Andreas Rienow

Supporting SLEUTH – Enhancing a cellular automaton with support vector machines for urban growth modeling

Abstract In recent years, urbanization has been one of the most striking change processes in the socioecological system of Central Europe. Cellular automata (CA) are a popular and robust approach for the spatially explicit simulation of land-use and land-cover changes. The CA SLEUTH simulates urban growth using four simple but effective growth rules. Although the performance of SLEUTH is very high, the modeling process still is strongly influenced by stochastic decisions resulting in a variable pattern. Besides, it gives no information about the human and ecological forces driving the local suitability of urban growth. Hence, the objective of this research is to combine the simulation skills of CA with the machine learning approach called support vector machines (SVM). SVM has the basic idea to project input vectors on a higher-dimensional feature space, in which an optimal hyperplane can be constructed for separating the data into two or more classes. By using a forward feature selection, important features can be identified and separated from unimportant ones. The anchor point of coupling both methods is the exclusion layer of SLEUTH. It will be replaced by a SVM-based probability map of urban growth. As a kind of litmus test, we compare the approach with the combination of CA and binomial logistic regression (BLR), a frequently used technique in urban growth studies. The integrated models are applied to an area in the federal state of North Rhine-Westphalia involving a highly urbanized region along the Rhine valley (Cologne, Dusseldorf) and a rural, hilly region (Bergisches Land) with a dispersed settlement pattern. Various geophysical and socio-economic driving forces are included, and comparatively evaluated. The validation shows that the quantity and the allocation performance of SLEUTH are augmented clearly when coupling SLEUTH with a BLR- or SVM-based probability map. The combination enables the dynamical simulation of different growth types on the one hand as well as the analyses of various geophysical and socio-economic driving forces on the other hand. The SVM approach needs less variables than the BLR model and SVM-based probabilities exhibit a higher certainty compared to those derived by BLR.

Geosimulation of urban growth and demographic decline in the Ruhr: a case study for 2025 using the artificial intelligence of cells and agents

AbstractThe Ruhr is an “old acquaintance” in the discourse of urban decline in old industrialized cities. The agglomeration has to struggle with archetypical problems of former monofunctional manufacturing cities. Surprisingly, the image of a shrinking city has to be refuted if you shift the focus from socioeconomic wealth to its morphological extension. Thus, it is the objective of this study to meet the challenge of modeling urban sprawl and demographic decline by combining two artificial intelligent solutions: The popular urban cellular automaton SLEUTH simulates urban growth using four simple but effective growth rules. In order to improve its performance, SLEUTH has been modified among others by combining it with a robust probability map based on support vector machines. Additionally, a complex multi-agent system is developed to simulate residential mobility in a shrinking city agglomeration: residential mobility and the housing market of shrinking city systems focuses on the dynamic of interregional housing markets implying the development of potential dwelling areas. The multi-agent system comprises the simulation of population patterns, housing prices, and housing demand in shrinking city agglomerations. Both models are calibrated and validated regarding their localization and quantification performance. Subsequently, the urban landscape configuration and composition of the Ruhr 2025 are simulated. A simple spatial join is used to combine the results serving as valuable inputs for future regional planning in the context of multifarious demographic change and preceding urban growth.

Comparing support vector machines with logistic regression for calibrating cellular automata land use change models

ABSTRACT Land use change models enable the exploration of the drivers and consequences of land use dynamics. A broad array of modeling approaches are available and each type has certain advantages and disadvantages depending on the objective of the research. This paper presents an approach combining cellular automata (CA) model and support vector machines (SVMs) for modeling urban land use change in Wallonia (Belgium) between 2000 and 2010. The main objective of this study is to compare the accuracy of allocating new land use transitions based on CA-SVMs approach with conventional coupled logistic regression method (logit) and CA (CA-logit). Both approaches are used to calibrate the CA transition rules. Various geophysical and proximity factors are considered as urban expansion driving forces. Relative operating characteristic and a fuzzy map comparison are employed to evaluate the performance of the model. The evaluation processes highlight that the allocation ability of CA-SVMs slightly outperforms CA-logit approach. The result also reveals that the major urban expansion determinant is urban road infrastructure.

