their work in the exciting field of airborne and terrestrial laser scanning and 3D medical-site.info (accessed 17 March Keywords: LiDAR; deadwood; airborne laser scanning; terrestrial Sampling- medical-site.info (accessed on Combining Airborne and Terrestrial Laser Scanning Technologies to Measure Forest Understorey Volume. Article (PDF Available) in Forests 8(4) · April.
|Language:||English, Spanish, Dutch|
|Distribution:||Free* [*Register to download]|
Request PDF on ResearchGate | On Jan 1, , George Vosselman and others published Airborne and Terrestrial Laser Scanning. Last month we reported back from the Intergeo trade show that it was all about three abbreviations: Airborne Laser Scanning. (ALS), Terrestrial Laser Scanning . estimation, Airborne Laser Scanning (ALS) and Terrestrial Laser Scanning (TLS) have attracted much attention. We examine the relationship.
As the depth of mobilization of from scanning point clouds obtained in different seasons. The benefits and shortcomings of of the beach show a healthy concave or problematic combining the two laser scanning methods for monitoring coastal convex shape.
This contribution investigates the suitability of novel remote sensing methods like terrestrial TLS and airborne laser I. Their key driving mechanism is a highly intermittent wave regime that is accompanied by aperiodic variations of the water level, seasonal ice cover , and frequent presence of conditions favorable for unexpectedly high run-up  and set-up .
The most interesting coastal segment in this respect is the southern coast of the Gulf of Finland where the beaches mostly overlie ancient dunes in deeply indented bays that are geometrically sheltered from a large part of the directions of strong winds.
The volume of sediment and the magnitude of littoral drift are modest here and the entire coast in question generally suffers from sediment deficit ,. A specific feature of this coast is that the strongest storms tend to blow from directions from which winds are not very frequent . Thus, differently from the classical examples of bay beaches , the bayhead beaches here are only partially sheltered from intense waves.
They often reveal step-like Figure 1. The Baltic Sea and the Gulf of Finland. The box indicates the study evolution: The scanning device can be transported by a major debris flow event  or assessments mounted on different platforms, i. In particular, one of the well- ALS or a ground vehicle mobile laser scanning. The altitudes of surveying routes could reach up to about objects such as buildings, bridges, statues, road surfaces a few kilometers, thus resulting in comparatively sparse data or other structures.
Repeated ALS measurements over the an accuracy of about 1 cm or even better  and at a same area makes possible the creation of exact surface models resolution up to a few mm. However, in order to save time for different time epochs and therefore enables the detection and decrease the file sizes, the resolution is usually set at a of inter-epoch volume and mass changes.
The advantage is that the entire area within a m range. However, in practice the maximum of interest can be included in the analysis while only changes reasonable distance to a vertical surface is about 50— m along single profiles can be identified using classical methods and to a horizontal surface about 25—50 m, varying according of coastal monitoring. Thus, the ALS enables more accurate to the specific scanner and the characteristics of the surface. Therefore, using shorter ranges enables acquiring finer details.
Therefore ALS data of various vintages may already be accessible for regions of interest. Monitoring of coastal 2. Airborne laser scanning processes using ALS data has already been carried out, e. In ALS the scanning device is placed on an aircraft, e. ALS is cost- This study attempts to widen the applicability and enhance effective in applications where large areas need to be covered, the quality of standard ALS data products by using in situ TLS for example measuring ground surfaces for compilation of data for minimizing systematic errors between ALS and TLS terrain models, infrastructure objects, etc.
Generally, an ALS datasets from different epochs. It been used, for example, to quantify the sediment volume usually varies from 0. Acquiring more high-resolution data requires more . We explore the potential of the joint use of the TLS time, flight hours and therefore more funds. The goal is to characterize not only the aircraft at any instant, a Global Navigation Satellite System overall intensity of coastal processes but also to shed light to GNSS receiver and an Inertial Measurement Unit IMU are the changes in the internal structure of the beach, in particular used.
GNSS receiver records the position and IMU the pitch, to identify whether the most representative parts of the beach roll and yaw of the aircraft.
Since the calculation involves gain or lose sediment. The means of elimination of systematic m, the divergence of the laser beam is much more and random errors of the ALS and TLS, creating and pronounced and the beam can reach up to a meter in diameter comparing of the resulting Digital Terrain Models DTM of in ground surface. This means that capturing very fine details test areas from scanning point clouds are discussed next.
