The NLCD 2001 is created by partitioning the U.S. into mapping zones. A total of 66 mapping zones within the conterminous U.S., with an additional 13 in Alaska, were delineated based on ecoregion and geographical characteristics, edge matching features and the size requirement of Landsat mosaics. Mapping zone ak01 encompasses a portion of Alaska. Questions about the NLCD mapping zone ak01 can be directed to the NLCD 2001 land cover mapping team at the USGS/EROS, Sioux Falls, SD (605) 594-6151 or mrlc@usgs.gov.
Conceptually, the descriptive tree is a classification tree generated by using the final minimum-map- unit land cover product (1 acre) as training data, and Landsat and other ancillary data as predictors. The goal of the descriptive tree is to summarize the effects of boosted trees (10 sequential classification trees) into a single condensed decision tree that can be used as a diagnostic tool for the classification process. This descriptive tree can be used to assess the relative importance of each of the input data sets on each land cover class. Such information may also be useful to customize the minimum-mapping-unit classification to meet a user's specific needs through raster modeling. Descriptive trees usually capture 60 to 80% of the information from the original land cover data.
A logic statement from a descriptive tree classification describes each classification rule for each classified pixel. An example of the logic statement follows:
IF tasseled-cap wetness > 140 and imperviousness = 0 and canopy density < 4, then classify as Water
Additional information may be found at <http://www.mrlc.gov/mrlc2k_nlcd.asp>.
To conduct the land cover classification using DT, a large quantity of training data is required. For mapping zone ak01, training data were collected from several combined sources including National Wetlands Inventory (NWI), Alaska Burned Area Emergency Response (BAER), Bureau of Land Management (BLM), Ducks Unlimited (DU), field collected data, and a variety of local data sources.
Note that the training data were used to map all land cover classes except for four classes in urban and sub-urban areas (developed open space, low intensity developed, medium intensity developed, high intensity developed). All urban and suburban land cover classes for Alaska were mapped through either a sub-pixel quantification of impervious surfaces using a regression tree modeling method, or through a series of manual classifications that may include roads, towns, and pipeline GIS data gathered for Alaska. Quality assessment of all landcover classes was limited to available high-resolution imagery, previous classifications, and any field collected ground truth.
Following the development of the best classification through decision tree modeling, additional steps were required to complete the final land cover product. The four classes in urban and suburban areas were determined by a calibration with known impervious areas in Alaska mapping zone 8, which includes the Anchorage area, a crosswalk of the manually interpreted areas, and a cursory impervious estimate for each mapping zone. From the interpreted developed product, developed areas were grouped into the following four categories: (1) developed open space (imperviousness approximately < 20%), (2) low-intensity developed (imperviousness approximately from 20 - 49%), (3) medium intensity developed (imperviousness approximately from 50 -79%), and (4) high-intensity developed (imperviousness approximately > 79%). Alaska landcover mapping also provided one additional challenge in that some areas of high relief in Alaska are in shadow at the time of nominal Landsat imagery collection. These shadowed areas were classified with the best available methods. However, a landcover containing shadowed areas that were classified will be available upon special request.
Other classes of forest and non-forest were combined with the urban classes to complete the land cover product. Finally visual inspection of the classification was made with areas/pixels that were wrongly classified delineated first as an "area of interest" (AOI), subsequently then limited manual editing was done to eliminate the classification error within the AOI.
The completed single pixel product was then generalized to a 1 acre (approximately 5 ETM+ 30 m pixel patch) minimum mapping unit product using a "smart eliminate" algorithm. This aggregation program subsumes pixels from the single pixel level to a 5-pixel patch using a queens algorithm at doubling intervals. The algorithm consults a weighting matrix to guide merging of cover types by similarity, resulting in a product that preserves land cover logic as much as possible.
Acquisition dates of Landsat ETM+ (TM) scenes used for land cover classification in zone ak01 are as follows:
Leafon-
Index 10 for Path 69/Row 11 on 06/25/2001 = Scene_ID 7069011000117650
Index 11 for Path 72/Row 11 on 08/30/2000 = Scene_ID 7072011000024350
Index 11 for Path 72/Row 12 on 08/30/2000 = Scene_ID 7072012000024350
Index 04 for Path 73/Row 12 on 07/02/1999 = Scene_ID 7073012009918350
Index 05 for Path 74/Row 10 on 07/14/2001 = Scene_ID 7074010000119550
Index 05 for Path 74/Row 11 on 07/14/2001 = Scene_ID 7074011000119550
Index 09 for Path 74/Row 12 on 08/02/2002 = Scene_ID 7074012000221450
Index 13 for Path 74/Row 12 on 06/08/1985 = Scene_ID 5074012008515910
Index 06 for Path 77/Row 10 on 08/01/2000 = Scene_ID 7077010000021450
Index 12 for Path 77/Row 10 on 07/03/2001 = Scene_ID 7077010000118450
Index 08 for Path 77/Row 12 on 06/30/2000 = Scene_ID 7077012000018250
Index 07 for Path 77/Row 11 on 07/14/1999 = Scene_ID 7077011009919550
Index 14 for Path 77/Row 12 on 08/31/1999 = Scene_ID 7077012009924350
Index 01 for Path 78/Row 12 on 07/29/2002 = Scene_ID 7078012000221050
Index 02 for Path 77/Row 12 on 08/31/1999 = Scene_ID 7077012009924350
Index 02 for Path 78/Row 12 on 07/29/2002 = Scene_ID 7078012000221050
Index 03 for Path 79/Row 10 on 07/01/2001 = Scene_ID 7079010000118250
Index 03 for Path 79/Row 11 on 07/01/2001 = Scene_ID 7079011000118250
Landsat data and ancillary data used for the land cover prediction -
Data Type of DEM composed of 1 band of Continuous Variable Type.
Data Type of Slope composed of 1 band of Continuous Variable Type.
Data Type of Aspect composed of 1 band of Categorical Variable Type.
Data type of Position Index composed of 1 band of Continuous Variable Type.
Alaska GPS roads
Alaska Pipeline data
Alaska Airports layer
Alaska Cities layer