Every observation kept on this site is part of a record being built for keeps.

The instrumental climate record — the one science relies on — goes back only to the late nineteenth century. Everything older comes from proxies: tree rings, ice cores, lake sediments. These are excellent for temperature trends and terrible for daily life. They cannot tell us when the orioles arrived, when the peepers first called, when the brush piles burned, when the bluebirds gathered the dog's old fur for their nests. The texture of a season — what it actually felt like to live through — survives only in what someone bothered to write down.

This has long been understood. Thoreau's flowering dates from 1850s Concord are now used to measure how much spring has shifted. Gilbert White's Selborne of 1789 is still cited by ecologists. The Chinese imperial weather diaries of the 1700s are data. Ship logs from the age of sail are data. Every faithful daily record becomes, in time, a primary source.

So we keep ours faithfully. The prose carries what the day was like; the date and place and time carry what the day was. Both matter. A graduate student in the year 2090, studying the climate decade we are now living through, will need observations from the watersheds of our era — what arrived early, what arrived late, what arrived at all. We are keeping ours so that the West River will be answerable, when the question comes.

To make the record durable, we intend to keep it in more than one place. On this site, where it lives publicly. In printed annuals, where it lives on shelves. In research library partnerships, where it lives among other historical collections. In the open protocol that underlies the entries themselves, where it lives independent of any company. We intend the archive to outlast us, and to outlast whatever organization happens to be maintaining it at any given moment.

If you are reading this from a long way off — we hope the entries are useful. They were kept by the people who lived here.