The is impossible and measurement errors are more likely to occur. First of all, in applications where the desired accuracy and data resolution are that of Laser scanning is a remote sensing method which utilizes characteristic to TLS, ALS does not meet the expected laser pulses to measure distances to objects.
Based on these demands. Also given the relatively large differences in spatial distances and the angles of laser beams, the coordinates of resolution, the methods for data processing have to be slightly measured points are calculated. This results in a 3D point altered.
Additionally, combining different laser scanning methods and campaigns raises problems of accuracy. The main issue is how to ensure that the datasets are aligned to each other and determine whether systematic errors are present between them. Unifying point clouds from different campaigns could be accomplished by finding distinguishable points in both point clouds and moving one of the clouds to the correct location. Unfortunately, ALS data is usually not dense enough to Figure 2.
Location scheme of Pirita Beach in Tallinn Bay. Therefore, it is necessary to detect and quantify the systematic the scarp of the coastal forest was covered by an about 0. A total volume of 30 m3 with a typical The systematic error of elevation can be determined by grain size of 0.
Such reference surfaces e. By comparing the elevation differences between references therein. As the grain size of this sand volume was the same surface measured by TLS and ALS campaigns, the considerably smaller than the native sand, a large volume was extent of the systematic error can be determined. Points near probably relatively rapidly in a few years transported into the edge of the surface should be discarded to minimize the deeper areas. For example, a recession of the scarp at the northern end of the beach, and part of the ALS pulse might reflect from a curbstone instead extensive storm damage to the coastal forest have continued of the pavement.
The larger the reference surface area the . Although alterations of natural conditions such as an better is the accuracy and reliability of linking of the ALS and increase in storminess in the s  may have caused TLS measurements.
For example, construction of Pirita Harbor bayhead of Tallinn Bay, Estonia, is a typical small, embayed substantially decreased the supply of river sand. The largest beach of the southern coast of the Gulf of Finland Fig. The sediment transport processes along the beach are presented in length of the sandy area is about 2 km and the dunes are ,. The wave climate in the vicinity of the beach is relatively low: The average net loss of sand The stability of Pirita Beach has been discussed for several from the entire beach was estimated to be in the range of decades see , and references therein.
Several attempts to the beach whereas no prevailing transport direction exists in refill the beach with material dredged from a neighboring the southern sections. Consequently, different sections of the harbor or transported from mainland quarries were undertaken beach may have different level of erosion or accretion.
The beach up to m, elevation up to 2 m above the mean water test area is entirely located within a single flight corridor level and the total sand volume has increased. This tendency Fig. ALS The changes are marked northwards from the mouth of a small stream about m to the north of Pirita The TLS survey was conducted with a pulse-based Leica Harbor see its location in Fig.
The beach survey was performed from — The most intense erosion occurred at the six scanning stations Fig. For future surveys, nine interface between the sandy and till coasts at the northern end reference points were established near the beach Fig. The heights of the points transport and erosion and deposition patterns, the central area were determined with respect to a nearby located levelling of the beach in the vicinity of the above-mentioned stream benchmark.
Also, this area possibly has variable erosion and accumulation patterns and was selected as test region for this study. The Estonian Land Board also performed the ground filtering and classification of the measured points. In this Figure 3. Flight corridors hatch of different ALS campaigns.
Note a Figure 4. The reference points numbered green circles and TLS stations match between the location of the , and flight corridors.
Reference points , , , and were The studied beach area and the parking lot used as a reference surface measured with GNSS Reference point no.
This surface has been changeless i. The reference surface was measured with TLS using reference points that were linked to points located at the beach. Doing so improved the accuracy of the process of detection of deformations of the beach surface.
Figure 5. The profile in the centre of the test area matches profile No. Table I which affects the To properly interpret the described data, it is important to resolution and accuracy of the data. Both raster based Digital Elevation small sediment supply may keep its immediate vicinity in an Models and vector based Triangulated Irregular Network almost equilibrium on the background of a gradually eroding TIN models were used.
The constructed DTMs were used to beach section. Also, the meandering of its mouth under bi- analyze and visualize the changes that had occurred through directional littoral flow  may to a certain extent affect the the years.
