Field Notes Journal Entry
Running Models in the Evening Garden
An evening reflection on ecological modelling, seasonal structure, and the relationship between computation and field observation
Running Models in the Evening Garden
This evening the garden is full of soft sound.
Woodpigeons are calling from somewhere beyond the garden wall. A collared dove repeats its steady refrain from the rooftops further down the road. In the undergrowth, a blackbird is turning leaves over with deliberate force, stopping occasionally to listen before continuing its search.
At the same time, a MacBook on the garden table is running ecological models.
There is something faintly absurd about this juxtaposition. Equations are being solved while the evening chorus continues around them. Seasonal curves are being fitted while a real blackbird moves through the border only a few feet away.
And yet the longer I work on these models, the less strange it seems.
Much of modern computational work feels detached from place. Data arrives already abstracted: spreadsheets, databases, plots, clusters. The living organisms that generated the observations can quietly disappear behind the machinery used to analyse them.
I have tried very hard to avoid that separation.
The purpose of these seasonal models is not to replace field observation, but to extend it. Long-term observation creates patterns too large and slow for memory alone to hold comfortably. The models simply help trace those larger structures: the rising and falling presence of species through the year, the asymmetries of arrival and departure, the subtle distinctions between residency, migration, detectability, and aggregation.
The blackbird in the garden is a good example.
Blackbirds are present throughout the year, yet they do not occupy the seasons in the same way. In spring they proclaim themselves constantly from rooftops and trees. By high summer they often become quieter and more concealed. During the moult they can seem to vanish almost entirely into cover before gradually re-emerging through autumn and winter.
The models attempt to describe that changing seasonal visibility mathematically, but the mathematics only exists because the lived pattern exists first.
That relationship matters to me.
I do not think of computation here as something opposed to natural history. If anything, it feels closer to an extension of the notebook tradition: another way of paying attention to recurring structures in the living world.
So the laptop continues quietly fitting curves while the evening deepens around it.
The pigeons continue calling. The blackbird continues searching the undergrowth. And somewhere inside the machine another small approximation of seasonal life slowly converges toward a solution.