Jonathan Tonkin
@jdtonkin
Professor of ecology at UCNZ | Using science to anticipate where our living world is headed | Writing Predirections newsletter on biodiversity, climate & water
The Primer provides a practical guide to generalized additive models for location, scale and shape (GAMLSS), including theory, implementation, diagnostics, and applications from medicine, ecology, climate science, and environmental management.
In our new Nature Reviews Methods Primers article led by Julian Merder, we introduce distributional regression, a framework that models the entire conditional distribution of a response rather than just its mean.
Most statistical models focus on a single question: How does an explanatory variable affect the mean response? But environmental and biological systems are often shaped not only by changes in the mean, but also by changes in variability, skewness, thresholds, and the frequency of extreme outcomes.
There’s something about seeing things in new detail that opens up new levels of wonder. The way these florets hold water droplets is just mind blowing to me.
Ecosystems don’t optimise around a single best strategy. In variable environments, they persist by spreading risk across multiple dimensions. Here, I look at 4 ways nature does this: Across genes & species, space, time, & strategies w/in organisms. predirections.substack.com/p/how-nature...
Removing barriers reconnects rivers. Good news: native species usually respond positively. Less-good news: invasive species can too. Our new global review shows that river restoration can create winners on both sides of the native/non-native divide, but results are generally positive.
In ecosystems, who arrives first can matter just as much as who is best adapted. Early arrivers can change their environment in ways that can favour some species and disfavour others.
Seagrass meadows are one of the planet's most overlooked ecosystems. Despite this, they are fundamental for coastal biodiversity and fisheries, help to protect shorelines, and even store vast amounts of carbon. 📷 Benjamin L. Jones 1/
Let this sink in for a minute. Globally, a subset of 34,836 populations of 5,495 species has declined by an average 73% in abundance between 1970 and 2020.
Wrapping up an incredible 10 days in Spain and Switzerland. Wish I could stay for the full @worldbioforum.bsky.social but time to head home. Leaving feeling incredibly inspired to do meaningful research. Reconnected with old friends and made many new ones. Now to finish writing my grant.
Spent the past week debating, brainstorming and writing about what makes a more predictive ecology. The creative juices flow when in the company of clever, creative people and walking among amazing geological features like this.
Super fun to hang out with @oldenfish.bsky.social at Friday Harbor, WA. What a place to work on a paper. There’s something about good friends + wild places that sharpens ideas. Stay tuned for something exciting... and different.
Your unread books are a map of who you're becoming: on tsundoku, antilibraries, and the joy of never catching up predirections.substack.com/p/your-unrea...
Very cool to experience Christchurch’s new stadium. Even some nice University of Canterbury advertising.
2026 is shaping up to be a very low flow year in snowmelt driven rivers of the western US. This is what 100 years of river data looks like on the Yampa River, Colorado. Note the thicker black line for 2026. The species that depend on these floods won’t be happy.
Colorado's mountains have just come through a winter of record low snowpack. The spring floods that cottonwoods, fish, and river ecosystems depend on won't come this year. Here's why the timing of seasons matters more than most climate coverage lets on. predirections.substack.com/p/climate-ch...
Absolutely delighted to share that Daniel Hernández-Carrasco has been named National Champion representing Aotearoa New Zealand for the Frontiers Planet Prize. What a good way to celebrate Earth Day! @frontiersin.bsky.social
Every Earth Day I wonder what we'll tell the next generation. That we knew and kept going anyway? We still have time to write a different ending.
How do we know the magnitude of a biological invasion? How do we compare invasions of tiny organisms like ants with that of larger ones like deer? Simple. Compare their total biomasses. In our new paper in Bioscience, we do just this. doi.org/10.1093/bios...
Emergent aquatic insects sustain riparian birds. Turns out freshwater restoration and conservation can help terrestrial ecosystems! www.nature.com/articles/s41... Christian Schürings and @oldenfish.bsky.social
Meet Dolomedes aquaticus—a fascinating fishing spider. They live on the banks of gravel-bed rivers in NZ & feed on aquatic insects by fishing from the margins. With back legs anchored and the rest of their body floating, they feed at night by sensing vibrations from prey through the water surface.