The resulting DTMs constructed from ALS make it location of the waterline and the volume of sediment present possible to highlight relatively long-term changes within exactly along the profile. However, making strong — whereas a comparison of similar results from the conclusions from the described behavior may not be justified.
Finally, the ALS data from different years enable neighbouring cross-sections located about 70 m to the south us to recognize an interesting shift in the nature of changes to the beach around the year Single profiles 2 Elevation [m] The study area is an about m long strip in the central part of the beach Fig. This area contains one coastal profile No. Data for this profile, available for the years of — Fig. Although certain fluctuations occurred in the exact position of Figure 6.
Beach profile No. Data courtesy of the the beach surface and the zero-height line, none of these Geological Survey of Estonia. These changes have clear structure along both profiles. As 0 mentioned above, the changes at the distance of 0—10 m from their beginning may not be particularly reliable but still a -1 certain loss of sand from the northern segment of the study 3 10 20 30 40 50 areas characterized by this section of profile 3.
Profile 3 Further down to the waterline, at a distance of 10—25 m from Elevation [m] 2 the beginning, the beach has kept its shape in — but 1 has considerably lost sand in Its height has decreased by about 30 cm, which means the loss approximately 3 m3 per 0 meter of the coastline during this year.
Although a part of this -1 difference may stem from inexact match of the ALS and TLS 3 10 20 30 40 50 data, it is likely that this section of the beach had negative Profile 3. This loss has been only partially 2 compensated by an accumulation in the vicinity of waterline 1 ALS 25—30 m from the beginning.
This match of -1 accumulation with the one for profile 3 and with Fig. Figure 7. Beach profiles see the location in Fig. This profile information about the nature and course of the changes to the demonstrates rapid changes along its entire subaerial length. Profile 3 in Fig. This material was almost totally portrayed in Fig.
The data at a distance of 0—10 m from the eroded in — A minor accumulating section in the starting points of each profile are affected by the particular TLS data in the immediate vicinity of the waterline may flying line and height of the plane carrying the ALS device. The match of TLS and ALS for relevant density of the point cloud was lower in and the along this profile suggests that the differences between scarp is evidently not properly reflected in the data.
The end 2. Spatial changes of the scanned profile depends on the instantaneous location As we are interested in the capacity of the ALS and TLS of the waterline that may vary by 10—20 m depending on the techniques to highlight and identify changes to beaches, we water level during the scan.
This differences between the segments to the north and south of the match inter alia once more confirms the reliability of the TLS stream mouth Fig. The entire study area gained sand with and ALS data for changes to the subaerial beach. Given the total The overall original shape of the beach in was scanned area of about m2, the used technique is thus moderately convex signaling a relatively healthy situation.
Erosion was observed only in a few spots about noted also above. The above discussion suggests that this may reflect see the right panel of Fig. Most likely, the sediment volume caused by the small stream. The Figure 9. The different in these intervals. In terms of visually observed wave accumulation of sand was relatively rapid and homogeneus heights, the years — were relatively mild and the year along and across the entire southern segment. The height of relatively stormy.
The annual mean for these years at the beach typically increased by 20—30 cm cf. It mostly occurred in the landward part of the beach. It is thus is therefore likely that in — relatively mild wave likely that most of the accumulation here was driven by conditions with a comparatively large proportion of swells aeolian transport. As mentioned above, a few areas of generated by typical south-westerly storms in the open Baltic decrease in the height of the beach in the central part of the Sea and the western Gulf of Finland dominated the coastal segment apparently reflect local smoothing of the beach processes in Pirita Beach and were favourable for recovery of surface.
The autumn was stormy and the ice period started Fig. The later than usual. Several strong wave storms affected not only entire study area lost sand. Only a small vicinity of the stream the open Baltic Sea  but also the Gulf of Finland . The amount of lost sand single wave in the Gulf of Finland was recorded near Helsinki per meter of coastline was almost constant along the study in a storm that repeated the all-time highest significant wave area.
Differently from the years —, the changes to height in this bay .
Although the latter storm blew from the beach height were distributed unevenly along the beach the east, it is likely that Pirita Beach was frequently impacted cross-section. The loss was largest in the landward part of the by severe and destructive wave conditions in — Although the ALS measured changes. This structure is characteristic to severe data may have relatively large uncertainty in the immediate wave conditions superposed with high local water level.
