HEP In the literature: contested

Can DAMA's background subtraction create a spurious yearly modulation?

In plain words

DAMA subtracts each year's average rate before looking for a yearly wave; if the true rate drifts slowly, this step can manufacture a wave that is not there. The question is whether this explains all or part of the signal.

Precise statement

DAMA forms residuals by subtracting the annual mean rate per energy bin, cycle by cycle (roughly September to September). For a time-dependent background $B(t)$ (e.g., decaying cosmogenic or radiogenic components, or slow gain drift), this procedure induces a sawtooth residual whose fit to $A$ cos(2 pi (t - t0)/1 yr) can give a nonzero A with a phase that depends on $\mathrm{d}B/\mathrm{d}t$. An answer is a quantitative yes or no: whether a background evolution consistent with DAMA's published spectra and stability data reproduces the reported amplitude and phase.

What would settle it

Analysis of DAMA's absolute, non-subtracted single-hit rate versus time per energy bin, or a full forward model of the DAMA background evolution fitted to their data.

Status in the literature

Buttazzo, Panci, Rossi and Strumia (JHEP 2020) showed the mechanism can produce a modulation, and COSINE-100 (Sci. Rep. 2023) showed that applying the DAMA method to COSINE-100 data induces an artificial modulation; DAMA and Messina, Nardecchia and Piacentini (JCAP 2020) dispute that it applies to DAMA data.

Related problems