hide
Free keywords:
-
Abstract:
Aims Gas exchange measurements on individual plants
depend largely on chamber systems, and uncertainties
and corrections in current flux calculation procedures
require further assessment.
Methods We present a practical study with novel
methods for analyses of flux uncertainties in an original
chamber design excluding soil fluxes and allowing simultaneous
measurements of whole-plant photosynthesis
and transpiration.
Results Results indicate that random errors caused by
IRGA noise and the lack of criteria to optimize the time
window (TW) of chamber enclosure lead to significant
flux uncertainties (12 %). Although enclosure should be
rapid to minimize plant disturbances, longer TWs
(3 min) increase confidence in flux estimates. Indeterminate
stabilization periods in existing calculation protocols
cause significant systematic errors. Stabilization
times were identified via the change-point detection method, and flux uncertainties were reduced. Photosynthesis was overestimated by up to 28 % when not
correcting the evolving CO2 molar fraction for water
vapour dilution. Leakage can compromise flux estimates,
but was negligible (ca. 2 %) here due to the large
chamber-headspace and relatively small values of both
collar contact length and closure time.
Conclusions A bootstrapping, resampling-based flux
calculation method is presented and recommended to
better assess random errors and improve flux precision. We present practical recommendations for the use of whole-plant chambers.