In vicinity of the coastal scarp near the forest, it is still likely that such situations waves erode unprotected sediment relatively the loss of beach height was unevenly distributed across the far from the coastline here in the vicinity of the coastal beach. Estimating uncertainty due to model form is also hard in the face of small samples of destructively harvested and weighed trees, particularly as these are potentially biased to smaller trees Ahmed et al.
Lutz et al. This is further compounded by the fact that bigger trees have fundamentally greater variance in AGB Mascaro et al. Databases of wood density do exist, but are still limited given the diversity observed in practice Chave et al. A further fundamental issue is that all estimates other than destructive harvests are indirect, leading to the so-called fallacy of misplaced concreteness, i. This leads to potentially unknown uncertainties as well as inconsistencies in comparing AGB estimates derived from different data and using different upscaling methods Mitchard et al.
These issues of heteroscedasticity, i. An illustration of the variation in estimated AGB values from different ASEs in Ter-Mikaelian and Korzukin for two deciduous broadleaf species: red maple Acer rubrum and paper birch Betula papyrifera.
The estimated biomass is plotted against dbh, the diameter-at-breast-height of the trees. The shaded region represents an estimate of what Ahmed et al. These estimates are based on analysis of the statistical properties of the underlying data and models but are potentially conservative in assumptions of measurement error distributions normally distributed and the ability of simple ASE models to capture large-scale ecological variations in tree form Burt ; Burt et al.
The uncertainty in AGB at the plot level is less than at the individual tree level, due to the impact of measurement errors reducing with increasing sample size. This plot-scale uncertainty, however, may be less important than the potential issues of bias and heteroscedasticity as outlined above, particularly for larger trees.
These issues mean that in developing accurate EO-derived estimates of AGB with realistic uncertainty, care must be taken to develop appropriate ASEs spanning the full range of biomass likely to be observed, with uncertainty due to ASE form, wood density and upscaling, quantified as far as possible Ahmed et al.
Terrestrial laser scanning TLS has the capability to address many of these problems, by providing tree- and plot-level AGB estimates which are independent of allometry, unbiased in terms of tree size distributions and with well-quantified uncertainty Burt ; Stovall and Shugart TLS estimates collected widely and reliably can reduce current uncertainties in terrestrial C stocks, enable improved calibration and validation of AGB products, particularly from EO, and form the basis of improved allometric models.
Here, we describe key developments in the use of TLS to estimate AGB, present analysis of uncertainties that should be addressed, and highlight challenges that remain. This in turn has led to the development of tools and methods to exploit the resulting point cloud data for ecological applications Calders et al.
This has allowed the use of TLS to move from generalised structural metrics of tree size and shape such as height, DBH, crown size and volume considered as an enclosing envelope , trunk size and shape, to full reconstruction of 3D tree architecture down to higher order branches, comprising the majority of tree woody biomass.
Reliable automation of 3D architecture reconstruction is important for comparisons with EO, where the benefits of TLS measurements of large numbers of tree measurements are key. A range of methods have been proposed to extract full-tree 3D architecture automatically from TLS data e. Disney outlines some of the various assumptions and trade-offs underpinning them and challenges remaining to improve them. These include: mm-accurate co-registration of multiple point clouds in forest environments, and optimising data collection to achieve this Wilkes et al.
As a general guide, taller and more dense canopies will require more dense TLS sampling to minimise the impacts of occlusion in the upper canopy. Wilkes et al. Increasing the scan density provides diminishing returns; the additional cost of data collection and processing more than offset the slight improvement in the very uppermost branches, which in any case contribute the least to AGB. Panels show: i iterative pass-through filtering to fit enclosing cylinder red ; ii plane fitting to remove ground points; iii , iv and v iterative clustering and region-based segmentation multiple colours representing separate segments to remove neighbouring vegetation; and finally vi cylinder fitting to determine the location of first branching and hence the height-to-crown Various tools are now available to process and analyse TLS data for 3D reconstruction.
Other non-commercial tools developed for 3D point clouds more generally particularly ALS and now point clouds derived from structure-from-motion, e. This has various features for downsampling, editing and manipulating point clouds as well as for extracting general metrics.
