Peter R. J. North

Three-dimensional forest light interaction model using a Monte Carlo method

A model for light interaction with forest canopies is presented, based on Monte Carlo simulation of photon transport. A hybrid representation is used to model the discontinuous nature of the forest canopy. Large scale structure is represented by geometric primitives defining shapes and positions of the tree crowns and trunks. Foliage is represented within crowns by volume-averaged parameters describing the structural and optical properties of the scattering elements. Simulation of three-dimensional photon trajectories allows accurate evaluation of multiple scattering within crowns, and between distinct crowns, trunks and the ground surface. The sky radiance field is treated as anisotropic and decoupled from bidirectional reflectance calculation. Validation has been performed on an example of dense spruce forest. Results show close agreement between model predictions and field measurements of bidirectional reflectance, high-resolution spectra and hemispherical albedo.

Amazon forests maintain consistent canopy structure and greenness during the dry season

The seasonality of sunlight and rainfall regulates net primary production in tropical forests. Previous studies have suggested that light is more limiting than water for tropical forest productivity, consistent with greening of Amazon forests during the dry season in satellite data. We evaluated four potential mechanisms for the seasonal green-up phenomenon, including increases in leaf area or leaf reflectance, using a sophisticated radiative transfer model and independent satellite observations from lidar and optical sensors. Here we show that the apparent green up of Amazon forests in optical remote sensing data resulted from seasonal changes in near-infrared reflectance, an artefact of variations in sun-sensor geometry. Correcting this bidirectional reflectance effect eliminated seasonal changes in surface reflectance, consistent with independent lidar observations and model simulations with unchanging canopy properties. The stability of Amazon forest structure and reflectance over seasonal timescales challenges the paradigm of light-limited net primary production in Amazon forests and enhanced forest growth during drought conditions. Correcting optical remote sensing data for artefacts of sun-sensor geometry is essential to isolate the response of global vegetation to seasonal and interannual climate variability.

Radiation transfer model intercomparison (RAMI) exercise

The community involved in modeling radiation transfer over terrestrial surfaces designed and implemented the first phase of a radiation transfer model intercomparison (RAMI) exercise. This paper discusses the rationale and motivation for this endeavor, presents the intercomparison protocol as well as the evaluation procedures, and describes the principal results. Participants were asked to simulate the transfer of radiation for a variety of precisely defined terrestrial environments and illumination conditions. These were abstractions of typical terrestrial systems and included both homogeneous and heterogeneous scenes. The differences between the results generated by eight different models, including both one-dimensional and three-dimensional approaches, were then documented and analyzed. RAMI proposed a protocol to quantitatively assess the consequences of the model discrepancies with respect to application, such as those motivating the development of physically based inversion procedures. This first phase of model intercomparison has already proved useful in assessing the ability of the modeling community to generate similar radiation fields despite the large panoply of models that were tested. A detailed analysis of the results also permitted to identify apparent “outliers” and their main deficiencies. Future undertakings in this intercomparison framework must be oriented toward an expansion of RAMI into other and more complex geophysical systems as well as the focusing on actual inverse problems.

Vegetation height estimates for a mixed temperate forest using satellite laser altimetry

Data from the Geoscience Laser Altimeter System (GLAS) aboard the Ice Cloud and land Elevation Satellite (ICESat) offer an unprecedented opportunity for canopy height retrieval at a regional to global scale. The data also provide useful information for forest stand level assessment at coincident locations. In this study height indices from light detection and ranging (LiDAR) waveforms were explored as a means of extracting canopy height; these were examined with reference to a mixed temperate forest in Gloucestershire, UK, containing planted stands with a mean age of 51 years and mean maximum height of 26.6 m. A method based on using a terrain index (TI; maximum minus minimum elevations from a 7×7 subset 10‐m resolution digital terrain model (DTM)) to adjust the waveform extent (WE; signal begin minus signal end) produced an R 2 value of 0.89 when regressed against field measurements of maximum canopy height at footprint locations (field height = 0.91(WE−TI)+4.86; root mean squared error (RMSE) = 2.99 m, coefficient significance p<0.001, intercept significance p>0.01). Multiple regression performed on both WE and TI with field measurements produced an R 2 of 0.90 and an RMSE of 2.86 m (field height = 1.0208WE−0.7310TI; coefficient significance p<0.001, intercept not significant). Maximum canopy height estimates using an automated approach to ground return identification based on iterative fitting of Gaussian peaks (GP1_2MAXAMP) to the waveform explained 74% of variance when compared to field measurements (field height = 1.05(GP1_2MAXAMP); RMSE = 4.53 m, coefficient significance p<0.001, intercept not significant). The ability of satellite LiDAR to retrieve data for such a complex and diverse area further indicates the potential of this technique for both carbon accounting and forest management.

