Description: GeneralFlight lines were planned using a priori information including known areas of exemplary thermokarst landscapes, existing permafrost study sites, a 2016-2017 Sentinel-2 mosaic of 10-m spatial resolution, and a NWT-wide change detection product based on Landsat Tasseled Cap trend analysis (courtesy of Government of Northwest Territories, NWT Centre for Geomatics). Surveys were planned throughout the Taiga Plains/Shield, Boreal/Taiga/Tundra Cordillera, and the Southern and Northern Arctic.The foci areas were covered to the degree possible given weather, aircraft and fuel supply limitations. The aerial survey spanned a total of 72 suitable flying days within a four-year period. Both Bell 407 and Aerospatiale A-Star helicopters were used for survey work. The airborne transects totaled nearly 39,000 km. From an regional representation perspective, the transects intersected seven Level II NWT ecoregions, seven Level III NWT ecoregions, and 104 Level IV NWT ecoregions (Ecosystem Classification Group, 2009).Survey altitude averaged 250 to 350 m Above Ground Level for all regions, but depended daily on the type, topography and weather conditions of the transect. Surveys were conducted at typical cruise speeds of 190-200 km/h (GPS ground speed). Hard-copy maps and Garmin handheld GPS units (62S, Montana 680) were used for navigation, and transect lines typically deviated from what was planned a priorid due to weather or site prioritization.Along the transects a detailed and large-scale record of thermokarst and permafrost landscape features was captured via digital images, geographic locations (tracklogs) and tabulated observational information using ESRI Survey123. Digital images were mainly captured with a Nikon D5600® 24 megapixel digital single-lens reflex camera with Nikon 18-200mm f/3.5-5.6G ED VR II lens and Nikon GP-1 external GPS unit. Supporting cameras included a Nikon D750® 24 megapixel digital single-lens reflex (DSLR) camera and a D850® DSLR with various lenses (A Tamron 24-70 f2.8, Nikon 50 mm f1.8, and a Nikon 24mm f2.8), as well as a Sony RX100 III® 20 megapixel 1” sensor compact camera with F1.8-2.8 Carl Zeiss Vario-Sonnar T* lens. The Nikon D5600 camera was predominantly operated behind the permafrost scientist (left passenger seat in cockpit) and primarily acquired observations of mass-wasting, hydrology, and periglacial features and landscapes. Supporting cameras were typically operated on the opposite (pilot) side in case of a three-person field team, acquiring similarly featured photos as well as photos depicting general landscape conditions, disturbances and vegetation cover. If the field team consisted of two staff both cameras were operated on the left side.Over a four-year period 42,361 images were collected that passed the post-processing sequence. These images have formed an invaluable record of current thermokarst, vegetation and landform conditions that can be used in many subject areas, including monitoring of environmental change.Data governanceData owner and discipline authority: NT Geological Survey.Data custodian: NWT Centre for Geomatics.Data stewards: shared responsibility between NT Geological Survey and NWT Centre for Geomatics.Date of last update: April 2024.Update cycle: periodically when new field data is collected.Suggested citation for photo dataset: Van der Sluijs, J., Kokelj, S.V., Rudy, A.C.A. 2024. Geomatics field- and web-methodologies supporting a systematic aerial inventory and characterization of thaw-sensitive permafrost terrain in the Northwest Territories.Post-processing informationSeveral photo acquisition conditions challenged data management of the acquired images, including a relatively slow GPS lock of the Nikon GP-1 leading to missing geo-tags (20-30% of cases in year 2020), non-unique photo filenames as well as the presence of blurry or non-relevant photos (e.g., ground photos, wildlife). A post-processing pipeline was implemented to: Screen out blurry or non-relevant photos using freeware program PixelPeeper v1.2 (Textures.com, 2016);Batch resize remaining raw photos to half-scale, conduct auto-colour corrections based on histogram stretching (to overcome haze and absence of circular polarizing or UV filter), Apply GNWT Visual Identity branding and compress photos to 90% quality using freeware program Irfanview (Skiljan, 2017);Automatically rename each photo based on date/time/sensor while preserving original filename for cross-reference to Survey123 photo comments using freeware program Darktable (Darktable Team, 2020);Screen photos for the presence of GPS location in EXIF metadata via a custom script developed using the ESRI ArcPy python 3.0 site package, and geo-tag those photos with missing tags using freeware program GeoSetter and the Garmin handheld GPS-tracklogs (interpolated synchronization up to 60 sec with time adjustments);ESRI ArcGIS® Pro script “GeoTagged Photos to Points” was used to generate a hyperlinked point feature class in a file geodatabase. In total 42,361 geo-referenced