Specific tools developed for addressing 3D reconstruction include: Treeseg Burt et al. Operates on single trees, or potentially at the stand-scale in less dense, more uniform canopies.
SimpleTree Hackenberg et al. Computree Othmani et al. Forestr Atkins et al. A key problem is how to assess the accuracy and uncertainty of the resulting 3D tree volumes. The only way to do this really is via destructive harvesting. The practical obstacles to this mean it is very difficult and so has been done in only a few cases. This is robust across varying canopy types, including large tropical trees and trees with buttressed trunks.
Buttressed trees are important for tropical forests as they can make up significant numbers of large trees in tropical plots. Disney et al. However, these fitting errors are randomly distributed and so tend to cancel out at the plot level. Less automated but more flexible volume fitting approaches such as mesh grids ibid. This is the strongest and most direct evidence that TLS can be used to test the existing ASEs, as well as to augment allometric datasets, to include more, larger trees.
This potentially addresses one of the fundamental limitations of ASEs highlighted above. More generally, Disney et al. This allows for sensitivity and uncertainty analysis and has been used to test some of the approaches above Raumonen et al.
The utility of this 3D simulation approach can also be tested by comparing simulations from the retrieved 3D architectural models with other, indirect estimates of canopy properties, including reflectance, leaf area index LAI , etc.
The advantage of this is that many of these EO-derived estimates of canopy properties have uncertainty which is much better-characterised than for estimates of AGB. Upper row: simulated image of a deciduous woodland left and UAV RGB image over the canopy on which this model simulation is based right. This is intended to be an indicative simulation rather than an exact representation, due to lack of leaf spectral and size information in the upper canopy.
Lower row: simulated left and measured right hemiphotos in leaf-off conditions from the same location within the canopy. This allows for new algorithm development and testing and benchmarking of simpler RT models, where canopy structural properties are known a priori Widlowski et al. They point out that TLS can figure prominently in both the measurement of plot-scale biomass and structure for improving EO retrievals, new allometric models, for pre-launch calibration and post-launch validation of EO estimates via plot measurements of biomass.
They note that in a TLS can provide unbiased by size class estimates of standard forestry properties DBH, height, etc. This illustrates how volume estimates derived from TLS measurements can be used to understand the EO signal, as well as quantifying forest structural differences across biomes. The two UK sites stand out as we would expect and hope! When expressed as a function of canopy height lower panel , the differences between the tropical and UK forests are less pronounced, except for the conifer which shows a characteristic dominant trunk and small crown biomass.
Information of this sort is critical for understanding the dependence of ASE-derived estimates of AGB based on height. These retrievals assume the dominant portion of the P-band backscatter signal over tropical forests arises from a peak of biomass in the upper part of the canopy Le Toan et al. Measuring these components accurately from TLS is likely to be a greater challenge, as outlined below.
Perhaps even more importantly, the TLS estimates have low or no bias Calders et al. They suggest that ASEs should be developed with sample sizes of at least individuals to ensure equations have high precision and low bias. The issue of bias is discussed by Burt et al. They note that the issue of heteroscedasticity means that uncertainty in ASE-derived AGB will always tend to be greater in higher biomass areas, but quantifying this uncertainty drastically improves the usefulness of resulting EO estimates.
Stovall et al. They conclude that quantifying wood density variability is essential for quantifying uncertainty in TLS-based biomass estimates. As noted above, by comparing at the volume level wherever possible, as facilitated by TLS, the issue of wood density can be separated from the problem of uncertainty in the ASE or TLS-derived volume. The challenges to wider adoption of TLS for AGB estimation are broadly technical, cost and scaling and are all linked.
The technical challenges lie in how to collect TLS data at plot scales and then retrieve structural properties, notably full 3D tree volume, with sufficient accuracy and quantified uncertainty.
Resolving smaller branches in tall, dense leaf-on canopies is a challenge for TLS, which will tend to increase uncertainty in these components of AGB at the tree-scale. Comparisons between TLS-derived and harvested AGB at the individual tree level seem to show little or no bias, in part because the most poorly resolved branches are the smallest. This is potentially improved with higher density point clouds and with more effective methods to separate wood and leaf points Boni Vicari et al.