Note
What is a desk-based assessment in archaeology?

A desk-based assessment, usually just called a DBA, is the paper-and-screen stage of an archaeological project. No trenches, no trowels, no boots on the actual ground. You're pulling together everything that's already known or recorded about a site before anyone decides whether, or how, to dig.
Most planning authorities in the UK expect a DBA as the first piece of evidence in the archaeological chain for a development. It's often called pre-determination work because it has to land before planning permission is granted. Some briefs label it Stage 1, with evaluation trenching and mitigation following as later stages if the DBA flags enough potential to warrant them.
What actually goes into it
A DBA pulls from a fairly fixed set of sources: the Historic Environment Record, historic mapping back through the tithe and enclosure maps where they exist, geotechnical borehole logs if the site's been surveyed for anything else, and whatever aerial photography happens to be sitting in the county archive or the National Monuments Record. You walk the site if access allows, note surface finds or earthworks, and write up a statement of archaeological potential, usually banded as low, moderate or high for different periods.
The aerial photo trawl is the part that varies most in quality from one DBA to the next. Historic England's archive holds decades of RAF and reconnaissance sorties, but coverage of any given field depends entirely on whether someone happened to be flying over it in a dry summer with a camera pointed down. For a lot of rural sites, especially ones that haven't come up for development before, there's nothing useful in the archive at all. You end up writing "no cropmark evidence identified" when the honest answer is closer to "nobody's ever looked properly."
Where the desk study runs out of road
A DBA is a documents exercise by design. It tells a planning officer what's known, and it's meant to set reasonable expectations for the fieldwork that follows. But because it leans so heavily on whatever aerial coverage already exists, a site can come out of the desk stage reading as low potential simply because the archive is thin, not because the ground is empty.
Cropmarks are weather-dependent and short-lived. Walls, ditches and infilled channels show up as differential crop stress for a narrow window most summers, sometimes only a few weeks, and if nobody flew that window over your parcel, the evidence just doesn't exist in any archive. The record falls short there, even though the ground underneath may still hold the feature.
This is the part of the DBA where a current pass over the study area, rather than a trawl through whatever's already filed, earns its place. Buried Feature Detection reads NIR imagery over your survey area through the growing season and returns a georeferenced cropmark layer, walls, ditches and palaeochannels flagged as differential growth, delivered as a layer you can drop straight into the desk-based assessment or drag into QGIS alongside your HER data and historic mapping. It doesn't replace the documentary research. It closes the part of it that depends on luck with the weather and the flying schedule.
A quick note on terminology
"DBA," "pre-determination archaeological assessment" and "Stage 1 archaeological assessment" all describe the same piece of work. Which term shows up in your brief usually comes down to the local planning authority's template, though most still trace back to the same CIfA standards underneath. The content expected is the same regardless of the label on the front page.
If your next DBA is due in a few months and the field hasn't had a proper aerial pass in a growing season, it's worth checking what's actually out there before you write "no cropmark evidence."