habitat codes include background, natural water, artificial water, fresh marsh, intermediate marsh, 

 brackish marsh, saline marsh, forest, swamp, shrub/scrub, shrub/scrub (spoil), agriculture/pasture, 

 developed, aquatic vegetation, inert, and beach. 



The land change statistics for the plan were obtained by using MATRIX to compare the 1956 

 and 1978 aggregated habitat maps. The habitat classes for 1956 and 1978 were reeoded to produce 

 maps consisting of two subjects-land and water-for each year. It is important to note that CMD 

 considers aquatic vegetation a water class when deriving land loss statistics. MATRIX compared 

 the 1956 and 1978 land and water maps to produce a change map consisting of four classes: 1) 

 water 1956 to water 1978, 2) land 1956 to water 1978, 3) water 1956 to land 1978, and 4) land 

 1956 to land 1978. MATRIX compared each cell in the 1956 map with its corresponding cell in 

 the 1978 map and assigned one of the four change classes to that cell. The 1956/1978 land and 

 water change map for the Little Pecan Plan (Figure 11) shows that the area had 9.07% land loss 

 and 0.70% land gain. The MATRIX program not only calculates numerical change values but also 

 produces a map showing where land loss has occurred. Plan managers can use a change map to 

 identify problem areas with high land loss rates or to differentiate between various types of land 

 loss occurring within a plan's boundaries. For example, the Little Pecan Plan change map shows 

 shoreline erosion on Little Pecan Lake, areas of interior marsh deterioration in the southern 

 portion of the plan, and the addition of oil field location canals in the eastern half of the plan. 



Other change combinations of 1956 and 1978 habitat classes can also be obtained using 

 MATRIX. Vegetation change maps can be produced showing changes in marsh type, forest, or 

 swamp over the 22-year period. Care should be used when interpreting change statistics as the 

 habitat data reflects conditions existing when the aerial photograph was taken. Meteorological 

 conditions or management practices could change water levels within a plan, causing a 

 misunderstanding of land loss or gain data. 



Land Loss Density Maps. ERDAS can also be used to generate land loss density maps. A land 

 loss density map is used to locate areas that have land loss in excess of a user-specified base value. 

 The land loss density map for the Little Pecan Wetland Management Plan (Figure 12) shows all 

 areas of land loss that are greater than 608 m 2 acres. The land loss density map for the plan was 

 created by scanning a land loss map of the plan with a roving window composed of 9 cells (30x30 

 m). The central cell of the window is sequentially centered on each cell of the land loss map. The 

 cell is then assigned a value based on the number of land loss cells occurring within the window. 

 For example, if 3 cells are counted within the window, the central pixel for the density map is 

 assigned a land loss density value of 3 which equates to 300 m 2 . The land loss density for a full 

 window equals 892 m 2 . By using filtering techniques to eliminate land loss values below a certain 

 size, a map can be produced which identified problem areas on management plans. Land gain 

 maps can also be created using the same method. 



Land and Water Interface Density Maps. A land and water interface density map is produced 

 using techniques similar to those used for creating the land loss density map. A roving cell window 

 is used to scan shorelines on the 1956 and 1978 plan maps. Shorelines which have the greatest 

 complexity are assigned a higher density value than simple shorelines. An example of a complex 

 shoreline would be the southwestern quadrant of the Little Pecan Wetland Management Plan. 

 Many small ponds and natural channel meanders occur there, increasing the total shoreline area. 

 ERDAS assigns a graduated color scale to these density values. The purpose for creating the 1956 

 and 1978 land and water interface density maps is to provide plan managers with a visual means 

 of identifying areas of complex shorelines within plans. These sights may be desirable habitats for 

 certain species. ERDAS can provide a visual representation of shoreline complexity within plans 



338 



