0808-B1
Inge Jonckheere[1], Stefan Fleck, Bart Muys and Pol Coppin
Rapid, reliable and objective estimation of Leaf Area Index (LAI) is of utmost importance in numerous studies of the Biosphere, as LAI is very often a critical parameter in process-based models of vegetation canopy response to global environmental change. This paper compiles current knowledge concerning the use of hemispherical canopy photography for LAI determination of forest canopies. The usefulness of LAI measurements has already been demonstrated in that context. We suggest that the use of a digital camera with high dynamic range has the potential to overcome a number of the technical problems described in previous studies about hemispherical photography.
Leaf Area Index (LAI), defined as the projected area of foliage per unit ground area (Watson, 1947), is the most common and, arguably, most useful comparative measure of vegetation structure in forest canopies. Accurate measurements of vertical LAI profiles of forests are of fundamental importance as a basis for estimating the exchange of carbon, water, nutrients, and light at the stand, landscape and ecosystem scale. LAI is thus a critical parameter in physiology-based models of forest responses to global environmental change. There are several procedures to estimate LAI, both directly and indirectly (see review of both methods in Chason et al., 1991). Apart from the LAI-2000 instrument (Li-Cor Inc., Nebraska), which captures and processes the information gathered by a hemispherical lens on the coarse resolution of only five concentric rings, a common optical method to assess LAI indirectly is by means of hemispherical photography.
Fig.1. Hemispherical image
Hemispherical canopy photography is a technique for studying plant canopies using the full information of a close to 180° field of view. Highly resolved photographs are taken through a hemispherical (fisheye) lens from beneath a canopy looking vertically upward. Thus, it has the advantage that it captures the forest canopy structure in a two-dimensional way without integration over large areas of the picture, making more detailed spatial analyses possible. A hemispherical photograph (Fig.1) acts as a permanent record of canopy geometry (Chen et al., 1997) and may be evaluated based on position, size, density, distribution of canopy gaps on the image. Gap fractions may be calculated based on light attenuation and contrast between features within the photo (sky vs. canopy) (Frazer, 1997). They are basically the result of light interception by leaves, and therefore a given LAI in combination with the known statistical distribution of foliage elements and their angles produces a certain gap fraction. The evaluation of hemispherical photographs is based on the inversion of this kind of light interception models (Chason et al., 1991).
The inversion techniques are based on relationships between gap fraction and canopy geometry. The models describe the probability of interception of radiation within canopy layers. The assumption of random spatial distribution of canopy elements led to the Poisson model, which is the only model used for the LAI-2000. Canopies with clumped leaves or more regularly dispersed leaves may be described by binomial models, using the negative or the positive binomial probability functions (Neumann et al., 1989). Clumped and regular canopies were also described by the Markov model of Nilson (1971). All the models require statistical information on the distribution of leaf angles and azimuths within the canopy, and the binomial and Markov models additionally require a parameter to describe the orderliness of the canopy (Neumann et al., 1989). Given these inputs and the solar elevation, the models can statistically estimate the solar radiation regime within the canopy if the LAI is given, or they may be inverted and compute the LAI based on the probability of light penetration. In essence hemispherical photographs produce a hemisphere of directions on a plane (Rich, 1990). The exact nature of the projection varies according to the lens that is used. The simplest and most common hemispherical lens geometry is the polar or equiangular projection (Fig. 2) (Frazer et al., 1997). A polar projection assumes that the zenith angle of an object in the sky is directly proportional to the distance along a radial axis within the image plane. In a perfect equiangular projection of a 180° field of view, the resulting circular image shows a complete view of all sky directions, with the zenith in the center of the image and the horizons at the edges.
Fig. 2. The polar hemispherical projection (After Rich, 1990).
Already in 1924, Hill designed the first hemispherical lens to study cloud cover within a hemispherical sky. Later, architects used hemispherical photos to assess so-called site-factors that estimate the solar radiation regimes at different positions within or near buildings. Forest ecologists and foresters conceived of using photographic techniques to study light environment under forest canopies. In that context, Evans and Coombe (1959) superimposed diagrams of the sun track on hemispherical photographs to study solar radiation penetration through forest canopy openings. Anderson (1964) provided the thorough theoretical basis for using hemispherical photographs for calculation of the penetration of solar beam (direct) and scattered (diffuse or indirect) components of solar radiation from visible sky directions. Others (Wang and Miller 1987) recommended the point-drop method (Miller and Lin 1985) as calibration for the hemispherical photographs in the calibration stands.
