GAMOLCS 2017 Abstracts


Short Papers
Paper Nr: 1
Title:

Improving SLEUTH Calibration with a Genetic Algorithm

Authors:

Keith C. Clarke

Abstract: A review of calibration methods used for cellular automaton models of land use and land cover change was performed. Calibration advances have been achieved through machine learning algorithms to either extract land change rules, or optimize model performance. Many models have now automated the calibration process, reducing the need for subjective choices. Here, the brute force calibration procedure for the SLEUTH CA-based land use change model was replaced with a genetic algorithm (GA). The GA calibration process populates a “chromosome” with five parameter combinations (genes). These combinations are then used for model calibration runs, and the most successful selected for mutation, while the least successful are replaced with randomly selected values. Default values for the constants and rates of the genetic algorithm were selected from SLEUTH applications. Model calibrations were completed using both brute force calibration and the GA. The GA model performed as well as the brute force method, but used vastly less computation time with speed up of about 3 to 22. The optimal values for GA calibration are set as the defaults for SLEUTH-GA, a new version of the model. This paper is a contraction of Clarke (in press), which reports on the full set of results.

Paper Nr: 2
Title:

Cartographic Scale and Minimum Mapping Unit Influence on LULC Modelling

Authors:

David García Álvarez

Abstract: Two models at two different scales (1:25.000 and 1.100.000) were calibrated using two different Land Use and Land Cover maps at such cartographic scales (SIOSE and CORINE) and with a different Minimum Mapping Unit (0.2-0.5ha and 25ha). Differences between models were assessed through cross-tabulation analysis (quantity and allocation disagreement) and spatial metrics (pattern disagreement). The models results have been very different depending on the scale considered, although most of the disagreement comes from the contrasting input maps. In any case, the scale at which the models were calibrated have proved to influence the pattern modelled and the quantity and allocation of changes.

Paper Nr: 3
Title:

Mapping Socio-biodiversity: Do Old Modelling Tools Suit New Challenges?

Authors:

Sónia Carvalho Ribeiro, William Leles da Costa, Amanda Ribeiro de Oliveira, Danilo da Silveira Figueira, Isabella Lorenzini da Silva Teixeira, Lilian Aline Machado, Herman Rodrigues Oliveira and Britaldo Silveira Soares Filho

Abstract: This work shows an original use of classical methods in land change modelling. The aim of this study is to model yields (productivity) and economic importance (annual rents) of rubber and Brazil nut in the Brazilian Amazon. Biophysical variables related to rubber and Brazil nut yields as well as market access (commercialization) were used to model favorability of productivity using Weights of Evidence (WofE) method. To favorability of productivity were assigned yields base on case study data. The economic model then combines the map of yields with output prices and costs of collection, processing, and transport to estimate annual rents per hectare for a specific forest plot. For estimating transport costs we used cost friction surface modelling tools. Our results show that yields for Brazil nut averages 8.19±7.41 kg ha-1year-1 and rent averages US$ 5.05±7.49 ha-1year-1. Rubber average yields is of 3.53 kg/ha/year and rubber rents average US$ 0.56±0.7ha-1year-1. Coupling biophysical and economic models allowed us to explore which environmental and governance improvements are needed to avoid deforestation and forest degradation in the Brazilian Amazon. Our results also show that despite some methodological issues and the recurrent call for “new” modelling approaches for addressing the complexity of socio ecological systems, “old” modelling tools such as Weight of Evidence and Cost Friction Surface, are still suited for addressing the challenge of mapping socio-biodiversity.

Paper Nr: 4
Title:

Modeling Land Change using One or Two Time Points based Calibration - A Comparison of Factors

Authors:

María Teresa Camacho Olmedo

Abstract: One of land change model parameters in calibration step relates to how changes over time and space are considered in the model. A land change model can be calibrated with the state at one time point or with the difference between two time points. The purpose is describing land use and cover (LUC) state patterns, i.e. one time point calibration, and LUC transition patterns, i.e. two time points. For a case study in Spain we obtained the collections of factors for two calibration periods at one time point (dates 2000 and 2006) and the collections of factors for two calibration periods between two time points (periods 1990-2000 and 2000-2006). Evidence likelihood is used to transform the explanatory variables into factors. The objective of this paper is to compare these four collections of factors to show how the choice of reference maps influences the factors and how these factors highlight the change patterns in two different calibration periods and in the calibration of two models. As a following step the detailed results for the different factors and LUC categories are analysed.