The Propagation of Foliar Biochemical Absorption Features in Forest Canopy Reflectance

Remotely sensed estimates of the foliar biochemical content of vegetation canopies could be used to derive indicators of ecosystem functioning at regional to global scales. In the past decade, a number of studies have reported strong correlations between the reflectance spectra of vegetation canopies and their foliar biochemical content. However, these studies have commonly employed multiple regression techniques or spectral indices to determine biochemical content, which have been found to be highly sensitive to variation in canopy architecture [such as leaf area index (LAI) and canopy closure] and understory. To date, these effects combined with the low signal-to-noise ratios of airborne spectrometers have inhibited the development of robust and portable spectral techniques for the estimation of canopy biochemical content. This paper reports on a theoretical study in which a leaf model, LIBERTY (leaf incorporating biochemicals exhibiting reflectance and transmittance yields), characterized specifically for conifer needles, was coupled with a hybrid geometric/radiative transfer bidirectional reflectance distribution function FLIGHT (forest light) model. By varying leaf biochemical content, LAI, canopy closure and understory, we analyzed the simulated canopy reflectance spectra to determine if the biochemical absorption features in leaf spectra were preserved at the canopy scale. Absorption features or wavelength regions that were both related to a specific biochemical of interest (water, lignin-cellulose) and persistent at the scale of both the leaf and the canopy were identified at a number of wavelengths or wavelength regions.

Retrieval of land surface bidirectional reflectance and aerosol opacity from ATSR-2 multiangle imagery

New satellite instruments that sample top-of-atmosphere radiance at a number of view angles offer the potential for improved retrieval of atmospheric aerosol opacity, land surface bidirectional reflectance, and biophysical parameters. This paper presents a method for simultaneous retrieval of aerosol opacity and land surface bidirectional reflectance, which utilizes the dual view capability of the second Along-Track Scanning Radiometer (ATSR-2). Analysis of a physically based model of light scattering results in two simple equations defining possible spectral variation of land surface bidirectional reflectance distribution function (BRDF). These are used as constraints to anew inversion of a model of atmospheric scattering to simultaneously retrieve atmospheric aerosol opacity and bidirectional reflectance from top-of-atmosphere radiance. The inversion assumes no a priori knowledge of the land surface cover. Sensitivity is evaluated using both simulated and field-measured data to reproduce expected ATSR-2 observations. Where an atmosphere of known aerosol scattering properties, but of unknown optical depth, is available, results show mean absolute error in retrieval of aerosol opacity of the greater of 0.02 or 15% relative error and bidirectional reflectance retrieval at 55 nm to an accuracy of <0.01. Where a number of candidate aerosol models are available, results show discrimination of dominant aerosol type is possible in 95% of cases considered. The methods perform best over dark surfaces, such as vegetation, but show accurate retrieval over soil and pixels containing a number of cover types.

Monte Carlo ray tracing in optical canopy reflectance modelling

This paper reviews the use of Monte Carlo (MC) methods in optical canopy reflectance modelling. Their utility, and, more specifically, MC ray tracing for the numerical simulation of the radiation field within a vegetation canopy, are outlined. General issues pertinent to implementation and exploitation of such methods are discussed, such as the descriptions of canopy structure and radiometric properties required for their use. Strategies for the reduction of variance, which form the core of the application of MC methods to canopy reflectance modelling are presented, and examples given of the type of information which may be obtained from canopy reflectance modelling using MC ray tracing. The use of MC methods in the development of models of canopy development, driven by fundamental properties such as radiation interception are discussed.