Various authors (e.g. Wang and Miller 1987) have analyzed hemispherical photographs to obtain LAI, often using some form of automated scanning of photographs. They invariably inverted a Poisson model to obtain LAI estimates. Mussche et al. (2001) concluded after a comparative study that the exponential model for light extinction was not appropriate and created an underestimation of LAI, which could be avoided using an other light extinction model (e.g. negative binomial model). Moreover they suggested that underestimation of LAI by hemispherical photographs could also partially be due to the exposure and development of the film.
With the advent of affordable digital technologies, standard graphic image formats, and more powerful desktop computing, digital image analysis techniques have been used increasingly to examine hemispherical canopy photographs (Canham 1995). In that context, analysis of hemispherical photographs has been successfully used in a diverse range of studies to characterize plant canopy structure and light penetration, as has been investigated by several researchers (e.g. Easter and Spies 1994). Chen et al. (1997) used the methodology with success in boreal forests, whereas Dufrêne and Bréda (1995) investigated the technique in European deciduous forests. van Gardingen et al. (1999) and Comeau et al. (1998) have implemented hemispherical photography in mixed woodlands. Planchais and Pontailer (1999) compared LICOR 2000 with hemispherical photographs in beech stands and found out that both indirect techniques gave the same estimation of gap fraction at all zenith angles. However, in studies requiring fine details of the canopy structure (e.g. determining the foliage angular distributions) or the light penetration, the advantage of spatial discrimination of hemispherical photographs has been proven useful (Andrieu et al. 1994). van Gardingen et al. (1999) have improved the estimating of LAI from hemispherical images by dividing each annulus into a number of small segments. Gap fraction of each segment is calculated and the average of their logarithms is calculated for each annulus (log-average method). Comparing to destructive sampling, the log-average method was shown, to significantly reduce the underestimation of leaf area index obtained from analysis of hemispherical images of clumped canopies. Conventional analysis of hemispherical photographs resulted in an underestimate of 50% compared to the destructive harvest, while the segmented analysis reduced this to 15%.
The LAI estimated from hemispherical photographs is sensitive to photographic exposure (Chen et al. 1991), but indicated exposure may vary among cameras and light meters (e.g. Wagner, 1998) and exposure may be metered either outside or below the canopy by different operators. The extent to which the photographs should be overexposed depends on the relative contribution of the sky and the canopy to the solid angle of the hemisphere and on the internal light meter of the camera. Exposure is the amount of light acting on the emulsion of the film and is determined by the lens aperture (f-number and shutter speed). Built-in light camera meters measure the illuminance of the subject being photographed and the camera calculates ‘automatic’ exposure settings assuming the light comes from a mid-gray surface (18% visible reflectivity) by converting to photographic exposure. A change of exposure value EVR represents a halving or doubling of the amount light reaching the film. Therefore to make an unobscured overcast sky (18% visible reflectivity) completely white (100% visible reflectivity) should require 2.5 stops of overexposure. The complete white sky is needed in order to allow a more accurate thresholding for the binarization of the image. The new advanced cameras however have more complex light programs. Chen et al. (1991) investigated this influence of exposure settings (shutter speed and lens aperture) and concluded that hemispherical photography can be a more accurate method to determine LAIeff in comparison with the LAI-2000, when the right exposure is achieved. They suggested 1-2 stops of overexposure relative to the automatic exposure metered outside the canopy should produce this outcome.
Furthermore, when traditional analogue hemispherical photography is used to determine LAI, a special problem apart from the time-consuming process arises, caused by the limited dynamic range. As such, camera exposure settings have a major impact on the LAI measurements and are a major cause of measurements errors as demonstrated by Chen et al. (1991). Moreover, the low dynamic range causes difficulties in distinguishing sunlit leaves from relative small, underexposed gaps in the canopy. The use of a digital camera would overcome some of these technical problems, mainly those concerning the development of the photographic film. Traditionally, hemispherical canopy photography has relied upon conventional black and white, or color films (negatives or diapositives), and CCD-scanners to produce digital images for analysis (Frazer et al. 1997). Today, however, high-resolution (up to 6 million pixels) digital cameras offer forest scientists a practical alternative to traditional film photography (Frazer et al. 2001). These new devices offer some advantages: (1) digital images make the expenses and time associated with photographic film and development, and scanning unnecessary and thereby eliminate errors that may occur during this procedure; (2) the potential of real time processing. Also the image procession and data extraction can occur directly in the field, thus creating a more streamlined process; and finally (3) the unlimited image treatment possibilities.
