For the last few months I have been holding down the Landscape Analysis desk at the pacific regional branch of the Canadian Wildlife Service, while the regular guy was off doing emergency planning. This was the last week of the contract and I’m once again a private citizen. Most of my time was given to a particular project: an upcoming study of the impacts of salmon gillnet fishing on diving seabirds. I wanted to take the job because the project has some really interesting research challenges. I was also curious to get an inside view of how conservation science is done in government.

Many of the projects I’ve been involved with in academia share similar themes to the work being done by the CWS: landscape-scale, data-driven assessment of trends and interactions in environmental systems. But doing research at CWS had a different feel than the academic labs I’ve known. Government research answers to different criteria and has a different audience. The primary motivation is not to satisfy your personal sense of cosmic wonder, and share what you learn with the peers who might read your journal submission. Nor are you hoping that some government conservation type might eventually take note of your theory, and embed it somehow in conservation regulation. In government you’re supposed to be doing the right thing, right now, on behalf of everybody. And not only do you have to do the right thing, you might have to demonstrate that the thing you’ve chosen to do is, in some plausibly objective sense, right. In practice that seems to mean being able to tie your choices back to previously identified criteria, which in turn link back to pre-existing directives, which are all presumably grounded on specific bullet points in legislation. It’s meant to be science as a service, not as an interest. The result might be a more conservative approach to research. You can take more risks when it’s only your own thesis goals on the line.

I don’t mean that as a criticism. Transparency and accountability are necessary to good governance, and if the business of the day has to move a little more slowly to ensure a documented lineage of decision making, so be it. But I was struck by a comment made by a poster on an internal forum, who suggested that space for innovation be deliberately built into the CWS’s work practices. If you have to be able to prove that you can answer a question before you’re allowed to expend resources asking it, then there are a class of complex questions which will just get left alone. And a lot of those types of questions tend to show up in conservation. I wasn’t in the Service very long, but my impression was that many people there are very aware of the benefits of asking difficult questions, and are finding ways to do so. I also got the impression that that objective is turning out to be, not surprisingly, tricky.

More concretely, I think I was seeing an institution in the midst of a protracted shift from keeper-of-records to having a more actively analytical capacity. Which would certainly be a good thing. In a changing biosphere it isn’t enough to know what the trajectory is. If your mandate is to intervene beneficially, you have to know what is causing the trends. Those two goals of memory and interpretation are not mutually exclusive, and a history of record keeping could be a key foundation for a great research agency.

This transition will not be helped by the legacy of Stéphane Dion’s time as environment minister. I gather (from occasional passing references, so maybe this narrative is inaccurate) that Dion was responsible for deliberately splitting the Service into two halves, one charged with the traditional duties of counting things, and one given the tasks of “science and technology”. There must have been some compelling  reasons to do so, but in the absence of those arguments I can’t fathom how anyone would think that severing the two hemispheres of the brain would yield good results. For starters, it’s impossible to do biological survey work without doing science. Sampling the world is not as simple as going into it, jabbing an index finger at things, and counting off. I could write an entire blog post about that, so I won’t.

Not surprisingly, data is a topic of ongoing planning and debate at the CWS. The Service has an extraordinary institutional memory. Their internal records for some species include decades of repeated censuses and surveys (in some cases repeated by the same person — imagine that!). They have stacks of tech reports with beautiful typewriter-printed tables, and gorgeously hand-drafted survey maps depicting the state of bird-colonized pacific atolls in the 1970s, tucked into their back-flap pockets. The people of the CWS hold a collective image of the historic trends of distribution and movement of bird species on the coast and across the country. And they are very actively in the process of transcribing that tech-report and spreadsheet data into a living digital resource but hoo-boy, doing so leads to some thorny posers around process and infrastructure.

Whilst chipping away at my own corner of that issue, I was struck by just how screwed up the state of GIS data storage is in general. How does any institution store spatial data in a truly cross-platform, future-proof way that protects it for posterity whilst simultaneously exposing it for analysis? Particularly if that data has any kind of significant relational structure to it? There are some beautiful ideas for the future, but the current options aren’t very satisfying. I had somehow always assumed that problem was solved somewhere, but now I’m confused how anybody does it.

The study I was contributing to will ramp up this summer, and depending on the results of that field season (and on the future of the salmon gillnet fishery, which is hardly clear) could be ongoing for years. I hope I’ve helped set up that program for success. Another difference between government work and working in academia or private industry is that little voice in your head, constantly wondering if what you’re doing would be judged a reasonable expenditure of the taxpayer’s dollars, if they could watch you doing it. For my own sake it was a definite win. The challenges were interesting — I was especially pleased to enhance my experience with database wrangling — and that inside-government perspective did indeed yield plenty to think about.