Tanged Up in Seasons—Alex Wells

This past May, I traveled back in time.  

From the top of East Rock Park, the stately houses and university buildings of New Haven barely poked out from beneath an overflow of green leaves. It was early May, but it already felt like summer in Connecticut. Seven days later, I’d returned to early April, the glowing leafscape replaced with a cobweb of bare aspen branches. Hiking up another 1000 feet, I slogged into the snowpack and back into the depths of winter. For this journey through time, I relied on neither magic spells nor a time machine. Instead, my 2005 Subaru was all it took to transport me out of one season in Connecticut and into another in the high peaks of Colorado. 

We generally think of the Earth’s seasons as being functions of time, with one sliding inexorably into the next. In the United States, summer officially starts on June 21st, unofficially on Memorial Day, and “meteorologically” on June 1st. However, the actual reality of one season moving into the next depends just as much on where you are as when. 

In Colorado’s Roaring Fork Valley—where I grew up and where I’m now working as a Western Resource Fellow—this seasonal entanglement is on full display in mid-May. At 7,000 feet above sea level on the valley floor, the aspens and cottonwood might be decked out in fresh green buds. But if you go a mile north and up a thousand feet, you’ll find leafless scrub oaks. Or switch to the other side of the valley, to a north-facing slope, and there may still be pockets of snow pooling stubbornly beneath the conifers.  Keep going up and you’ll be wading through snow well before you hit a summit. Eventually, the “green wave” of spring washes over all of Colorado, but its time of arrival can vary by months between the low valleys and the high peaks. A number of studies have shown wildlife will “surf” this wave of green vegetation as springtime stretches out across the landscape, taking advantage of this natural variability (Aikens et al., 2020). 

All of this is just within a single valley. If you zoom out further, the clean divisions between winter, spring, summer, and fall break down further, even within the United States. If you’re an elementary schooler in Miami or Los Angeles, you might get a January lesson about the four seasons and then step outside into a city at 70° Fahrenheit and full of leafy trees (at least apart from the exotic species adapted to somewhere else’s seasons). That’s not to say that there are no seasons there, just that the local variations don’t map neatly onto the idealized, archetypal seasons of London or Boston.  

Or, instead of looking across the whole country, you can focus on a single point on the landscape. A patch of trees on a hillslope or a hollow that was carved by ancient glaciers. Even in that spot, you’re likely to find hints of spring in the dead of winter or bleak premonitions of winter in the height of summer. In a chapter of Pilgrim at Tinker Creek, Annie Dillard writes “What we think of the weather and behavior of life on the planet at any given season is really all a matter of statistical probabilities; at any given point, anything might happen. There is a bit of every season in each season. Green plants—deciduous green leaves—grow everywhere, all winter long, and small shoots come up pale and new in every season. Leaves die on the tree in May, turn brown, and fall into the creek. The calendar, the weather, and the behavior of wild creatures have the slimmest of connections. Everything overlaps smoothly for only a few weeks each season, and then it all tangles up again.” 

Seasons are fractal, becoming ever more intricate as you zoom in and out over time and space. While the four seasons offer a clean division of the calendar or a straightforward classroom lesson, they are approximations at best. Stick figure drawings in place of photographs. Convenient and easy to draw, but superficial and bland.  

However, just as much as our common understanding of seasons can be detached from the reality on the ground, we have enormous influence over what that reality actually looks like. In Colorado, the April 1st snowpack has been between 3% and 23% lower in the 21st century than it was in the latter half of the 20th. This trend of shrinking winter is projected to continue, with peak snowpack shifting somewhere from a few days to several weeks earlier by 2050 (Bolinger et al., 2024). This isn’t solely a function of human-caused climate change—though statewide temperatures already average multiple degrees Fahrenheit warmer than they did 50 years ago. It is also a function of land management and use. In the sandy drylands of the Colorado Plateau, the combination of overgrazing, energy production, off-highway vehicle usage, and a changing climate is resulting in huge amounts of dust deposited on the winter snowpack in the Southern Rockies (Duniway et al., 2019). Covered with dust, snow reflects less sunlight and melts out faster. Studies have suggested that this speeds up the arrival of the green wave, and the disappearance of snow, by three to six weeks in Colorado (Bolinger et al., 2024).  

These shifting patterns are enormous and unintentional. Collateral damage from the way people in the United States and beyond accumulate wealth, produce energy, eat dinner, and have fun. As terrifying as these unintended climatic consequences are, it is also possible for humans to shift the progression of seasons in a way that is intentional and positive for local people and ecosystems. And no, I’m not even talking about geoengineering. 

In the Intermountain West, the restoration of stream ecosystems is pushing watersheds back in time, closer to how they were a hundred years ago when atmospheric carbon was below 300 ppm, rather than over 400 ppm. From the rangeland streams of Nevada to the high-elevation headwaters of Colorado, process-based restoration (PBR) has been shown capable of extending the flow of streams late into the dry season (American Rivers, 2024). In one project on an Idaho ranch, the construction of a handful of beaver dam analogs (essentially human-made structures that mimic beaver dams) and the subsequent relocation of a family of beavers to the newly created ponds turned an ephemeral trickle into a verdant wetland that flows 42 days longer in the year. Of course, many PBR projects do not have such a profound (or even measurable) impact on the timing of streamflow, and these data suggest that they cannot offset the impact of climate change on streams (Hafen, 2017; Dittbrenner, 2019). Nonetheless, these projects remind us that how the seasons play out is malleable and intricate, shifting with space and human action, rather than a simple drone of winter, spring, summer, fall… 

New Haven, CT on May 8th from the top of East Rock Park (300 ft) versus the San Juans, CO on May 19th, from the top of Twin Peaks (10,800 ft). Photos by Alex Wells. 


Aikens, E.O., et al. 2020. Wave-like Patterns of Plant Phenology Determine Ungulate Movement Tactics. Current Biology, 30(17). doi.org/10.1002/ecs2.2650  

American Rivers, 2024. Restoring Western Headwater Streams with Low-Tech Process-Based Methods: A Review of the Science and Case Study Results, Challenges, and Opportunities.  

Bolinger, R.A., Lukas, J.J., Schumacher, R.S., Goble, P.E. 2024. Climate Change in Colorado, 3rd edition. Colorado State University. doi.org/10.25675/10217/237323.   

Dillard, A. 1974. Pilgrim at Tinker Creek. Harper’s Magazine Press. 

Dittbrenner, B. 2019. Restoration potential of beaver for hydrological resilience in a changing climate, PhD Dissertation, University of Washington. 

Duniway, M.C., Pfennigwerth, A.A., Fick, S.E., Nauman, T.W., Belnap J., Barger, N.N. 2019. Wind erosion and dust from US drylands: a review of causes, consequences, and solutions in a changing world. Ecosphere 10(3). doi.org/10.1002/ecs2.2650  

Hafen, K. 2017. To What Extent Might Beaver Dam Building Buffer Water Storage Losses Associated with a Declining Snowpack? Master of Science Thesis, Utah State University.