One of the main problems cited in literature of hemispherical photography for determination of LAI is the selection of the optimal brightness threshold in order to distinguish leaf area from sky area thus producing a binary image. Recently, negative color images taken by video and digital camera were often used for the hemispherical photographs. Previous research demonstrated that with a high resolution of a digital camera, the choice of the threshold level would be less critical because the frequency of mixed pixels is reduced in comparison to the aggregation of pixels in cameras with lower resolution (Blennow, 1995). In relation to analogue cameras, these digital sensors have better radiometric image quality (linear response, greater dynamic range, wider spectral sensitivity range. The dynamic range is the range of discrete brightness levels an imaging system can distinguish. A normal photographic film generally does not provide a dynamic range of much more than 6 bits. A commercial consumer-priced digital camera offers a dynamic rate of 8 bits (256 levels; e.g. Nikon Coolpix 950, Nikon, Japan). Englund et al. (2000) evaluated the difference between digital and film hemispherical photography in measuring forest light environments and concluded that digital photography was effective and more convenient and inexpensive than film cameras, but they recommended caution when comparisons are made between the two techniques. Frazer et al. (2001) investigated both types of cameras for analysis of forest canopy gap structure and light transmission and found out that digital and film measures were correlated better under more open canopies as well as under overcast sky conditions. Moreover, digital photographs were extremely difficult to threshold, and no single color plane seemed to improve the contrast between sky and canopy elements. He worked with an 8-bit digital camera (Nikon Coolpix 950, Nikon Inc, Japan) and the sharpness of the digital image was generally poor compared to the film. So digital imaging provides several advantages over film-based imaging: economical processing, high resolution, rapid-product turn-around, and high dynamic range, but nevertheless the intended application and use of the photographs must be carefully considered before selecting a photo system for hemispherical photography. A professional digital sensor characterized by a high dynamic range can offer 12-16 bits (e.g. Kodak DCS660, Kodak, USA). It would improve the separability between vegetation elements and sky. Modern photographic film, filters, and digital image enhancement technologies offer remarkable opportunity to improve hemispherical image quality and contrast. These improvements in turn would facilitate a higher success rate in the classification of sky and non-sky pixels during the threshold process. The potential for digital image enhancement is increased using true-colour images because combinations of techniques can be applied to any one or all of the three RGB planes. Image enhancement methods include the application of a) digital filters to mathematically recombine neighbouring pixels, b) overlays to splice multiple RGB planes or even separate images and c) tools that modify the frequency and magnitude of pixel spectra.
Leaf Area Index is an important measure of canopy structure because tree morphology, leaf orientation and distribution influence LAI estimates. Trees of different species can have therefore very different LAI values. Clumping of needles or leaves affect LAI estimation in conifer species and to a lesser extent in deciduous species and seems to be the main cause of errors in the LAI estimation.
Hemispherical photography, a technique used in the scope of indirect methods, has proven to be a powerful indirect method for measuring various components of canopy structure and understory light regime. Numerous advances in hemispherical analysis have taken place over the last decade, which are directly related to evolving computer, photographic, and digital technologies and scientific modeling methods. Hemispherical photographs can be archived, reprocessed when improved models become available and used to perform other measurements, for example architecture and light regime below the canopy. Further testing and defining of a standardized field protocol for digital hemispherical photography are however needed.
Anderson, M.C., 1964. Studies of the woodland light climate, I, The photographic computation of light condition. J. Ecol., 52: 27-41.
Andrieu, B., Sohbi, Y., Ivanov, N., 1994. A Direct Method to Measure Bidirectional Gap Fraction in Vegetation Canopies. Remote Sens. Environ., 50: 61-66.
Blennow, K., 1995. Sky view factors from high-resolution scanned fish-eye lens photographic negatives. J Atmos. Ocean. Tech., 12: 1357-1362.
Canham, C. D., 1995. GLI/C: Software for calculation of light transmission through forest canopies using color fisheye photography. Unpublished manuscript. Institute of Ecosystem Studies, Millbrook, NY. 7p
Chason, J. W., Baldocchi, D. D., Huston, M. A., 1991. Comparison of Direct and Indirect Methods for Estimating Forest Canopy Leaf-Area. Agric. For. Meteorol., 57: 107-128.
Chen, J. M., 1996. Optically-Based Methods for Measuring Seasonal Variation of Leaf Area Index in Boreal Conifer Stands. Agric. For. Meteorol., 80: 135-163.
Chen, J. M., Black, T. A., 1991. Measuring Leaf-Area Index of Plant Canopies With Branch Architecture. Agric. For. Meteorol., 57: 1-12.
