models that simulate how a species or system will behave in the future under different management decisions.

Specifically you will:

  • Develop innovative concepts, theories and techniques to facilitate optimal adaptive management over time for hard to detect invasive and threatened species populations.
  • Contribute to the development of adaptive management recommendations to help practitioners protect biodiversity.
  • Publish findings in high impact journals, present finding at both national and international conferences and participate in interdisciplinary working groups.
  • Contribute to a dynamic, innovative and effective research team working with CSIRO Ecosystem Sciences.
  •  Participate in CSIRO’s postdoctoral training program.

Location:   Dutton Park, Brisbane, QLD, Australia
Salary:      AUD$81K – AUD$88K plus up to 15.4% superannuation
Tenure:     3 year specified term
Reference: Q13/03434

To be successful in this position you will need:

  • A PhD in artificial intelligence, ecology, conservation, computational sustainability or related field of decision theory (e.g. applied mathematics, computer science, economics or related discipline).

Note:  Owing to the terms of CSIRO Postdoctoral Fellowships, you must not have more than 3 years relevant post doctoral experience.

  • Demonstrated research achievement in decision theory, optimal resource allocation, adaptive management or ecological modelling. In particular, demonstrated research achievement in one or more of Markov decision processes (MDP), partially observable Markov decision processes (POMDP), stochastic dynamic programming, reinforcement learning and adaptive management.
  • Demonstrated ability to initiate research characterised by originality, creativity and innovation. Publish the findings from research in international peer reviewed journals or selective conference proceedings.
  • Enthusiasm for applying advanced computational and decision theoretic tools to ecological problems.
  • High-level written, oral and interpersonal communication skills, including demonstrated experience in preparing briefings for a range of audiences, and ability to work effectively in a team.

Position Details – Q13/03434

Seminar at the Global Change Institute, University of Queensland

I’m excited to present my work at the Global Change Institute tomorrow (26/09/2013). A good opportunity to communicate and reflect on my work so far. I can promise that the slides will have no equations :-). Here’s the abstract:IMG_1115

At the forefront of linking conservation science with quantitative tools from the field of artificial intelligence (AI), Dr Iadine Chadès will introduce the process of making smart conservation decisions under imperfect knowledge and resource constraints. During her PhD, Dr Chadès developed new methods to tackle complex optimisation problems for mobile robots using Markov decision processes (MDP). She discovered that these models can also be used to improve decision-making in modern conservation science – teaching a robot to navigate utilizes the same mathematics as choosing the best conservation actions to save threatened species under uncertainty. Eager to contribute to conservation science, she changed career and turned towards decisions in ecology. Combining expertise in AI with ecological and economic models, this seminar will look at complex applied conservation problems and the solutions that can be applied to efficiently eradicate invasive weeds, control mosquito-borne diseases and protect threatened species from extinction.

Accounting for complementary to maximize monitoring power for species management

Congratulations to Ayesha Tulloch. We have a new paper in Conservation Biology that addresses how to monitor management actions. I really like that paper and I hope it will become a good reference, check it out! Note that we also provide the Matlab code.

One challenge faced by researchers and conservation practitioners is designing and implementing effective monitoring programs particularly when funds are limited. Decisions about how to monitor are hindered by uncertainty in management outcomes. This research demonstrates a new framework for addressing the uncertainties in selecting species for monitoring change due to a management action or policy, using network theory and decision analysis.

Tulloch A.I.T., Chadès I., Possingham H.P. (2013) Accounting for Complementarity to Maximize Monitoring Power for Species Management. Conservation Biology 27, 988-999. Abstract

Adaptive management of migratory birds

We have a new paper published at IJCAI (top Artificial Intelligence conference, ranked A*, probably the most selective conference, congratulations to Sam and team!).

Nicol S, Buffet O, Iwamura T, Chadès I (2013). Adaptive management of migratory birds under sea level rise. Proceedings of IJCAI-13, Beijing, China. (PDF);

Or read the blog version on the computational sustainability website.

In this paper we are posing an adaptive management challenge to the AI community:

  • Why do we care? Because solving adaptive management problem is a complex optimisation problem and efficient methods are lacking!
  • What are we hoping? We hope that future AI research will account for the specific description of adaptive management problems using our problem as a classic benchmark problem.
  • How can you help, what’s next? If you have a complex problem feel free to submit a challenge to the AI community!

 

Sam at IJCAI 2013
Sam at IJCAI 2013 – Photo: O. Buffet

Migratory connectivity magnifies habitat loss

Graph representing the migratory flyway of the eastern curlew

Tak‘s paper is out! Don’t miss the bottleneck index that we derived – a handy tool to predict the most important nodes.

Iwamura, T., Possingham, H.,  Chadès, I., Minton, C., Murray, N., Rogers, D., Treml, E., Fuller, R. (2013) Migratory connectivity magnifies the consequences of habitat loss from sea-level rise for shorebird populations Proc R Soc B 280: 20130325

Abstract:

Sea-level rise (SLR) will greatly alter littoral ecosystems, causing habitat change and loss for coastal species. Habitat loss is widely used as a measurement of the risk of extinction, but because many coastal species are migratory, the impact of habitat loss will depend not only on its extent, but also on where it occurs. Here, we develop a novel graph-theoretic approach to measure the Continue reading Migratory connectivity magnifies habitat loss