Mapping and monitoring forest extent is a common requirement of regional forest inventories and public land natural resource management, including in Australia. The state of Victoria, Australia, has approximately 7.2 million hectares of mostly forested public land, comprising ecosystems that present a diverse range of forest structures, composition and condition. In this paper, we evaluate the performance of the Random Forest (RF) classifier, an ensemble learning algorithm that has recently shown promise using multi-spectral satellite sensor imagery for large area feature classification. The RF algorithm was applied using selected Landsat Thematic Mapper (TM) imagery metrics and auxiliary terrain and climatic variables, while the reference data was manually extracted from systematically distributed plots of sample aerial photography and used for training (75%) and accuracy (25%) assessment. The RF algorithm yielded an overall accuracy of 96% and a Kappa statistic of 0.91 (confidence interval (CI) 0.909-0.919) for the forest/non-forest classification model, given a Kappa maximised binary threshold value of 0.5. The area under the receiver operating characteristic plot produced a score of 0.91, also indicating high model performance. The framework described in this study contributes to the operational deployment of a robust, but affordable, program, able to collate and process OPEN ACCESSRemote Sens. 2013, 5 2839 large volumes of multi-sourced data using open-source software for the production of consistent and accurate forest cover maps across the full spectrum of Victorian sclerophyll forest types.
The presence of the aggressive, colony-forming honeyeater, Manorina melanophrys (bell miner), in the canopies of unhealthy eucalypts has been well reported. There is, however, some debate as to the actual mechanisms producing these unhealthy trees. To investigate further some of the processes that may be contributing to this form of canopy dieback, two field trials were carried out in Olney State Forest, near Wyong, New South Wales. The study site contained Eucalyptus saligna (Sydney blue gum) with canopy dieback and was occupied hy a large colony of bell miners. Close examination of the foliage revealed a large and diverse suite of phytophagous insects, including at least 16 species of psyllid (Hemiptera: Psyllidae). In the first trial, the use of bird exclusion cages over selected branches significantly improved leaf survival compared to leaves exposed to a relatively high density of bell miners. It is proposed that colonization by bell miners may interfere with the efficacy of both other insectivorous birds (through aggressive interspecific territoriality) and the invertebrate predators and parasitoids. Interference with such regulatory factors may enable some phytophagous insect populations to rise to sustained damaging levels. In the second trial, an insecticide application combined with reduced competition from the dense understorey and neighbouring trees was required to significantly improve trunk diameter and crown condition scores. After 12 months, neither treatment, by itself, significantly improved both growth measures. Possibly both treatments were required because the E. saligna trees were suffering from another source of stress (e.g. drought) in addition to the relatively high level of insect attack.
Dothistroma needle blight is a serious foliar disease in Australian Pinus radiata plantations causing defoliation, decreased productivity and, in extreme cases, tree death. Conventional methods of monitoring forest health such as aerial survey and ground assessments are labor intensive, time consuming, and subjective. Remote sensing provides a synoptic view of the canopy and can indicate areas affected by damaging agents such as pests and pathogens. Hyperspectral airborne remote sensing imagery (CASI-2) was acquired over pine stands in southern New South Wales, Australia which had been ground assessed and ranked on an individual tree basis, according to the extent of Dothistroma needle blight. A series of spectral indices were tested using two different approaches for extracting crown-scale reflectance measurements and relating these to ground-based estimates of severity. Dothistroma needle blight is most severe in the lower crown and statistically significant relationships were found between crown reflectance values and ground estimates using a 'halo' approach (which ignored each tree crown's brightest central pixels). Independent accuracy assessment of the method indicated that the technique could successfully detect three levels of Dothistroma needle blight infection with an accuracy of over 70%.
The physiological status of forest canopy foliage is influenced by a range of factors that affect leaf pigment content and function. Recently, several indices have been developed from remotely sensed data that attempt to provide robust estimates of leaf chlorophyll content. These indices have been developed from either hand-held spectroradiometer spectra or high spectral resolution (or hyperspectral) imagery. We determined if two previously published indices (Datt 1999), which were specifically developed to predict chlorophyll content in eucalypt vegetation by remote sensing at the leaf scale, can be extrapolated accurately to the canopy. We derived the two indices from hand-held spectroradiometer data of eucalypt leaves exhibiting a range of insect damage symptoms. We also derived the indices from spectra obtained from high spectral and spatial resolution Compact Airborne Spectrographic Imager 2 (CASI-2) imagery to determine if reasonable estimates at a scale of < 1 m can be achieved. One of the indices (R 850/R 710 index, where R is reflectance) derived from hand-held spectroradiometer data showed a moderate correlation with relative leaf chlorophyll content (r = 0.59, P < 0.05) for all dominant eucalypt species in the study area. The R (850)/R (710) index derived from CASI-2 imagery yielded slightly lower correlations over the entire data set (r = 0.42, P < 0.05), but correlations for individual species were high (r = 0.77, P < 0.05). A scaling analysis indicated that the R (850)/R (710) index was strongly affected by soil and water cover types when pixels were mixed, but appeared to be invariant to changes in proportions of understory, which may limit its application.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.