LASmoons: David Bandrowski

David Bandrowski (recipient of three LASmoons)
Yurok Tribe
Native American Indian Tribe in Northern California, USA

Wild spring-run Chinook salmon populations on the South Fork Trinity River in Northern California are near the brink of extinction. The South Fork Trinity River is the most remote and the largest un-dammed river in the State of California, federally designated as a wild and scenic river, and is a keystone watershed within the Klamath River basin supporting one of the last remaining populations of wild spring-run Chinook salmon. Ecosystem restoration is urgently needed to improve watershed health in the face of climate change, land use, and water diversions. This drastic decline of the wild salmon species motivated the Yurok Tribe and its partners to take action and implement this project as a last opportunity to save this species before extinction. Spring-run Chinook are extremely important for the Yurok people culturally, spiritually, and for a subsistence food source.

sample of the available photogrammetry data

Due to budgetary constraints, airborne LiDAR is not available; therefore the Yurok Tribe has been using aerial drones and Structure for Motion (SfM) photogrammetry to develop DTM models that can be used in determining available salmon habitat and to develop prioritized locations for restoration. The watershed has extremely heavy vegetation, and obtaining bare-earth surfaces for hydraulic modeling is difficult without the proper tools. The goal is to use LAStools to further restoration science and create efficient workflows for DTM development.

 length of river mapped: 8 Kilometers
+ number of points: 150,856,819
+ horizontal datum: North American Datum 83 – California State Plane – Zone 1 (usft)
+ vertical datum: North American Vertical Datum 88

LAStools processing:
1) data quality checking [lasinfo, lasview, lasgrid]
2) classify ground and non-ground points [lasground and lasground_new]
3) remove low and high outliers [lasheight, lasnoise]
4) create DTM tiles at appropriate resolution [las2dem]
5) create a normalized point cloud [lasheight]

LASmoons: Sebastian Flachmeier

Sebastian Flachmeier (recipient of three LASmoons)
UniGIS Master of Science, University of Salzburg, AUSTRIA
Bavarian Forest National Park, administration, Grafenau, GERMANY

The Bavarian Forest National Park is located in South-Eastern Germany, along the border with the Czech Republic. It has a total area of 240 km² and its elevation ranges from 600 to 1453 m. In 2002 a project called “High-Tech-Offensive Bayern” was started and a few first/last return LiDAR transects were flown to compute some forest metrics. The results showed that LiDAR has an advantage over other methods, because the laser was able to get readings from below the canopy. New full waveform scanner were developed that produced many more returns in the lower canopy. The National Park experimented with this technology in several projects and improved their algorithms for single tree detection. In 2012 the whole park was flown with full waveform and strategies for LiDAR based forest inventory for the whole National Park were developed. This is the data that is used in the following workflow description.

The whole Bavarian Forest National Park (black line), 1000 meter tiles (black dotted lines), the coverage of the recovered flight lines (light blue). In the area marked yellow within the red frame there are gaps in some of the flightlines. The corresponding imagery in Google Earth shows that this area contains a water reservoir.

Several versions of the LiDAR existed on the server of the administration that didn’t have the attributes we needed to reconstruct the original flight lines. The number of returns per pulse, the flight line IDs, and the GPS time stamps were missing. The goal was a workflow to create a LAStools workflow to convert the LiDAR from the original ASCII text files provided by the flight company into LAS or compressed LAZ files with all fields properly populated.

 ALS data flown in 2012 by Milan Geoservice GmbH 650 m above ground with overlap.
+ full waveform sensor (RIEGL 560 / Q680i S) with up to 7 returns per shot
+ total of 11.080.835.164 returns
+ in 1102 ASCII files with *.asc extension (changed to *.txt to avoid confusion with ASC raster)
+ covered area of 1.25 kilometers
+ last return density of 17.37 returns per square meter

This data is provided by the administration of Bavarian Forest National Park. The workflow was part of a Master’s thesis to get the academic degree UniGIS Master of Science at the University of Salzburg.