Estimation of fAPAR, LAI, and vegetation fractional cover from ATSR-2 imagery

Abstract We examine methodologies for estimation of vegetation cover, leaf area index (LAI), and fraction of absorbed photosynthetically active radiation ( f APAR ), considering the spectral sampling and dual-view capability of the ATSR-2 sensor. A set of simulated ATSR-2 reflectance measurements and corresponding vegetation parameters is defined using a Monte Carlo ray-tracing model. The case of semiarid vegetation is considered allowing for varying fractional cover, structure, and presence of standing litter. The error in estimation of vegetation properties using vegetation indices, linear spectral unmixing, and model inversion is compared over this dataset, quantified by a measure of signal to noise (S/N). For the estimation of f APAR , the NDVI gave best S/N among vegetation indices (S/N 4.5). Linear mixture modelling based on library spectra showed considerable improvement over vegetation indices for estimation of total vegetation cover. LAI is not retrieved with much accuracy by any method in the presence of standing litter and variable fractional cover. Model inversion has potential to be the most accurate method for retrieving all parameters, but only if the model approximates reality within 15%. Overall, the S/N in estimating parameters by any method is considerably lower than the S/N in instrument calibration (20/1). Use of the dual-view showed potential to improve estimates, but requires accurate registration.

Analyzing the effect of structural variability and canopy gaps on forest BRDF using a geometric-optical model

Abstract A geometric-optical reflectance model is presented which estimates the bidirectional reflectance distribution function (BRDF) of forest canopies by modeling four shadowing pattern components (i.e., illuminated crown, illuminated ground, shadowed crown, and shadowed ground. The model represents the forest canopy as a group of discrete tree crowns and uses a deterministic ray tracing procedure to generate a two-dinwnsional scene of shadow pattern components. It provides the type of shadow fraction information used to calculate the reflectance of forests when it is treated as an area-weighted sum of the shadow component reflectances. The model is designed to test the effect of canopy structure parameters on the bidirectional reflectance of forests and can easily be adapted to accommodate different crown shapes, crown sizes, stem densities, and distributions. The model was used to isolate and quantify the effects of crown shape, canopy cover, stein distribution pattern, and canopy gaps on the shadowing pattern and the canopy reflectance, using a set of statistically generated forest scenes and data gathered for a natural tropical lowland forest scene. The illuminated crown component and the illuminated ground component have the greatest impact on the reflectance values in the red and near-infrared region respectively. The red BRDF, mainly influenced by the illuminated ground component, proves to be sensitive to canopy cover, tree pattern distribution, and canopy gaps. The near-infrared BRDF, affected by the illuminated crown component, is sensitive to canopy cover, crown-center height distribution, and crown shape. The sensitivity of the BRDF to canopy gaps and crown clumping was investigated in more detail. Reflectance-measurements close to the hotspot are found to be insensitive to clumping while retaining sensitivity to crown cover. Forward scattering measurements are fcntnd to be more sensitive to gap size and frequency. The spatial effects of plantations, logging, or road features on the BRDF is further analyzed.

Aerosol optical depth and land surface reflectance from multiangle AATSR measurements: global validation and intersensor comparisons

This paper presents the results and satellite intercomparisons for the retrieval of aerosol optical depth (AOD) and land surface bidirectional reflectance using the Multiangle Advanced Along-Track Scanning Radiometer (AATSR). The algorithm developed is based on inversion of a physical model of light scattering that requires no a priori knowledge of the land surface. The model is evaluated for a number of sites around the world to test its operation over a range of aerosol types and land covers including dark and bright surfaces. Validation is performed using Aerosol Robotic Network ground-based sun-photometer measurements and by intercomparison with independent estimates of AOD derived from spaceborne instruments including Multiangle Imaging Spectroradiometer (MISR), Moderate Resolution Imaging Spectroradiometer (MODIS), and Total Ozone Mapping Spectrometer (TOMS) aerosol products. Results show good agreement (Pearsons correlation coefficient r2=0.70 for all sites combined) between the AATSR-derived estimates of AOD and the sun-photometer measurements. There is also a high correlation (r2=0.84) between the AATSR- and MISR-derived AOD estimates, but the correlations of the AATSR-derived AOD with MODIS-derived AOD and TOMS aerosol index are lower. In addition, the ability of the sensor to discriminate between different aerosol types is evaluated. Moreover, the estimates of the aerosol properties are used for atmospheric correction of the top-of-atmosphere reflectance. The AATSR surface reflectances are compared with the MODIS bidirectional reflectance distribution function/Albedo and MISR surface products and are shown to correspond with root-mean-square errors of 0.03 and 0.06 or better, respectively. The retrieval method is applied on an image basis resulting in an image of surface reflectance and a separate map of AOD. A map of AOD at 550 nm covering the Sahel and southern Sahara region is presented to demonstrate operation at regional and potentially global scales