Chen, J. M., T. A. Black, Adams, R. S., 1991. Evaluation of hemispherical photography for determining plant area index and geometry of a forest stand. Agric. For. Meteorol., 56: 129-143.
Chen, J. M., Black, T. A., 1992. Defining Leaf-Area Index for Non-Flat Leaves. Plant Cell. Environ., 15: 421-429.
Chen, J. M., Rich, P. M., Gower, S. T., Norman, J. M., Plummer, S., 1997. Leaf Area Index of Boreal Forests: Theory, Techniques, and Measurements. J. Geophys. Res.-Atmos., 102: 29429-29443.
Comeau, P. G., Gendron, F., Letchford, T. A., 1998. Comparison of Several Methods for Estimating Light Under a Paper Birch Mixed wood Stand. Can. J. For. Res., 28 (12): 1843-1850.
Dufrêne, E., Bréda, N., 1995. Estimation of Deciduous Forest Leaf-Area Index Using Direct and Indirect Methods. Oecologia, 104: 156-162.
Englund, S. R., O’Brien, J. J., Clark, D. B., 2000. Evaluation of digital and film hemispherical photography for predicting understorey light in a Bornean tropical rain forest. Agric. For. Meteorol. 97, 129-139.
Easter, M. J., Spies, T. A., 1994. Using hemispherical photography for estimating photosynthetic photon flux density under canopies and in gaps in Douglas-fir forests of the Pacific Northwest. Can. J. Forest Res., 24: 2050-2058.
Evans, G. C.K, Coombe, D. E., 1959. Hemispherical and woodland canopy photography and the light climate, J. Ecol., 47: 103-113.
Fassnacht, K.S., Gower, S.T., Norman, J.M., McMurtrie, R.E., 1994. A comparison of optical and direct methods for estimating foliage surface area index in forests. Agric. For. Meteorol., 71: 183-207.
Frazer, G. W., Trofymow, J. A., Lertzman, K. P., 1997. A method for estimating canopy openness, effective leaf area index, and photosynthetically active photon flux density using hemispherical photography and computerized image analysis techniques. K.P. Natural Resources Canada, Canadian Forest Service, Pacific Forestry Centre, Victoria, BC. Information Report BC-X-373: 73 p.
Frazer, G. W.; Fournier, R. A., Trofymow, J. A., Hall, R. J. A., 2001. Comparison of Digital and Film Fisheye Photography for Analysis of Forest Canopy Structure and Gap Light Transmission. Agric. For. Meteorol., 109: 249-263.
Miller, D. H., Lin, J. D., 1985. Canopy architecture of a red maple edge stand measured by a point-drop method. IN: The Forest-Atmosphere Interaction. B.A. Hutchison and B.B. Hicks (eds) Reidel, Boston: 59-70.
Mussche, S., Samson, R., Nachtergale, L., De Schrijver, A., Lemeur, R., Lust, N. A., 2001. Comparison of Optical and Direct Methods for Monitoring the Seasonal Dynamics of Leaf Area Index in Deciduous Forests. Silva Fennica, 35: 373-384.
Neumann, H. H., Den Hartog, G. D., Shaw, R. H., 1989. Leaf-Area Measurements Based on Hemispheric Photographs and Leaf-Litter Collection in a Deciduous Forest During Autumn Leaf-Fall. Agric. For. Meteorol., 45: 325-345.
Nilson, T., 1971. A theoretical analysis of the frequency gaps in plant stands. Agric. For. Meteorol., 8: 25-28.
Rich, P. M., 1990.Characterizing Plant canopies with hemispherical photographs. Remote Sensing Reviews, 5: 13-29.
van Gardingen, P. R., Jackson, G. E., Hernandez-Daumas, S., Russell, G., Sharp, L., 1999. Leaf Area Index Estimates Obtained for Clumped Canopies Using Hemispherical Photography. Agric. For. Meteorol.; 94: 243-257.
Wagner, S., 1998. Calibration of Grey Values of Hemispherical Photographs for Image Analysis Agric. For. Meteorol., 90: 103-117.
Wang, Y. S., Miller, D. R., 1987. Calibration of the hemispherical photographic technique to measure leaf area index distributions in hardwood forests. Forest Sci., 33: 210-216.
Watson, D. J., 1947. Comparative physiological studies in the growth of field crops. I: Variation in net assimilation rate and leaf area between species and varieties, and within and between years. Ann. Bot., 11: 41-76.
Email: [email protected]
|
[1] Department of Land
Management, Katholieke Universiteit Leuven, Vital Decosterstraat 102, 3000
Leuven, Belgium. Tel: +32-16-329749; Fax: +32-16329760; |