digital images were available after the post-processing workflow. On average, a geo-referenced image was collected every 1-2 km, or about every 25-40 seconds, albeit it that the actual density varies depending on several factors. Photos acquired with the main camera for Survey123 inventory purposes (Nikon D5600) have file names containing "NIKON" or NIKON_D5600" depending on whether multiple Nikon cameras were used on that particular flight day.The accuracy of the single-frequency GPS geotags (generally; 10 m horizontal; 30 m vertical) is known to decrease with increasing speed of movement. In general users can expect a horizontal accuracy of about 50-100 m for the photo locations, however, the absolute accuracy of the geotags was not tested and is therefore unknown. The Nikon GP-1 geotags, although featuring built-in functionality, were sometimes not available or of suspect quality (e.g., a 150-800 m offset away from the flightline). Missing Nikon geotags were interpolated using handheld GPS tracklogs. Nikon photos >150 m from a flightline were manually adjusted. Location information for the non-geotagged Sony photos was also determined using interpolated time synchronization between the photo timestamp and the handheld GPS tracklogs. Time synchronization was resolved to within several seconds, meaning that data users can expect a horizontal offset due to the high speed of movement (e.g., a ground speed of 200 km/h is synonymous to covering a 555 m distance in a 10-sec window). For the purposes of regional permafrost observations, the offsets were deemed unimportant, yet users should be aware of these offsets for applications requiring more precise positions. Ground elevations (Zground) and Altitudes were approximated in reference to the European Space Agency's Copernicus Global Digital Elevation Model (GLO-90) without corrections for vertical datum differences, thus uncertainties and invalid altitude values may exist. Citation for GLO-90: European Space Agency, Sinergise (2021). Copernicus Global Digital Elevation Model. Distributed by OpenTopography. https://doi.org/10.5069/G9028PQB. Accessed: 2022-06-02.Field observations recorded via Survey123 form were spatially joined to photo records if these were located within 200 m of the photo location. Survey123 records (n = 7,936) were first screened for cases without photos or photos related to previous observations, leaving 7,880 candidates. The extensive attribute table of the Survey123 application was deduplicated into three fields: 1) Type, 2) Char, and 3) Mod. These fields describe the type (Table 2), characteristics (Table 3) and modifiers (Table 4) of the field observations. These attributes were spatially linked to 8,700 of the 42,361 images (21%) that met the 200 m maximum distance requirement using a combination of the Near (Analysis) tool and an attribute table join between [UniqueID] and [IN_FID].AttributesTable 1: Overview of attribute table.FieldDescriptionDirectionDirection (in degrees) the device was pointing during photo captureXGPS longitude of observation (in degrees; WGS84)YGPS latitude of observation (in degrees; WGS84)ZGPS height of observation (in metres; WGS84 ellipsoid)ZgroundEarth’s elevation below observation (in metres; EGM2008)AltitudeAltitude (in metres), subtracting Zground from ZUniqueIDUnique integer ID of photo, used in Near toolIN_FIDSame as UniqueIDNEAR_FIDClosest Survey123 observation (maximum distance = XXX0NEAR_DISTDistance (in metres) to Survey123 observationGlobalidUnique identifier from Survey123TrdateTransect date (MM/DD/YYYY)TrtimeTransect (local) timeThemeThermokarst theme (MW: mass wasting; HY: hydrological; PG: periglacial)TkcommentComment field for notesTypeType of thermokarst featureCharObserved primary characteristics of thermokarst featureModObserved secondary modifiers of thermokarst featureTable 2: Types of thermokarst features (Type).Mass wasting (MW)Hydrological (HY)Periglacial (PG)CodeDescriptionCodeDescriptionCodeDescriptionSlumpThaw slumpLake-thawThaw lakeUplandUplandSlide-shallowShallow slideLake-geometricGeometric lakeLowlandLowlandSlide-deepDeep slideLake-terrainTerrain-aligned lakeFloodplainFloodplainRock-avalancheRock avalancheOtherOtherSlopesSlopesDebris-flow-fanRock glacierRock-glacierDebris flow fanGullyTerrainGully terrainTable 3: Characteristics of thermokarst features (Char).Mass wasting (MW)Hydrological (HY)Periglacial (PG)CodeDescriptionCodeDescriptionCodeDescriptionTranTranslationalLake-rampRamparted lakeCSCollapse scarRotRotationalShore-colShore collapseCBCollapse basinRetRetrogressiveLake-slumpSlumpExpansionExpansionALSALDExpansionExpansionShore-colShoreline collapse (thick peat)BSBlock SlabLake-drainDrainedPTLPalsaComComplexLake-lowLoweredPTHLow centre polygonsComthawComplex-thawPP-polyPatterned ponding polygonalPTRHigh centre polygonsUndifUndifferentiatedPP-sbPatterned ponding stringbogPP-polyRectilinear polygonsBedconBedrock controlPP-unPatterned ponding undifferentiatedPP-sbPatterned ponding (polygonal)MolMollardLittoral-terrLittoral