The use of shadows in high spatial resolution, remotely sensed, imagery to estimate the height of individual Eucalyptus trees on undulating land
The shadows cast by 180 individual Eucalyptus trees, of varying canopy condition, on undulating land in south-eastern Australia were used to infer their heights from 50-cm spatial resolution, multispectral aerial imagery (blue = 0.4-0.5 μm; green = 0.5-0.6 μm; red = 0.6-0.7 μm; near infrared = 0.7-1 μm). A geometrical shadow model was developed incorporating the local slope and aspect of the ground from a digital elevation model at each tree location. A method of deriving 'local tree time' to infer the solar elevation angle, in situations where the image acquisition time is not available, was also developed. Based on a measurement of the shadow length from the geometric centre of the tree crowns, and ignoring the role of the crown periphery in distorting the shadow shape, the tree heights were estimated with a root mean square error of ±5.6m (±27%) with some overestimated by as much as 50%. A geometric correction for shadow distortion assuming spherical crown geometry provided an improved estimate with a root mean square error of ±4.8m (±23%).
Comparison of Canopy Volume Measurements of Scattered Eucalypt Farm Trees Derived from High Spatial Resolution Imagery and LiDAR
Studies estimating canopy volume are mostly based on laborious and time-consuming field measurements; hence, there is a need for easier and convenient means of estimation. Accordingly, this study investigated the use of remotely sensed data (WorldView-2 and LiDAR) for estimating tree height, canopy height and crown diameter, which were then used to infer the canopy volume of remnant eucalypt trees at the Newholme/Kirby 'SMART' farm in north-east New South Wales. A regression model was developed with field measurements, which was then applied to remote-sensing-based measurements. LiDAR estimates of tree dimensions were generally lower than the field measurements (e.g., 6.5% for tree height) although some of the parameters (such as tree height) may also be overestimated by the clinometer/rangefinder protocols used. The WorldView-2 results showed both crown projected area and crown diameter to be strongly correlated to canopy volume, and that crown diameter yielded better results (Root Mean Square Error RMSE 31%) than crown projected area (RMSE 42%). Although the better performance of LiDAR in the vertical dimension cannot be dismissed, as suggested by results obtained from this study and also similar studies conducted with LiDAR data for tree parameter measurements, the high price and complexity associated with the acquisition and processing of LiDAR datasets mean that the technology is beyond the reach of many applications. Therefore, given the need for easier and convenient means of tree parameters estimation, this study filled a gap and successfully used 2D multispectralWorldView-2 data for 3D canopy volume estimation with satisfactory results compared to LiDAR-based estimation. The result obtained from this study highlights the usefulness of high resolution data for canopy volume estimations at different locations as a possible alternative to existing methods.
A Comparative Study of Land Cover Classification Techniques for "Farmscapes" Using Very High Resolution Remotely Sensed Data
High spatial resolution images (~10 cm) are routinely available from airborne platforms. Few studies have examined the applicability of using such data to characterize land cover in "farmscapes" comprising open pasture and remnant vegetation communities of varying density. Very high spatial resolution remotely sensed imagery has been used to classify land cover classes on a ~5000 ha extensive grazing farm in Australia. This "farmscape" consisted of open pasture fields, scattered trees, and remnant vegetation (woodlands). The relative performances of object-based and pixel-based approaches to classification were tested for accuracy and applicability. Maximum likelihood classification (MLC) was used for pixel-based classification while the k-nearest neighbor (k-NN) technique was used for object-based classification. A range of image sampling scales was tested for image segmentation. At an optimal sampling scale, the pixel-based classification resulted in an overall accuracy of 77 percent, while the object-based classification achieved an overall accuracy of 86 percent. While both the object- and pixel-based classification techniques yielded higher quantitative accuracies, a "more realistic" land cover classification, with few errors due to intermixing of similar classes, was achieved using the object-based method.
Evaluating remote sensing technologies for improved yield forecasting and for the measurement of foliar nitrogen concentration in sugarcane
AN ANALYSIS OF time series Landsat imagery acquired over the Bundaberg region between 2010 and 2015 identified variations in annual crop vigour trends, as determined by greenness normalised difference vegetation index (GNDVI). On average, early to mid-April was identified as the crucial period where crops achieved their maximum vigour and as such indicated when single image captures should be acquired for future regional yield forecasting. Additionally, the regional crop GNDVI averaged from Landsat images between February to April, produced a higher coefficient of determination to final yield (R2 = 0.91) than the average crop GNDVI value from a single mid-season SPOT5 image capture (R2 = 0.52). This result indicates that the time series method may be more appropriate for future regional yield forecasting. For improved prediction accuracies at the individual crop level, a univariate model using only crop GNDVI values (SPOT5) and corresponding yield (t/ha) produced a higher prediction accuracy for the 2014 Bundaberg harvest than a multivariate model that included additional historic spectral and crop attribute data. For Condong, a multivariate model improved the prediction accuracy of individual crops harvested in 2014 by 41.8% for one-year-old cane (Y1), and 46.2% for two-year-old cane (Y2). For the non-invasive measure of foliar nitrogen (N%), the specific wavelengths 615 nm, 737 nm and 933 nm (Airborne hyperspectral), and 634 nm, 750 nm and 880 nm (ground based field spectroscopy) were found to be the most significant. These results were supported by satellite imagery (Worldview-2 and Worldview-3) acquired over three replicated field trials in Mackay (2014 and 2015) and Tully (2015), where the vegetation index (VI) REN2NDVIWV, a ratio of the rededge band (705-745 nm) and the Near-IR2 band (860-1040 nm), produced a higher correlation to nitrogen concentration (%) than NDVI.
An allometric model for estimating DBH of isolated and clustered Eucalyptus trees from measurements of crown projection area
Owing to its relevance to remotely-sensed imagery of landscapes, this paper investigates the ability to infer diameter at breast height (DBH) for five species of Australian native 'Eucalyptus' from measurements of tree height and crown projection area. In this study regression models were developed for both single trees and clusters from 2 to 27 stems (maximum density 536 stems per ha) of 'Eucalyptus bridgesiana', 'Eucalyptus caliginosa', 'Eucalyptus blakelyi', 'Eucalyptus viminalis', and 'Eucalyptus melliodora'. Crown projection area and tree height were strongly correlated for single trees, and the log-transformed crown projection area explained the most variance in DBH (R² = 0.68, mean prediction error ±16 cm). Including tree height as a descriptor did not significantly alter the model performance and is a viable alternative to using crown projection area. The total crown projection area of tree clusters explained only 34% of the variance in the total (sum of) the DBH within the clusters. However average crown projection area per stem of entire tree clusters explained 67% of the variance in the average (per stem) DBH of the constituent trees with a mean prediction error ±8 cm. Both the single tree and tree cluster models were statistically similar and a combined model to predict average stem DBH yielded R² = 0.71 with a mean prediction error (average DBH per stem) of ±13 cm within the range of 0.28-0.84 m. A single model to infer DBH for both single trees and clusters comprising up to 27 stems offers a pathway for using remote sensing to infer DBH provided a means of determining the number of stems within cluster boundaries is included.