This paper proposes an automatic thresholding method for the discrimination of sky and canopy elements in color hemispherical photographs taken with a digital camera (Nikon Coolpix 950). The exposures for photography were principally determined on the basis of zenith luminance. DIF photo , which is diffuse transmittance calculated from the hemispherical photographs, was related to DIF sensor , which is diffuse transmittance measured directly with a photosynthetic photon flux density sensor. First, the thresholds for calculation of DIF photo were manually assessed in the photographs to obtain the best match with DIF sensor . At the lower pixel-value level in the pixel histograms from the photographs, L-shaped curves were always recognized, and the threshold occurred at the point with the maximum curvature. Second, an automatic thresholding algorithm, taking into account the position of the thresholds, was computerized. Third, the relationships between DIF photo and DIF sensor were field-tested across a wide range of light conditions. The method was effective in a planted coniferous forest and a natural broad-leaved forest and under overcast, twilit, and sunny sky conditions. The coefficients of determination between DIF sensor and DIF photo were greater than 0.99. However, DIF photo taken with Auto-Exposure was overestimated under dense canopy.Résumé : Cet article propose une méthode automatique de fixation du seuil pour distinguer le ciel et la canopée sur des photographies hémisphériques en couleurs prises avec une caméra digitale (Coolpix 950 de Nikon). L'exposition pour les photographies a été déterminée surtout sur la base de la luminance au zénith. DIF photo , qui est la transmittance diffuse calculée à partir des photographies hémisphériques, était relié à DIFcapteur, qui est la transmittance diffuse mesurée directement avec un capteur densité de flux photonique photosynthétique. Dans un premier temps, les seuils pour calculer DIF photo ont été évalués manuellement sur les photographies pour obtenir la meilleure relation avec DIFcapteur. Au niveau le plus bas de la valeur des pixels dans les histogrammes de pixels provenant des photographies, les courbes en forme de L étaient toujours détectées et le seuil était situé au point de courbure maximale. Dans un deuxième temps, un algorithme de détermination automatique du seuil qui tenait compte de la position des seuils a été calculé. Dans un troisième temps, les relations entre DIF photo et DIF capteur ont été testées sur le terrain pour une large gamme de conditions de lumière. La méthode était efficace dans une plantation de conifères et une forêt feuillue naturelle, sous des conditions de ciel nuageux, crépusculaire et ensoleillé. Les coefficients de détermination entre DIF capteur et DIF photo étaient supérieurs à 0,99. Cependant, avec une exposition automatique DIF photo était surestimé sous un couvert végétal dense.[Traduit par la Rédaction] Ishida 2216

The effects of air pollution on the growth of mountain trees were investigated at Buna-daira (1,180 m a.s.l.), about half the way up Mt. Tateyama, located in Japan. Every year, about 1 million tourists are transported by highland buses through the Tateyama-Kurobe Alpine route. Since the route opened in 1971, some tree species along the road have declined and have been blighted, suggesting that bus exhaust was the cause. However, the level of regional and long-range transboundary air pollution has also increased significantly over the last few decades. The atmospheric NO 2 concentration at the roadside in the forest was highly correlated with the traffic density of buses and penetration of the exhaust into the forest was detected. However, the maximum average NO 2 concentration was lower than 3.5 ppbv during the peak traffic period in the year. At Bunadaira, the total stem cross-sectional area at breast height (BA) of the forest was nearly unchanged from 1999 to 2006, but the BA of Fagus crenata decreased 10% and that of Cryptomeria japonica increased 6%. Neither tree growth nor tree death was significantly correlated with distance from the road. The cause of the decline of F. crenata could not be attributed to the effects of road, i.e., air pollution emitted from the buses or edge effects of the road. This area was more affected by regional, longrange transport of air pollution (O 3 , SO 2 , etc.). The average atmospheric O 3 concentration in autumn was higher than 40 ppbv and the recent increase in the O 3 concentration may be an important factor of F. crenata decline through the changes in the interspecific relationships between F. crenata and C. japonica, O 3 sensitive and tolerant species, respectively.

No abstract

We studied the relationship between potential photosynthetically active radiation (PAR), estimated from hemispherical photographs, and shoot extension rates of juveniles of 12 seral tree species. We distinguished between direct light site factor (DIR), diffuse light site factor (DIF), and gap light index (GLI). We used a log‐linear model to relate growth to DIR, DIF and GLI. Potential PAR explained shoot growth rates significantly in 33 cases out of 36 (12 species × 3 PAR). DIF explained the shoot extension rates better than DIR and GLI for 10 of the 12 species. The mean values of maximum shoot extension rates of the tolerant, the intermediate and the intolerant species were 52.4 cm year–1, 64.3 cm year–1 and 87.7 cm year–1, respectively. The maximum shoot extension rates of the tolerant and the intolerant species were recorded in about 50% DIF and more than 70% DIF, respectively. The minimum light level for seedling establishment of the intermediate and intolerant species was more than 20% DIF. We explained distribution characteristics and shoot extension of the species in relation to light in terms of shade tolerance. Predictions based on light‐growth curves and maximum growth curves were similar to field observations made elsewhere, suggesting that these models may be useful to predict extension growth of juvenile trees in mixed species stands.

International audienc