Agricultural and Forest Meteorology 100 (2000) 337–347 Effect of variations of PAR on CO 2 exchange estimation for Scots pine T. Vesala a,∗ , T. Markkanen 1a , L. Palva 2b , E. Siivola b,2 , S. Palmroth 3c , P. Hari c,3 a Department of Physics, University of Helsinki, P.O. Box 9, FIN-00014 University of Helsinki, Finland b Applied Electronics Laboratory, University of Technology, Otakaari 5, FIN-02150 Espoo, Finland c Department of Forest Ecology, University of Helsinki, P.O. Box 24, FIN-00014 University of Helsinki, Finland Received 29 March 1999; received in revised form 4 October 1999; accepted 9 October 1999 Abstract A set-up with 161 photosynthetically active radiation (PAR) sensors was used to investigate spatio-temporal variations of irradiance for five horizontal arrays within a Scots pine canopy. The measured PAR was converted to CO 2 exchange using of a shoot-scale photosynthetic response curve and the vertical distribution of the needle area. The net ecosystem exchange was simultaneously measured by the eddy covariance technique. The effect of spatial and temporal averaging of the PAR values and the number of sensors were analysed under different conditions as regards cloudiness and the shading by the foliage. In 1/2 h CO 2 exchange values for the entire canopy, a maximum overestimation of 30% resulted from a spatial averaging over horizontal arrays of 2–5 m and occurred under clear-sky conditions and significant foliage shading. Under partly cloudy conditions, the largest overestimation occurred for a case of little shading and the inaccuracy resulting from 1/2 h temporal averaging exceeded that of spatial averaging. ©2000 Elsevier Science B.V. All rights reserved. Keywords: Irradiance distribution; PAR; Photosynthesis; Scots pine; Net ecosystem exchange; Eddy covariance 1. Introduction Photosynthesis depends on various environmental factors, with solar radiation changing most temporally and spatially. The PAR may vary a lot already above canopies, and the foliage itself generates a compli- cated non-stationary pattern of direct, diffuse and scat- tered radiation (see e.g., Hutchison and Matt, 1977; ∗ Corresponding author. Tel.: +358-9-1918518; fax: +358-9-1918518. E-mail addresses: timo.vesala@helsinki.fi (T. Vesala), tiina.markkanen@helsinki.fi (T. Markkanen), lauri@jeeves.hut.fi (L. Palva), sari.palmroth@helsinki.fi (S. Palmroth), pertti.hari@helsinki.fi (P. Hari). 1 Fax: +358-9-1918518. 2 Fax: +358-9-4512307. 3 Fax: +358-9-1917605. Ross, 1981; Caldwell and Pearcy, 1994). Since the light response curve of photosynthesis is curvilinear, the application of averaged PAR measurements will give biased results for estimates of the exchange of CO 2 . The magnitude of the error is dependent both on the shape of the light response (degree of curva- ture) and the PAR distribution within the period (vari- ation of radiation) (Smolander, 1984). In principle, all information concerning spatio-temporal properties of light is valuable and therefore any averaging should be avoided in the measuring procedure, although in practice this cannot be fully achieved. The purpose of the study was to estimate the spatio-temporal variations of PAR in relation to canopy CO 2 exchange and, to estimate how the num- ber of measuring points and the spatial and temporal averaging of readings influence the calculations of 0168-1923/00/$ – see front matter ©2000 Elsevier Science B.V. All rights reserved. PII:S0168-1923(99)00146-X