Estimation of Leaf Area Index in a Mountain Forest of Central Japan with a 30-m Spatial Resolution Based on Landsat Operational Land Imager Imagery: An Application of a Simple Model for Seasonal Monitoring

Irina Melnikova, Yoshio Awaya, Taku Saitoh, Hiroyuki Muraoka, Takahiro Sasai
2018 Remote Sensing  
An accurate estimation of the leaf area index (LAI) by satellite remote sensing is essential for studying the spatial variation of ecosystem structure. The goal of this study was to estimate the spatial variation of LAI over a forested catchment in a mountainous landscape (ca. 60 km 2 ) in central Japan. We used a simple model to estimate LAI using spectral reflectance by adapting the Monsi-Saeki light attenuation theory for satellite remote sensing. First, we applied the model to Landsat
more » ... el to Landsat Operational Land Imager (OLI) imagery to estimate the spatial variation of LAI in spring and summer. Second, we validated the model's performance with in situ LAI estimates at four study plots that included deciduous broadleaf, deciduous coniferous, and evergreen coniferous forest types. Pre-processing of the Landsat OLI imagery, including atmospheric correction by elevation-dependent dark object subtraction and Minnaert topographic correction, together with application of the simple model, enabled a satisfactory 30-m spatial resolution estimation of forest LAI with a maximum of 5.5 ± 0.2 for deciduous broadleaf and 5.3 ± 0.2-for evergreen coniferous forest areas. The LAI variation in May (spring) suggested an altitudinal gradient in the degree of leaf expansion, whereas the LAI variation in August (mid-summer) suggested an altitudinal gradient of yearly maximum forest foliage density. This study demonstrated the importance of an accurate estimation of fine-resolution spatial LAI variations for ecological studies in mountainous landscapes, which are characterized by complex terrain and high vegetative heterogeneity. Because leaves are the main surface through which energy and mass are exchanged between terrestrial vegetation and the atmosphere, the leaf area index (LAI) is an essential variable used to integrate ecosystem functions such as photosynthesis, transpiration, and autotrophic respiration from the scale of an individual leaf to total tree and canopy scales [4] [5] [6] [7] . LAI is an indicator of forest ecosystem structure, i.e., the geometrical structure and density of foliage. Several definitions of LAI are found in the literature [8] [9] [10] , and the definitions depend on the leaf form (flat or needle) and methods applied for the estimation (true LAI and effective LAI). LAI was first defined by Watson [8] as the total one-sided area of leaf tissue per unit ground surface area. This definition is applicable to flat leaves, both sides of which have the same area. For leaves with other geometries such as needles, Myneni et al. [10] used the term LAI to mean the maximum projected leaf area per unit ground surface area. In this paper, we use the definition of LAI by Myneni et al. [10] . An accurate estimation of the LAI of forest ecosystems is crucial for studies of the structure and function of ecosystems and is required for ecosystem simulation models [11] [12] [13] . The spatial and temporal variations of LAI are estimated by in situ [4, 14] and remote sensing approaches [6, 7, [15] [16] [17] [18] . The in situ approach consists of direct methods, such as a model tree method, litter-fall collection [4, 19, 20] , and indirect methods, such as contact methods (e.g., inclined point quadrat and allometry [14] ) and non-contact optical methods. Optical methods make use of the measurements of PAR transmission through canopies by application of the Monsi-Saeki theory [21] . The Monsi-Saeki theory describes the relationship between the amount of leaves and light penetration in a plant canopy. It states that a total amount of photosynthetically active radiation (PAR), W/m 2 , intercepted by the canopy, depends on the incident incoming amount of PAR, canopy structure (density of leaves and vertical profile), and leaf optical properties. The equation describing the relationship between PAR-transmittance and amount of leaves (i.e., LAI) of a given plant canopy (Equation (7) in this study) is derived from Beer-Lambert law for the light attenuation in a solution (liquid containing certain concentration of chemical matter). In this study, we use the term "Monsi-Saeki theory" as it is developed for a plant canopy in the original theory. The optical methods include field measurements of light attenuation through the canopy by PAR sensors (namely PAR transmittance), analysis of hemispherical photographs [2,20], or use of commercial instruments [20] . The remote sensing approach is crucial for evaluating spatial variations of LAI over areas from landscape to global scales [16] and includes empirical, semi-empirical, and physical methods. Empirical, or statistical, methods have been used to estimate the spatial variation of LAI based on statistical relationships between LAI and spectral vegetation indices (VIs) derived from passive optical remote sensing imagery. The most frequently used VIs are the normalized difference vegetation index (NDVI), the enhanced vegetation index (EVI), and the soil adjusted vegetation index (SAVI) [6, 12, 17, 18] . While the main advantage of the empirical methods is their simplicity, the passive optical sensors cannot penetrate the dense canopy and obtain the spectral reflectance of the lower canopy and understory. Thus, in high LAI values (LAI ≥ 3), spectral signals, and consequently, LAI-VIs relationships, become saturated [22] [23] [24] . In addition, VIs are dependent on many factors, such as canopy geometry, leaf and soil optical properties, and sun position [25] , which restrict the universal applicability (generality) of the empirical models. Gray and Song [24] have introduced a fine-resolution empirical method for estimating LAI that overcomes the "saturation" limitation by combining image spectral information from Landsat imagery and spatial information from high-resolution panchromatic IKONOS imagery. However, pre-processing of images from multiple sensors increases computational and data requirements, especially in mountainous areas with complex atmospheric and topographic characteristics. In particular, the presence of clouds and aerosols (atmospheric effects) and the shadows of the mountain slopes (topographic effects) lead to a requirement of atmospheric and topographic corrections of each image separately because of the sensor differences in the spectral and spatial resolutions. Above all, if high-resolution imagery is derived with a great off-nadir viewing angle, trees are projected obliquely and cannot be orthorectified without very accurate digital surface elevation models (DEMs).
doi:10.3390/rs10020179 fatcat:ikhpq3zbljbfvihpobfxciyrqq