terracePP-unPatterned ponding (stringbog)DebDebris tongueLake-thawbThaw basinPF-polyPatterned ponding (undifferentiated)GlacGlacialDisc-sedDiscoloured sedimentPF-sbPatterned forest (polygonal)DendDendriticDisc-slDiscoloured shallowPFSPatterned forest (stringbog)ParParallelDisc-docDiscoloured DOCFTPatterned forest / shrub (undifferentiated)RcsCreeping/eyelinerDisc-bioDiscoloured algalSOLSolifluctionNoneNoneNoneNoneSSPatterned groundITInvoluted terrainCM-pingoConical mound (pingo)CM-otherConical mound (other)LITLithalsaPALFlowtracksNoneNoneTable 4: Modifiers of thermokarst features (Mod).Mass wasting (MW)Hydrological (HY)Periglacial (PG)CodeDescriptionCodeDescriptionCodeDescriptionStatusWaterbody sizeDrainage typeActActiveLakeLakeDrain-dendDendritic-parallelRecRecentPondPondDrain_rectRectangle-trellisInactInactiveCoastCoastlineDrain_multiMulti-basinalOldOldDrain_noneNoneAncAncientSpatial extentDnaDoes not applyDisDiscreteDegradation levelBasBasinDeg-dnaDoes not applySize/AbundanceLandLandscapeDeg-nonNoneSmaSmallDeg-lowLowMedMediumDeg-medMedMegMegaDeg-highHighIndIndividualFewFewSpatial extentMnyManyDisDiscreteBasBasinIce presentLandLandscapeVisYVisibleVisNNot visibleCnaCould not assessIce typeMassMassiveSegSegregatedWedWedge-iceUndifUndifferentiatedHeadwall heightBlw55to205-20 mOver20>20 mHydrological Connectivity0No connection1Connection2Downstream deposition (Debris tongue)GeomorphologyLakeLakesideFluvFluvialCoastCoastalMtnMountain
Copyright Text: The geomatics data collection and processing procedures were produced by J. van der Sluijs, S.V. Kokelj and A.C.A. Rudy, Northwest Territories Centre for Geomatics and Northwest Territories Geological Survey as a data product to support the Thermokarst Collective Permafrost Mapping Project.
Description: Flight lines were planned using a priori information including known areas of exemplary thermokarst landscapes, existing permafrost study sites, a 2016-2017 Sentinel-2 mosaic of 10-m spatial resolution, and a NWT-wide change detection product based on Landsat Tasseled Cap trend analysis (courtesy of Government of Northwest Territories, NWT Centre for Geomatics). Surveys were planned throughout the Taiga Plains/Shield, Boreal/Taiga/Tundra Cordillera, and the Southern and Northern Arctic.The foci areas were covered to the degree possible given weather, aircraft and fuel supply limitations. The aerial survey spanned a total of 72 suitable flying days within a four-year period. Both Bell 407 and Aerospatiale A-Star helicopters were used for survey work. The airborne transects totaled nearly 39,000 km. From an regional representation perspective, the transects intersected seven Level II NWT ecoregions, seven Level III NWT ecoregions, and 104 Level IV NWT ecoregions (Ecosystem Classification Group, 2009).Survey altitude averaged 250 to 350 m Above Ground Level for all regions, but depended daily on the type, topography and weather conditions of the transect. Surveys were conducted at typical cruise speeds of 190-200 km/h (GPS ground speed). Hard-copy maps and Garmin handheld GPS units (62S, Montana 680) were used for navigation, and transect lines typically deviated from what was planned a priorid ue to weather or site prioritization.Along the transects a detailed and large-scale record of thermokarst and permafrost landscape features was captured via digital images, geographic locations (tracklogs) and tabulated observational information. Data from one or several Garmin handheld GPS receivers (.GPX format) were imported into a file geodatabase as seperate feature classes and manually inspected for quality and consistency. These feature classes were merged whereby GPS locations associated with ground visits were deleted. A "Date" string attribute was added for each sample. Lines were generated from the cleaned observations using ArcGIS "Points to Line" tool, with the Date field as Line Field and the DateTime field as the Sort Field. Transects with missing handheld GPS information were manually digtized based on GPS tags from photo records, field survey forms, and field notes. Flight line distances (in kilometer) were calculated based on NAD83 CSRS with respective UTM zone projection.Data governanceData owner and discipline authority: NT Geological Survey.Data custodian: NWT Centre for Geomatics.Data stewards: shared responsibility between NT Geological Survey and NWT Centre for Geomatics.Date of last update: April 2024.Update cycle: periodically when new field data is collected.Suggested citation for photo dataset: Van der Sluijs, J., Kokelj, S.V., Rudy, A.C.A. 2024. Geomatics field- and web-methodologies supporting a systematic aerial inventory and characterization of thaw-sensitive permafrost terrain in the Northwest Territories.
Copyright Text: The geomatics data collection and processing procedures were produced by J. van der Sluijs, S.V. Kokelj and A.C.A. Rudy, Northwest Territories Centre for Geomatics and Northwest Territories Geological Survey as a data product to support the Thermokarst Collective Permafrost Mapping Project.