Monitoring programs tend to begin where the paperwork begins. A utility adopts a sampling schedule, a laboratory fixes its detection limits, and the first results become the reference against which every later change is judged. Anything that happened before that file was opened tends to fall outside the record. Riparian vegetation, wildlife use of a shoreline, and the slow movement of nutrients through a watershed can all change for decades before anyone thinks to measure them. By the time a trend is visible, the earlier condition is often known only from memory or from a few scattered notes.
A recent Guardian report on Indigenous students retracing the Lewis and Clark route offers a useful case. According to its summary, two centuries after the 1804 expedition that Thomas Jefferson sent Meriwether Lewis and William Clark to explore the land west of the Mississippi, Indigenous students are conducting a new wildlife survey along that route. The headline makes the underlying logic plain: “We can’t protect what we cannot see.” The sentence is as much a statement about monitoring as about conservation. A protection decision needs an observation, and an observation needs a method, a place, and a time.
The new survey is valuable on its own terms, but its strongest use comes from comparison. A count taken today describes what exists now. It says much less about direction unless the observers can connect it to earlier counts, older accounts, or documented conditions on the same ground. The same logic applies to water. A turbidity reading in a reservoir, a benthic sample from a stream, or a shoreline observation near an intake gains meaning only when a prior value exists that was collected in a comparable way.
Old Records Carry Their Own Metadata
Legacy data is not automatically usable. Before an older record enters a comparison, it needs the same fields a modern file would require: the date, the location with its precision, the collector, the method, the instrument, and any detection or reporting limit in force at the time. A handwritten field notebook may give a site name and a season but no coordinates. A lab sheet from a previous decade may report results in units or with qualifiers that no longer match current practice. Those gaps do not make the record worthless, but they change what it can support. An older record may justify a statement about direction or rough magnitude, while a claim about a precise change in concentration usually requires methods that match.
Observer and method shifts deserve particular care. A survey run by a new team, using different protocols or a different sampling window, can show differences that reflect the change in people and procedures rather than in the environment. Good practice is to run a bridging period in which old and new methods operate side by side for a defined stretch, and to record the result of that overlap. Without one, the first year of a new program can look like an anomaly when it is really a change in how the site is seen.
Recover the Record Before the Gap Grows
Records fade faster than the conditions they describe. Lab notebooks are discarded during office moves, volunteer logs sit in private drawers, and the engineer who knew why a sample was taken at a particular bend in the river retires. Practical recovery starts with the places that hold documents: state and federal archives, utility annual reports, old permit files, and university field station records. Community-held knowledge and records belong in the same inventory, and they should be approached on terms set by the people who hold them, including how the material will be stored, credited, and used.
Equipment transitions are another point where the record breaks. When an analyzer is retired or a sensor is swapped, the replacement should be bridged against the old unit using the same standards and the same sample set. Service providers can help here. Calibration history, service logs, and cross-check results for a retiring instrument are part of the record a utility will need later, and independent regional dealers such as Jones Air & Water may be able to supply that history when a unit leaves service.
A workable baseline register lists each monitored condition, the earliest usable record, its method and confidence, and the gap that separates it from current practice. Confidence labels matter. A record marked as directional should not carry the weight of a validated result in a compliance file or a rate case. The headline’s point holds only when the observer can actually see the thing being protected, and the same standard applies to any utility asking whether a source is changing. Start the register before the next change, while the people who ran the earlier program can still be asked what they meant.
Further reading: independent regional dealers such as Jones Air & Water.