LAStools processing:

The LiDAR was provided as 1102 ASCII text files named ‘spur000001.txt’ to ‘spur001102.txt’ that looked like this:

more spur000001.txt
4589319.747 5436773.357 685.837 49 106 1 215248.851500
4589320.051 5436773.751 683.155 46 24 2 215248.851500
4589320.101 5436773.772 686.183 66 87 1 215248.851503

Positions 1 to 3 store the x, y, and z coordinate in meter [m]. Position 4 stores the “echo width” in 0.1 nanoseconds [ns], position 5 stores the intensity, position 6 stores the return number, position 7 stores the GPS time stamp in seconds [s] of the current GPS week. The “number of returns (of given pulse)” information is not explicitly stored and will need to be reconstructed in order, for example, to identify which returns are last returns. The conversion from ASCII text to LAZ was done with the txt2las command line shown below that incorporates these rationals:

  • Although the ASCII files list the three coordinates with millimeter resolution (three decimal digits), we store only centimeter resolution which is sufficient to capture all the precision in a typical airborne LiDAR survey.
  • After computing histograms of the “return number” and the “echo width” for all points with lasinfo and determining their maximal ranges it was decided to use point type 1 which can store up to 7 returns per shot and store the “echo width” as an additional attribute of type 3 (“unsigned short”) using “extra bytes”.
  • The conversion from GPS time stamp in GPS week time to Adjusted Standard time was done by finding out the exact week during which Milan Geoservice GmbH carried out the survey and looking up the corresponding GPS week 1698 using this online GPS time calculator.
  • Information about the Coordinate Reference System “DHDN / 3-degree Gauss-Kruger zone 4” as reported in the meta data is added in form of EPSG code 31468 to each LAS file.
txt2las -i ascii\spur*.txt ^
        -parse xyz0irt ^
        -set_scale 0.01 0.01 0.01 ^
        -week_to_adjusted 1698 ^
        -add_attribute 3 "echo width" "of returning waveform [ns]" 0.1 0 0.1 ^
        -epsg 31468 ^
        -odir spur_raw -olaz ^
        -cores 4

The 1102 ASCII files are now 1102 LAZ files. Because we switched from GPS week time to Adjusted Standard GPS time stamps we also need to set the “global encoding” flag in the LAS header from 0 to 1 (see ASPRS LAS specification). We can do this in-place (i.e. without creating another set of files) using the following lasinfo command:

lasinfo -i spur_raw\spur*.laz ^
        -nh -nv -nc ^
        -set_global_encoding 1

To reconstruct the missing flight line information we look for gaps in the sequence of GPS time stamps by computing GPS time histograms with lasinfo and bins of 10 seconds in size:

lasinfo -i spur_raw\spur*.laz -merged ^
        -histo gps_time 10 ^
        -o spur_raw_all.txt

The resulting histogram exhibits the expected gaps in the GPS time stamps that happen when the survey plane leaves the target area and turns around to approach the next flight line. The subsequent histogram entries marked in red show gaps of 120 and 90 seconds respectively.

more spur_raw_all.txt
bin [27165909.595196404,27165919.595196255) has 3878890
bin [27165919.595196255,27165929.595196106) has 4314401
bin [27165929.595196106,27165939.595195957) has 435788
bin [27166049.595194317,27166059.595194168) has 1317998
bin [27166059.595194168,27166069.595194019) has 4432534
bin [27166069.595194019,27166079.59519387) has 4261732
bin [27166239.595191486,27166249.595191337) has 3289819
bin [27166249.595191337,27166259.595191188) has 3865892
bin [27166259.595191188,27166269.595191039) has 1989794
bin [27166349.595189847,27166359.595189698) has 2539936
bin [27166359.595189698,27166369.595189549) has 3948358
bin [27166369.595189549,27166379.5951894) has 3955071

Now that we validated their existence, we use these gaps in the GPS time stamps to split the LiDAR back into the original flightlines it was collected in. Using lassplit we produce one file per flightline as follows:

lassplit -i spur_raw\spur*.laz -merged ^
         -recover_flightlines_interval 10 ^
         -odir strips_raw -o strip.laz

In the next step we repair the missing “number of returns (per pulse)” field that was not provided in the ASCII file. This can be done with lasreturn assuming that the point records in each file are sorted by increasing GPS time stamp. This happens to be true already in our case as the original ASCII files where storing the LiDAR returns in acquisition order and we have not changed this order. If the point records are not yet in this order it can be created with lassort as follows. As these strips can have many points per file it may be necessary to run the new 64 bit executables by adding ‘-cpu64’ to the command line in order to avoid running out of memory.

lassort -i strips_raw\strips*.laz ^
        -gpstime -return_number ^
        -odir strips_sorted -olaz ^
        -cores 4 -cpu64

An order sorted by GPS time stamp is necessary as lasreturn expects point records with the same GPS time stamp (i.e. returns generated by the same laser pulse) to be back to back in the input file. To ‘-repair_number_of_returns’ the tool will load all returns with the same GPS time stamp  and update the “number of returns (per pulse)” attribute of each return to the highest “return number” of the loaded set.

