James R. Freemantle

Mapping Forest Background Reflectance in a Boreal Region Using Multiangle Compact Airborne Spectrographic Imager Data

Forest background, consisting of understory, moss, litter, and soil, contributes significantly to optical remote sensing signals from forests in the boreal region. In this paper, we present results of background reflectance retrieval from multiangle high-resolution Compact Airborne Spectrographic Imager sensor data over a boreal forest area near Sudbury, ON, Canada. Modifications of the background by white and black plastic sheets at two sites provide two extreme limits for the development and testing of an algorithm for retrieving the background information from multiangle data. Measured background reflectances in red and near-infrared bands at six sites in the vicinity of these modified sites are used to validate the algorithm. We also explore the effect of uncertainties in the input forest structural parameters on this retrieval. The results document: 1) capability of the algorithm to retrieve meaningful background reflectance values for various forest stand conditions, particularly in the low to intermediate canopy density range; 2) the effect of background bidirectional reflectance distribution function on retrieved values; 3) performance of the algorithm using data with different cross angle values; and 4) verification of the internal consistency of the geometric-optical 4-Scale model used. The results provide an important platform for the operational estimation of the vegetation background reflectance from the bidirectional reflections observed by the Multiangle Imaging Spectroradiometer instrument.

Improving Clumping and LAI Algorithms Based on Multiangle Airborne Imagery and Ground Measurements

Measurements at more than one angle capture the directional anisotropy of solar radiance reflected from vegetated surfaces. According to our recent research, we propose that the best two view angles for vegetation structural mapping are the following: 1) the hotspot, where the Sun and view directions coincide, and 2) the darkspot, where the sensor sees the maximum amount of vegetation structural shadows. The Normalized Difference between Hotspot and Darkspot (NDHD), an angular index generated from Compact Airborne Spectrographic Imager (CASI) data, is found to be highly correlated with the field-measured foliage clumping index. The foliage clumping index characterizes the nonrandomness in the spatial distribution pattern of leaves. It is of comparable importance as the leaf area index (LAI) for quantifying radiation interception and distribution in plant canopies, and it also affects estimated LAI mapping using remote sensing data. As the clumping index can vary considerably within a cover type, it is highly desirable to map its spatial distribution for various ecological applications. We have generated clumping index maps based on the previous algorithms and empirical relationships between field-measured ¿ and CASI-derived NDHD. Through intensive validation using field data, we demonstrate that the combination of the hotspot and darkspot reflectances has the strongest response to changes in vegetation structure. Two crown structural characteristics, namely, crown height and within-crown density, are major factors that impact the NDHD and clumping index difference between the mature and young (regrowth) coniferous forests. The study area is located near Sudbury in the northern Ontario, Canada.

A comparison of NDVI and MTVI2 for estimating LAI using CHRIS imagery: a case study in wheat

Leaf area index (LAI) is a key variable in crop growth models. The derivation of reliable LAI maps from satellite imagery would provide a means of spatially extrapolating these models. During the 2004 and 2005 growing seasons, hyperspectral data from the Compact High Resolution Imaging Spectrometer (CHRIS) sensor were acquired over a wheat crop in southern Alberta. The ability to obtain reliable LAI estimates from CHRIS data was investigated using a preexisting relationship between LAI and the modified transformed vegetation index (MTVI2). The performance of the recently develop MTVI2 in estimating LAI was compared to that of the more commonly used normalized difference vegetation index (NDVI). Both narrow and broad bands simulated from the CHRIS data were used in calculating the NDVI. All vegetation indices provided good relationships with LAI (R2 = 0.70–0.90). The errors in the estimated LAI values varied with vegetation index and were dependent on growth stage. The MTVI2 performed better than the NDVI at full canopy closure. Confirming previous reports in the literature, the NDVI tended to saturate at LAI values of 37–4, which resulted in an underestimation of LAI at full canopy closure. There was no benefit to using narrow spectral bands in the NDVI. Estimation of LAI using the MTVI2 was underestimated late in the season during the seed-filling period.

Boreal forest mapping at the BOREAS study area using seasonal optical indices sensitive to plant pigment content

In this study, forest vegetation classification was investigated based on seasonal variation of pigments as inferred from visible and near-infrared spectral bands. This analysis was carried out on data collected over the southern study area of the Boreal Ecosystem-Atmosphere Study (BOREAS) with the Compact Airborne Spectrographic Imager (casi) in May and July 1994 and with the medium-resolution imaging spectrometer (MERIS) in May and August 2003. Three modified normalized difference vegetation indices (mNDVIs), which are sensitive to relative proportions among pigments and pigment content, and a red-edge spectral parameter, the wavelength at the reflectance minimum (λ0), were used. Accuracy assessments of the derived land cover maps were performed using a forest inventory map provided by the Saskatchewan Environment and Resource Management Forestry Branch Inventory Unit (SERM–FBIU). The forest vegetation classification using seasonal optical indices (mNDVIs and λ0), as derived from the casi data in May and July, shows an overall accuracy of 84% for all cover types identified, namely dry conifer, wet conifer, mixed stands, aspen, fen, and the disturbed and regenerated area. The classification results also demonstrate that classification using reflectance parameters sensitive to pigment absorption features outperforms that using the reflectance itself. In addition, the classification using seasonal information is better than that using information obtained from a single date, and the spatial patterns were consistent with those achieved using multidate MERIS imagery. The forest vegetation classification using seasonal changes in optical indices (mNDVIs and λ0) derived from the MERIS imagery in May and August revealed a reasonably high overall classification accuracy (72%) for all vegetation cover types identified, namely conifer, mixed stands, and fen.

Vegetation classification using hyperspectral and multi-angular remote sensing data

In this study, vegetation cover type classification was investigated using CHRIS data over agricultural scenes acquired across the 2004 growing season. Spectral indices sensitive to crop chlorophyll content and leaf area index were first calculated from CHRIS nadir data in May, June and July. The seasonality of these indices was analyzed and employed to identify crop types in the study area. To further improve the classification accuracy, the angular signatures of the vegetation canopies were derived from the nadir and off- nadir data in the red and near-infrared band using the kernel-driven Ross_Thick and Li-Sparse model. The coefficients of the kernel-based model were then used for crop type classification, together with the spectral indices derived from nadir data. Preliminary results show that the additional angular information can slightly improve the classification accuracy.

Remote Sensing Derived Leaf Area Index and Potential Applications for Crop Modeling

Integration of meteorological and remote sensing data in crop growth models offers a potentially powerful tool for yield monitoring. Leaf area index (LAI) is a key variable in crop growth models. The derivation of reliable LAI maps from satellite imagery would provide a means of spatially extrapolating these models. As part of a two-year project to develop an intelligent sensorweb system for yield prediction in agricultural crops and rangeland, the ability to obtain reliable LAI estimates from Compact High Resolution Imaging Spectrometer (CHRIS) data was investigated. Throughout the 2004 and 2005 growing season, data from the CHRIS sensor were acquired over two contrasting sites in Alberta, a wheat crop and rangeland. The modified triangular vegetation index (MTVI2) was used to derive LAI values which were compared to ground- based LAI data collected weekly or tri-weekly in wheat and monthly on the rangeland. A strong relationship was observed between ground-based and remote sensing derived LAI in the case of wheat (r=0.91-0.93). In, the rangeland, where senescent vegetation is a confounding factor, LAI was consistently overestimated using the CHRIS imagery.