Skip First Level Navigation | Skip All Navigation

Intermediate ArcGIS 9
   Landcover Analysis

Course Introduction
Getting Started
Overview of data
 - MELCD products
 - NLCD products
 - Satellite data
 - Review of legacy data
 - TIFF data format

Spatial Analyses
<Previous: Identify tool    Next: NLCD products>

MELCD 2004 Products

Concepts:
   MELCD 2004 landcover
   MELCD 2004 imperviousness
   minimum mapping unit

In these next sections, we will cover all of the products available to end-users from the combined Maine Landcover Dataset 2004 (MELCD 2004) and National Landcover Dataset 2001 (NLCD 2001).

The MELCD 2004 products include a 5-meter resolution landcover dataset and a 5-meter resolution imperviousness dataset.  This project was an integrated extension of the NLCD 2001 project for Maine.

MELCD 2004 Landcover
The landcover data are based on 2004 SPOT imagery fused (pan-sharpened) with 2001 LandSat data (5m black-and-white data were merged with 30-m color and infrared data, then classified).  The pixel resolution is 5-meters, with a minimum mapping unit (MMU) of .89 acres.  The MMU is the smallest area that can be accurately mapped.  In practice, some areas smaller than the MMU are mapped in this layer, but they cannot be judged at the same level of accuracy as the rest of these data.  Accuracy as a whole, based on a sample of 1671 points, is 75% according to the accuracy report.  Maximum scale recommended for this dataset is 1:24,000; it is designed for analyses at the town or subwatershed level.  For more complete information, read the metadata.  The classification includes 23 distinct classes:

Pixel
value
   Class
2        Developed, High Intensity  (80-100% impervious)
3        Developed, Medium Intensity  (50-79% impervious)
4        Developed, Low Intensity  (21-49% impervious)
5        Developed, Open Space  (developed areas, but 0-20% impervious - city parks, golf courses, baseball fields, etc.)
6        Cultivated Crop  (production of annual crops such as corn, potatoes, strawberries, and tilled barren fields)
7        Pasture/Hay  (grasses are major vegetation, managed for harvesting as hay or grazing)
8        Grassland/Herbaceous  (unmanaged grasslands - rare in Maine)
9        Deciduous Forest  (> 20% tree canopy cover, > 75% of trees are deciduous)
10      Evergreen Forest  (> 20% tree canopy cover, > 75% of trees are evergreen)
11      Mixed Forest   (> 20% tree canopy cover, 25-75% are deciduous)
12      Scrub/Shrub   (woody vegetation < 5m tall is > 20% of cover - typically regenerating fields, cuts, or rights-of-way)
13      Wetland Forest  (freshwater wetland with > 20% tree canopy cover)
15      Wetland  (all other wetlands)
16      Road/Runway  (impervious road or runway, but not in developed areas)
19      Unconsolidated Shore (rocky shore, mudflats, sand beach, exposed lake shoreline)
20      Bare Ground  (open quarries and pits, granite outcrops and peaks)
21      Open Water   (water bodies typically > 10m wide)
22      Blueberry Field   (commercial blueberry operations)
23      Recent Clearcut   (forested area with > 90% canopy removal 2001-2004)
24      Light Partial Cut   (forested area with 20-50% canopy removal 1995-2001)
25      Heavy Partial Cut  (forested area with 50-100% canopy removal 1995-2001)
26      Regenerating Forest  (forested area with canopy increase 1995-2001)
27      Alpine  (shrubby or grassy vegetation above treeline on mountains)

image of MELCD data
MELCD 2004 Landcover data showing Bangor, Maine.

MELCD 2004 Imperviousness Data
The imperviousness data indicates, at a 5-meter pixel resolution, whether or not a pixel is 50% or more impervious (pixel value of 0) or <50% impervious (pixel value of 1) based on 2004 satellite data.  Like the landcover data,
the pixel resolution is 5-meters, with a minimum mapping unit (MMU) of .89 acres.  The MMU is the smallest area that can be accurately mapped.  In practice, some areas smaller than the MMU are mapped in this layer, but they cannot be judged at the same level of accuracy as the rest of these data.  Accuracy as a whole, based on a sample of 1444 points, is 94% according to the accuracy matrix.  Maximum scale recommended for this dataset is 1:24,000; it is designed for analyses at the town or subwatershed level.  For more complete information, read the metadata.

image showing MELCD imperviousness
MELCD 2004 imperviousness data in Bangor, black is impervious (with rivers in blue for reference - they would be considered pervious).


<Previous: Identify tool    Next: NLCD products>