Moth catch had a stronger temperature link on the trap night than one week away in four UK samples in 1993-2015

We tested whether the temperature link to recorded moth catch was strongest on the trap night. Four separate models used 28,570 nights at seven English and two Scottish sites in 1993-2003 and 2004-2015. Models controlled each trap week, current weather and temperature seven days before and after. The current slope minus the mean of the two displaced slopes was 0.157, 0.125, 0.191 and 0.151 log expected catch per degree C. All four intervals were above zero after correction for four tests. The separate displaced slope intervals included zero. Shorter groups, other offsets, curved temperature terms and source-quality exclusions kept the result. Later records contained no explicit zero catches. A positive-catch-only check kept the early contrasts positive. This supports a short-term catch response but cannot separate flight activity, capture, rapid emergence and record selection. Weekly controls also remove slow biological responses. The result does not estimate a causal climate effect, population change or air pollution effects. The package supplies code, aggregate evidence, exact source hashes and a checked reproduction route. Source record tables stay outside the signed bundle.