--- name: physiology-signals title: Physiology Signals description: Physics and pitfalls of physiology sensors in plant phenotyping - chlorophyll fluorescence (dark vs light adapted), reflectance spectra and vegetation indices, red edge, water bands, and relative units. Use before interpreting fluorescence, multispectral, or hyperspectral traits. keywords: - level:L1 - phenotyping - fluorescence - spectral --- # Physiology signals ## Chlorophyll fluorescence - **Dark-adapted** measurements give the maximum quantum yield of PSII, Fv/Fm = (Fm - Fo) / Fm. Healthy leaves of most species sit in a narrow high band; lower values indicate stress or damage. Check the protocol: was the plant dark-adapted, and for how long? A value labelled Fv/Fm from a light-adapted step is not Fv/Fm. - **Light-adapted** protocols (light curves, quenching analysis) give operating efficiency and quenching parameters (PhiPSII / QY at a step, NPQ, qP, qN, qL, Y(NPQ), Y(NO)). Values depend on the light step; compare the same step across plants. Step names usually encode the light level or the recovery phase -- read the protocol docs; describe your reading as uncertain if the export does not define them. - **Raw fluorescence levels** (Fo, Fm, Ft) are in arbitrary units that depend on camera gain and leaf area; use ratios for comparisons. - **Plausibility**: yields are fractions in [0, 1]; values outside are artefacts (division by near-zero on dim pixels, empty masks). Check vendor whole-plant values against mask-median values when both exist. - **Timing**: fluorescence changes over the day; compare same-session values. ## Reflectance spectra and indices - **Reflectance** should lie in [0, 1] after white/dark calibration; values above 1 or below 0 point to calibration or specular problems. - **Shape checks** on a median plant spectrum: low in the visible red, a steep rise at the **red edge** (roughly 680-750 nm), a high near-infrared plateau, and **water absorption** dips in the shortwave infrared (near 1450 and 1940 nm). A spectrum without these is not vegetation or not reflectance. - **Indices** (NDVI, PRI, PSRI, SIPI, MCARI, OSAVI and variants) differ between implementations: check the formula and the exact bands in the docs before comparing to literature values. Record the formula and source references with the analysis; a card is optional. - **Wavelength labels**: parse wavelengths from metadata or column names; look for near-duplicate band columns (see `onboard-dataset`). - **Water signals** from a few bands (ratios of reflectance near 950-970 nm vs near 1450 nm, or similar) are relative indicators, not percent water content, unless the docs give a calibration. ## Units - `a.u.` (arbitrary units), vendor ratios, and indices are relative: compare within one instrument, protocol, and session; do not compare magnitudes across instruments or facilities. - Report the method (sensor, protocol step, mask, aggregation) with every result using the available metadata or analysis notes. An `observations` table or a column named `method` is not required.