Urban Development in West Africa—Monitoring and Intensity Analysis of Slum Growth in Lagos: Linking Pattern and Process

Upgrading all slums in Lagos by 2030 will an ambitious task, given that more than 70% of its residents resides in slums. Furthermore, there is no recent study identifying neither the slums nor their temporal growth/development pattern in Lagos that can backstop any slum management initiative. This study aims to contribute by applying object-based image analysis and intensity analysis to map and link patterns and processes of slum growth in Lagos. RapidEye imagery from 2009 and 2015 were used to create maps for each time point for six land use categories (water, vegetated area, open space, road, slum, and other urban). Intensity analysis was applied to quantify the annual intensity of changes at the category and transition level. An overall accuracy (and kappa coefficient) of 94% (0.9) and 89% (0.86) were achieved for the 2009 and 2015 land use and land cover maps, respectively. This study showed that slums in Lagos have increased spatially during the time interval studied, with a total net gain of 9.18 square kilometers, influenced by the increase in population, mainly due to in-migration to Lagos. However, this study also revealed that slums were actively losing and gaining land area between 2009 and 2015, with an annual gain and loss intensity of 10.08 and 6.41, respectively, compared to the uniform intensity of 3.15. The gain was due to poor maintenance of buildings and encroachment onto available spaces (water and open space), while the loss was attributed to gentrification and demolition processes. A systematic process of transition was observed between slums and other urban (and open space) areas in the interval studied, and this process was mainly influenced by the Lagos state government. This analysis is crucial for designing policy interventions to manage slum growth in Lagos.

Eine Zukunft, drei Perspektiven – räumlich explizite Prognose des Flächenverbrauchs für die Metropolregion Rheinland 2030

Zur Vorhersage zukunftiger Veranderungen der Landnutzung wurde eine Simulation unter drei unterschiedlichen Grundannahmen durchgefuhrt: A) nachhaltige Entwicklung, B) Business as usual, C) Wunsch nach Wohnen im Grunen. In allen drei Szenarien nimmt die Siedlungs- und Verkehrsflache zu, wobei in Szenario A vor allem bestehende Siedlungsraume verdichtet werden, wahrend in Szenario C vorrangig neue, „grune“ Flachen erschlossen werden.

Dreißig Hektar täglich? Ein Blick in die Vergangenheit und Status quo

Im vorliegenden Beispiel wurden Satellitenbilder der Metropolregion Rheinland fur 1975, 1984, 2001 und 2015 klassifiziert. Die Ergebnisse zeigen, dass die versiegelten Flachen seit 1975 konstant zunehmen, wahrend Acker-, Wiesen- und Weideflachen abnehmen. Diese Landnutzungsveranderungen haben nicht nur Einfluss auf unsere unmittelbare Lebensumwelt, sie haben auch messbare Folgen fur den Strahlungs- und Energiehaushalt und somit fur das Mikro- und Mesoklima.

Landschaftsverbrauch und Klimaanpassung: der Blick über die Metropolregion hinaus

Aufgrund des derzeit beobachtbaren Klimawandels betrifft das Themenfeld Flachenverbrauch und Landschaftsstrukturwandel insbesondere den dicht besiedelten Raum und verlangt nach Masnahmen zur Erhohung der Resilienz. Um die vorhandenen Potenziale auszuschopfen und die individuelle Handlungsbereitschaft zu wecken, ist es sinnvoll, ein Bewusstsein fur den Klimawandel und die Querschlusse zum Landnutzungswandel sowie den damit einhergehenden Anpassungsbedarf zu schaffen.

Technical note: using ISS videos in Earth observation – implementations for science and education

ABSTRACT A large variety of various passive and active satellite sensors producing panchromatic, multispectral or hyperspectral images of the Earth’s surface are currently in space and are used in different scientific fields such as earth science (e.g. geography, hydrology, geology, oceanography and glaciology), disaster management, as well as for military, commercial and economic purposes. In contrast, video files are a rather exotic data format in the field of Earth observation. Occasionally, video cameras are used in airborne remote sensing, but only recently video Earth observation from space has been established. This paper aims at the integration of video data in the scientific workflow, revealing advantages and disadvantages of moving images. Being the only available source for continuous video Earth observation data, the NASA experiment High Definition Earth Viewing on-board the International Space Station is the basis for our evaluation of the usability of video data. Following the rather coarse resolution of these data, we exemplify some potential fields of application in science and education. We show a basic workflow how to produce 3D models and stereoscopic videos of the Earth’s surface. As a pre-study for videos with better spatial as well as radiometric resolution, the delivered products serve not only scientific purposes, but are integrated in school lessons to evoke the pupils’ fascination for earth sciences and space.

In Memoriam: Gunter Menz

Remote Sensing Research Group, Department of Geography, University of Bonn, Meckenheimer Allee 166,53115 Bonn, Germany; andreas.rienow@ruhr-uni-bochum.de (A.R.); agatha.okhai@gmail.com (A.A.);hendrik.fluegel@googlemail.com (H.F.); s.heinemann@uni-bonn.de (S.H.); s6tohenn@uni-bonn.de (T.H.);k.hentze@geographie.uni-bonn.de (K.H.); f.kirimi@geographie.uni-bonn.de (F.K.);s6sokoch@uni-bonn.de (S.K.); bkoner@uni-bonn.de (B.K.-K.); s3anortw@uni-bonn.de (A.O.);schultz@giub.uni-bonn.de (J.S.); kstaar@uni-bonn.de (K.S.); kuria@geographie.uni-bonn.de (K.B.T.);anna.vetter@uni-bonn.de (A.V.)