lasreturn -i strips_sorted\strips*.laz ^
          -repair_number_of_returns ^
          -odir strips_repaired -olaz ^
          -cores 4

In a final step we use las2las with the ‘-files_are_flightlines’ option (or short ‘-faf’) to set the “file source ID” field in the LAS header and the “point source ID” attribute of every point record in the file to the same unique value per strip. The first file in the folder will have all its field set to 1, the next file will have all its field set to 2, the next file to 3 and so on. Please do not run this on multiple cores for the time being.

las2las -i strips_repaired\strips*.laz ^
        -files_are_flightlines ^
        -odir strips_final -olaz

It’s always useful to run a final validation of the files using lasvalidate to reassure yourself and the people you will be sharing the data with that nothing funky has happened during any of these conversion steps.

lasvalidate -i strips_final\strip*.laz ^
            -o strips_final\report.xml

And it can also be useful to add an overview in SHP or KML format to the delivery that can be created with lasboundary as follows:

lasboundary -i strips_final\strip*.laz ^
            -overview -labels ^
            -o strips_final\overview.kml

The result was 89 LAZ files (each containing one complete flightline) totaling 54 GB compared to 1102 ASCII files (each containing a slice of a flightline) totaling 574 GB.

LASmoons: Maria Kampouri

Maria Kampouri (recipient of three LASmoons)
Remote Sensing Laboratory, School of Rural & Surveying Engineering
National and Technical University of Athens, GREECE

The Aralar Natural Park, famous for its stunning landscapes, is located in the southeast of the province of Gipuzkoa, sharing a border with the neighboring province of Navarre. Inside the park there are nature reserves of exceptional importance, such as beech woods, large number of yew trees, very singular species of flora and fauna and areas of exceptional geological interest. Griffon vultures, Egyptian vultures, golden eagles and even bearded vultures (also known as lammergeier) can be seen flying over this area. European minks and Pyrenean desmans can be found in the streams and rivers that descend from the mountain tops.

The concept of biodiversity is based on inter- and intra-species genetic variation and has been evolving over the past 25 years. The importance of mapping biodiversity in order to plan its conservation, as well as identifying patterns in endemism and biodiversity hot-spots, have been pillars for EU and global environmental policy and legislation. The coupling of remote sensing and field data can increase reliability, periodicity and reproduce-ability of ecosystem process and biodiversity monitoring, leading to an increasing interest in environmental monitoring, using data for the same areas over time. Natural processes and complexity are best explored by observing ecosystems or landscapes through scale alteration, using spatial analysis tools, such as LAStools.

DTM generated with restricted version of las2dem above point limits

The aim of this study is to investigate the potential use of LiDAR data for the identification and determination of forest patches of particular interest, with respect to ecosystem dynamics and biodiversity and to produce a relevant biodiversity map, based on Simpson’s Diversity Index for Aralar Natural Park.

 approximately 123 km^2 of LiDAR in 1km x 1km LAS tiles
+ Average point density: 2 pts/m^2
+ Spatial referencing system: ETRS89 UTM zone 30N with elevations on the EGM08 geoid. Data from LiDAR flights are These files were obtained from the LiDAR flight carried out in 2008 by the Provincial Council of Gipuzkoa and the LiDAR flights of the Basque Government.

LAStools processing:
1) data quality checking [lasinfolasoverlaplasgridlasreturn]
2) classify ground and non-ground points [lasground]
3) remove low and high outliers [lasheight, lasnoise]
4) identify buildings within the study area [lasclassify]
5) create DTM tiles with 0.5 step in ‘.bil’ format [las2dem]
6) create DSM tiles with 0.5 step in ‘.bil’ format [las2dem]
7) create a normalized point cloud [lasheight]
8) create a highest-return canopy height model (CHM) [lasthin, las2dem]
9) create a pit-free (CHM) with the spike-free algorithm [las2dem]
10) create various rasters with forest metrics [lascanopy]

The generated elevation and forest metrics rasters are then combined with satellite data to create a biodiversity map, using Simpson’s Diversity Index.