Paper Nr: 5
Title:

Land Change Modeling Handling with Various Training Dates

Authors:

Martin Paegelow

Abstract: Popular modeling tools for land change simulation, especially those using Markov chains, undertake model training based only on two land use / cover (LUC) maps. This paper analyses uncertainty and potential errors caused by taking into account only two former, model known, LUC maps. This is illustrated by a simple data set of six LUC maps allowing various Markovian transition matrices; a range even larger by considering different confidence levels. Results underline the randomness in choice of only two training dates. Authors propose alternative methods to Markov chains integrating all available LUC maps in order to simulate forecasting scenarios. To do so, they incorporate all possible LUCC (land use / cover change) budgets to perform simple arithmetic combinations between the six training dates. Comparing Markov chain transitions based on two training dates and alternatively performed change rates taking into account all training dates results to important differences. This study underlines the importance of the choice of training dates during model calibration for path-dependent simulations.

Paper Nr: 6
Title:

Modelling Transport-based Land-use Scenarios in Bogota

Authors:

Francisco Escobar and Daniel Paez

Abstract: Economic growth experienced in Colombia since 2001 has impacted on heavier traffic levels in the capital city of Bogota which in turn have worsened air pollution indicators and environmental public health conditions. Different political options competing at municipal elections have included their respective proposals for public transport in their programs. Impact expected from each of these scenarios makes it necessary to implement models allowing their assessment. Given this need, we present the Bogota Land Development Model (BoLD), a practical implementation of a Land-use Cover Change (LUCC) simulation based on two different public transport scenarios; a highway-based network and a suburban rail system. Transport scenarios are combined with options to expand the city into natural reserves. Customized geospatial analyses were developed for calculating accessibility distance decay factors based on overtime-spatial decay determination (OSDD) method. Results of the scenarios are presented both in maps and in “mobility circles”. Validation of the results suggests that OSDD and the mobility circles appear to contribute to better information to decision-making when evaluating urban scenarios driven by transport projects.

Paper Nr: 7
Title:

Using Constraint Cellular Automata to Simulate Urban Development in a Cross-border Area

Authors:

J.-P. Antoni, V. Judge, G. Vuidel and O. Klein

Abstract: Urban sprawl and space consumption have become key issues in sustainable territorial development. Traditional planning approaches are often insufficient to anticipate their complex spatial consequences, especially in cross-border areas. Such complexity requires the use of dynamic spatial simulations and the development of adapted tools like LucSim, a CA-based tool offering solutions for sharing spatial data and simulations among scientists, technicians and stakeholders. Methodologically, this tool allows us to simulate future land use change by first quantifying and then locating the changes. Quantification is based on Markov chains and location on transition rules. The proposed approach is implemented on the Strasbourg-Kehl cross-border area and calibrated with three contrasting prospective scenarios to try to predict cross-border territorial development.

Paper Nr: 8
Title:

Future Land Use Change Dynamics in Natural Protected Areas - Madrid Region Case Study

Authors:

Marta Gallardo and Javier Martinez-Vega

Abstract: Natural protected areas are declared to safeguard their environment, goods and services. However, sometimes they are affected by land use changes related to human activity, which affects their ecosystem functions and their sustainability. Problems such as fragmentation or low habitat connectivity are some of its consequences. Developing future land use scenarios is essential if a preventive approach to the management of protected areas is to be adopted. In this paper, three different land use change scenarios in natural protected areas in Madrid region are modelled: a “business as usual” scenario, an economic crisis scenario and a green scenario. All protected areas are studied, from National and Nature Parks to Special Areas of Conservation and Special Protection Areas; changes in a buffer area of 5 km around PAs are also studied. The CLUE model (based on logistic regression) is used. Biophysical, socio-economic and accessibility factors and incentives and restrictions are considered. In recent decades, the region of Madrid has experienced intense urban and infrastructure development (48,332 ha). Protected areas have been affected by this urbanization process (almost 5,000 ha) and its surroundings (30,000 ha). These findings should alert land use planners and the managers of protected areas to the potential threats.