This dataset represents yellow-cedar decline mapped across Southeast Alaska from 1986 through 2024 during annual aerial detection surveys and three remote sensing efforts in 2004, 2006, and 2020.Two cumulative decline datasets have been created from the same source data. This dataset has appended all years of data with detailed information on mapping methods and it retains overlap in decline mapped among various years and efforts. This layer is best for users hoping to filter or vary symbology by year, method, or other data fields. The companion dataset (Cumulative Yellow-Cedar Decline Union 1986 to 2024) accounts for decline overlap between survey years and methods, better for users focused on quantification of decline extent without overlap. Both datasets originate from the same source data and occupy the same spatial footprint.

Survey Types and Methods

Aerial surveys in Alaska typically cover 10-15% of the forested area of the state each year. Aerial surveyors from the Alaska Division of Forestry & Fire Protection and USDA Forest Service - Forest Health Protection document insect, disease, and abiotic damage in the forest from about 1000 feet altitude using a digital mobile sketch-mapping tablet and software. This forest health dataset includes both polygon and point data. Points have a buffered area based on tree number. Initially, aerial surveys were conducted using paper 1:250,000 USGS quadrangle maps. In 1999, aerial surveys in Alaska transitioned to mapping with mobile tablets and software, called digital mobile sketch-mapping. Digital mobile sketch-mapping displays the plane’s location via GPS, allows observers to zoom to various scales, and generally improves the accuracy and resolution of polygon placement compared to paper mapping. The survey type is identified as aerial survey, photo interpretation, or satellite imagery interpretation in the attribute table. During some survey years, the spatial distribution of decline that was added to the cumulative decline layer produced that year exceeded the acreage mapped during that year's annual aerial survey, originating from continued efforts to digitize decline mapped during prior survey efforts. Decline added through these digitization efforts is identified in the method field (DigitizedOutsideAerialDetectionSurvey). Other specificmethodsof mapping identified in the attribute table include AerialSurveyMobleSketchMapping, AerialSurveyPaperMapping, MidScaleAerialPhotoInterpretation, and SatelliteImageryScanandSketch.

Remote sensing efforts to map yellow-cedar decline include two mid-scale aerial photo interpretation efforts (25,388 acres along Peril Strait in 2004 and 9,057 acres on Mt. Edgecumbe on Kruzof Island in 2006). Details about the scale and methodology of photo interpretation mapping efforts are summarized in the report by P. Hennon and D. Wittwer: Evaluating key landscape features of a climate-induced forest decline (Project WC-EM-07-01). In: Potter, Kevin M.; Conkling, Barbara L., eds. 2013. Forest Health Monitoring: national status, trends, and analysis 2010. Gen. Tech. Rep. SRS-GTR-176. Asheville, NC: U.S. Department of Agriculture Forest Service, Southern Research Station. p117-122. The finer-scale polygons identified through photo interpretation were buffered 50 m to complement the scale of aerial survey. In 2020, decline was mapped on parts of Prince of Wales and Annette Islands using high-resolution satellite imagery (‘scan and sketch surveys’) due to the global pandemic. These details are recorded (method) and the methodology is summarized in Appendix 1 of the Forest Health Conditions in Alaska- 2020 report (FS-R10-FHP. 2020. Forest Health Conditions in Alaska 2020. Anchorage, AK. USDA-FS, Alaska Region. R10-PR-46. pp76).

Survey Year, Decline Type, Decline Severity, and Acres Impacted

The specific survey year of decline detection is also provided, apart from a period in the 1986 to 1987 and from 1991 to 1994 during which decline data was digitized from paper aerial survey maps. The 363,000 acres mapped during this period include both active and older decline. Aerial surveys typically focus on actively damaged and dying trees, but there was a need to understand the full spatial extent of the problem while the cause was investigated and compared to broader landscape patterns. The decline type identifies whether active decline, old decline, or a combination of old and active decline were mapped or tracked separately during a particular survey effort. For most of the 2000s, decline mapping has been limited to actively dying trees with discolored tree crowns only.

Decline severity was recorded with mapped decline starting in 2001, two years after the conversion from paper to digital mapping. Severity has been assigned in different ways over time, initially based upon the number of trees per acre affected (rated low, medium, and high) and later based on the percentage of host trees affected (rated very light, light, moderate, severe, and very severe). This dataset has merged these original severity assessments (DeclineSeverity2) into a simplified low, moderate, high system (DeclineSeverity) by lumping the very light and light ratings as low and the severe and very severe ratings as high. This is our first attempt to combine these systems, understanding that changes in methodology and variation in severity rankings among surveyors complicates interpretation.

Yellow-cedar decline root freezing injury can take 10-15 years to kill individual trees and is progressive within affected forests; therefore, there is overlap in the impacted areas mapped over time. About one-quarter of the decline extent mapped from 1986 to 1987 and from 1991 to 1994 was also captured through subsequent survey efforts. The lag between freezing injury and crown symptoms development, and the multi-year persistence of crown symptoms within affected trees and forests, makes it difficult to attribute mapped decline to specific weather events or years. It is important to note that this cumulative dataset has overlapping decline polygons that will be double-counted if the Acres column is summed, artificially inflating the spatial extent of decline. Accounting for decline overlap across time, there is a total extent of 728,000 acres of yellow-cedar decline in this dataset. This is slightly higher than the decline extent presented in the Forest Health Conditions in Alaska - 2024 report (718,500 acres) because of the inclusion of the photo interpretation efforts along Peril Strait and Mt. Edgecumbe.

Decline Mask or Clip Using USGS National Land Cover Classes

This cumulative yellow-cedar decline layer was clipped/restricted to areas occurring within upland forest and forested wetland cover classes in theUSGS National Land Cover Database-modified dataset(modified by Frances Biles, Geographer, PNW Research Station USDA Forest Service for the 2016Climate Adaptation Strategy for Yellow-Cedar in Alaska, PNW-GTR-917). This cover class refinement eliminates portions of mapped decline polygons that are unlikely to support yellow-cedar and reduces the total acreage of cumulative yellow-cedar decline by about 68,500 acres or 8.6%, and has been applied to cumulative yellow-cedar decline estimates published by the USDA-FS and partners over the past decade.The 2016 Climate Adaptation Strategy provides a detailed summary of yellow-cedar and yellow-cedar decline, as well as modeled projections of yellow-cedar habitat suitability into the 2050s and 2080s based on two key risk factors for decline development: snowpack and hydrology. Another significant refinement of the yellow-cedar decline cumulative layer occurred in 1996 that reduced the extent of decline by 120,000 acres, mostly by replacing very large polygons with smaller, carefully positioned polygons. The 1996 refinement was applied to mapped polygons that occurred prior to this effort such that the older errors were not reincorporated into the final dataset.

SurveyType: AerialSurvey, AerialPhotoInterpreation, SatelliteImageryInterpretation

Method: AerialSurveyMobleSketchMapping, AerialSurveyPaperMapping, DigitizedOutsideAerialDetectionSurvey, MidScaleAerialPhotoInterpretation, SatelliteImageryScanandSketch

SurveyYear: Year of detection. Assigned to a specific year except for the period in the 1980s and from 1991 to 1994.

DeclineType: Active, Old, Old_and_Active

DeclineSeverity: Low, Moderate, High

DeclineSeverity2: Low, Moderate, High, Very Light, Light, Severe, Very Severe

Acres: International Survey Acres, Alaska Albers Equal Area Conic, NAD83