Michael N. DeMers

Forty-eight years of landscape change on two contiguous Ohio landscapes

This study analyzes the current and historic structure of two contiguous, rural landscapes covering approximately 242 km2 in central Ohio, USA: a till plain landscape with relatively homogeneous topography and soils, and a moraine landscape with greater geomorphological diversity and heterogeneity. These landscapes were chosen because they were both heavily dominated by agriculture during 1900–1940 and were both initially surveyed by the metes-and-bounds system. They differed, however, in the temporal pattern of settlement and development and in the inherent agricultural capability of their soils. We combined analysis of aerial photographs from 1940, 1957, 1971, and 1988 with historical archives and other available mapped data in a GIS data base to facilitate analysis of both spatial and temporal patterns of change. On the moraine, the agricultural matrix decreased over time as forest, urban/suburban areas, and industry increased. In contrast, on the till plain agricultural landcover increased through 1988, with concommitant decreases in upland forest and oak savanna. The moraine landscape exhibited greater diversity and equitability than the till plain on each date. The till plain had its greatest diversity and equitability in 1940, whereas the moraine increased in diversity and equitability during each time period. The undulating topography of the moraine encouraged landcover dynamism rather than stability, whereas the more homogeneous till plain exhibited considerable inertia. Patch and matrix shape remained constant and predominantly angular over the 48 year study period. Differences in the physical environment, especially topography and soil capability, and the socioeconomic environment, especially agricultural policies and patterns of urbanization, resulted in these two contiguous landscapes having different trajectories of change. It is clear from this study that socioeconomic factors must be combined with the physical setting to fully understand patterns of change in human-dominated landscapes.

Comparative evolution of statewide Geographic Information Systems in Ohio

Abstract This paper describes four autonomous Geographic Information Systems operating at a statewide level in Ohio (USA). All were initiated at various times in the 1970s and 1980s. They are operated by branches of state government, namely the Department of Natural Resources, the Ohio Environmental Protection Agency, the Ohio Department of Transportation, and the Public Utilities Commission of Ohio. A critical review of the operations and institutional contexts of these four systems shows evolution from in-house software development to third party suppliers, from coarse resolution grid cells to high accuracy vector data, from large dedicated GIS staffs to application-specific users augmented by smaller GIS support staffs, and from agency-dependent databases to inter-agency database sharing as technology improves and user needs and databases increase.

The importance of site assessment in land use planning: a re-examination of the SCS LESA model

Abstract The dual paradigm of the USDA Soil Conservation Service Land Evaluation and Site Assessment (LESA) system is the most striking aspect of the model design. Examination of established LESA databases and models for Douglas County, Kansas (USA) indicates the importance of the site assessment subsystem to the planning mission. This emphasis indicates a concern on the part of the US Soil Conservation Service to develop a planning model which is workable within the framework of local needs and design. Results show that the two subsystems, while distinctly separate, work together to plan the proper use of a dwindling agricultural resource base. Suggestions are made for research methodologies to improve the final grading system of LESA to reflect local objectives more adequately while responding to the national need.

Comparative Analysis of Land Evaluation Systems for Douglas County

The USDA Soil Conservation Service LESA system is evaluated and analyzed in two phases. First, using ten years of building permits in two targeted zones in Douglas County, Kansas, a variety of LESA structures are compared. Redundant factors and the impact of different weighting systems are examined. Phase Two follows with a systematic method to determine the Site Assessment factors and their interactions. Results indicate that certain evaluation structures are more reliably consistent and that the systematic approach in Phase Two improves LESA considerably. In addition, it is concluded that the final grading mechanism needs attention, to assure proper protection. Planners are cautioned to consider these issues when implementing LESA for their area.

Policy implications of LESA factor and weight determination in Douglas County, Kansas

The Land Evaluation and Site Assessment System (LESA) developed by the United States Department of Agriculture Soil Conservation Service is designed as an aid to local planning officials to protect important agricultural lands. Preliminary LESA development for Douglas County, Kansas indicates the overall usefulness of the model for its designated purpose within Douglas County. However, model development also indicates several inherent problems with the original system, stemming predominantly from arbitrary weight recommendations and predetermined factor suppositions which do not reflect the needs of a given planning area. The effect of these recommendations and suppositions are most evident when there is a conflict of scale due to incorporation of both national and local standards through the LESA model.

SEDRULE: a rule-based system for interpreting some major sedimentary environments

Abstract SEDRULE is a simple expert system for identification of major sedimentary environments from good outcrops. Although rudimentary, the program, written in muLISP, illustrates the forward-chaining search method. Initially designed as a teaching aid for the novice field sedimentologist, it also can be used to teach some basic principles of LISP programming within sedimentology. Furthermore, this program helps analyze the basic tenets of interpreting depositional environments, thereby helping to examine expert reasoning, which is a goal of AI research. Finally, the program expands its knowledge base as the students own abilities grow, so it provides a lifelong tool for field work.

Adding robustness to linear planning models through simulation

Abstract The planning community concerned with subjective land evaluation criteria can benefit from an iterative simulation model based on a modification of the Kane Simulation language (KSIM). Within a linear planning model designed to protect prime agricultural lands, procedure implementation demonstrates promise for future applications. The interaction matrix is useful for refining factor selection and weighting and for incorporation of factor interactions within weights. Model temporality aids in determining proper planning strategies by allowing examination of future implications and permitting adjustments prior to policy implementation. To implement the model consistently and efficiently, further refinement is necessary. Some method of factor reduction prior to application of the cross-impact matrix is necessary to reduce model development time, to eliminate extensive explanations of the KSIM graphics, and to promote general model understanding.

Resolution tolerance in an automated forest land evaluation model

Abstract The United States Department of Agriculture Soil Conservation Service (SCS) Land Evaluation and Site Assessment (LESA) model was developed in direct response to the National Farmland Protection Policy Act as a consequence of ever increasing demands on the land for nonagricultural uses. LESAs land evaluation (LE) subsystem is a portion of the model designed to determine physical quality of the land for agricultural uses. Previous research has shown that parts of the LESA model can be implemented in automated geographic information systems (GIS). However, no implementation of the forest LE subsystem has been implemented on a CIS, therefore requiring evaluation of land for forestry to be performed manually, one land parcel at a time. Nor have any studies indicated the necessary cell resolution for later incorporation of the LE subsystem within the LESA model. This paper describes a prototype implementation of the forest LE subsystem within a micro-computer based GIS at four separate resolutions. The objective of this project was to test its adaptability to GIS implementation and to ascertain its sensitivity to resolution changes. Results show that the forest LE subsystem is readily adaptable both to GIS implementation and final merger with the site assessment (SA) portion of the LESA model. The model showed little change in outcome as a result of resolution change. Categorical simplification of the modeling process and the broad class values utilized produces this lack of resolution sensitivity, demonstrating that resolution can be selected based on the needs of the SA subsystem.