LASmoons: Sebastian Kasanmascheff

Sebastian Kasanmascheff (recipient of three LASmoons)
Forest Inventory and Remote Sensing
Georg-August-Universität Göttingen, GERMANY

Forest inventories are the backbone of forest management in Germany. In most federal forestry administrations in Germany, they are performed every ten years in order to assure that logging activities are sustainable. The process involves trained foresters who visit each stand (i.e. an area where the forest is similar in terms of age structure and tree species) and perform angle count sampling as developed by Walter Bitterlich in 1984. In a second step the annual growth is calculated using yield tables and finally a harvest volume is derived. There are three particular reasons to investigate how remote sensing can be integrated in the current inventory system:

  1. The current process does not involve random sampling of the sampling points and thus does not offer any measure of the accuracy of the data.
  2. Forest engineers hardly ever rely on the inventory data as a stand-alone basis for logging planning. Most often they rely on intuition alone and on the total volume count that they have to deliver for a wider area every year.
  3. In the last ten years, the collection of high-resolution LiDAR data has become more cost-effective and most federal agencies in Germany have access to it.

In order to be able to integrate the available remote-sensing data for forest inventories in Germany, it is important to tell apart different tree species as well as estimate their volumes.

Hesse is one of the most forested federal states in Germany.

The goal of this project is to perform an object-based classification of conifer trees in Northern Hesse based on high-resolution LiDAR and multi-spectral orthophotos. The first step is to delineate the tree crowns. The second step is to perform a semi-automated classification using the spectral signature of the different conifer species.

 DSM (1m), DTM (1m), DSM (0.2 m) of the study area
+ Stereo images with 0.2 m resolution
+ high-resolution LiDAR data (average 10 points/m²)
+ forest inventory data
+ vector files of the individual forest stands
+ ground control points (field data)
All of this data is provided by the Hessian Forest Agency (HessenForst).

LAStools processing:
1) merge and clip the LAZ files [las2las]
2) classify ground and non-ground points [lasground]
3) remove low and high outliers [lasheight, lasnoise]
4) identify buildings within the study area [lasclassify]
5) create a normalized point cloud [lasheight]
6) create a highest-return canopy height model (CHM) [lasthin, las2dem]
7) create a pit-free (CHM) with the spike-free algorithm [las2dem]

LASmoons: Manuel Jurado

Manuel Jurado (recipient of three LASmoons)
Departamento de Ingeniería Topográfica y Cartografía
Universidad Politécnica de Madrid, SPAIN

The availability of LiDAR data is creating a lot of innovative possibilities in different fields of science, education, and other field of interests. One of the areas that has been deeply impacted by LiDAR is cartography and in particular one highly specialized sub-field of cartography in the domain of recreational and professional orienteering running: the production of high-quality maps for orienteering races (Ditz et al., 2014). These are thematic maps with a lot of fine detail which demands many hours of field work for the map maker. In order to reduce the fieldwork, LiDAR data obtained from Airborne Research Australia (ARA) is going to be used in order to obtain DEM and to extract features that must be included in these maps. The data will be filtered and processed with the help of LAStools.

Final map with symbolism typical for use in orienteering running

The goal of this project is to extract either point (boulders, mounds), linear (contours, erosion gullies, cliffs) and area features (vegetation density) that should be drawn in a orienteering map derived from high-resolution LiDAR. Different LiDAR derived raster images are being created: 0.5m DTM, vegetation density (J. Ryyppo, 2013), slope, Sky-View factor (Ž. Kokalj et al., 2011), and shaded relief. The area used is in Renmark, South Australia and the produced map is going to be used for the Australian Orienteering Championships 2018.

Sky-View factor of DTM for same area as shown above.

4 square kilometers of airborne LiDAR data produced by Airborne Research Australia at 18 pulses per square meter using the full waveform scanning LiDAR Q680i-S laser scanner from RIEGL
+ 60 hours of check and validation work in the field

LAStools processing:
1) tile into 500 by 500 meter tiles with 20 meter buffer [lastile]
2) classify isolated points as noise [lasnoise]
3) classify point clouds into ground and non-ground [lasground]
4) create a Digital Terrain Model (DTM) [las2dem]
5) normalize height of points above the ground [lasheight]
6) compute vegetation density metrics [lascanopy]
7) create hillshades of the raster DTMs [blast2dem or GDAL]

Ditz, Robert, Franz Glaner, and Georg Gartner. (2014). “Laser Scanning and Orienteering Maps.” Scientific Journal of Orienteering 19.1.
JRyyppo, Jarkko. (2013). “Karttapullautin vegetation mapping guide”.
Kokalj, Žiga, Zaksek, Klemen, and Oštir, Krištof. (2011). Application of sky-view factor for the visualization of historic landscape features in lidar-derived relief models. Antiquity. 85. 263-273.