Spatial neighborhood effect and scale issues in the calibration and validation of a dynamic model of Phragmites australis distribution – A cellular automata and machine learning approach

Altartouri, Anas; Nurminen, Leena; Jolma, Ari

doi:10.1016/j.envsoft.2015.04.010

Cited by 18 publications

(8 citation statements)

References 92 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The same similarity was observed for the Kappa Simulation index of our model (23.9%) and of others (e.g., 23% in Malek et al [96]; 12%-42% in Ke et al [98] and 15%-52% in Ke et al [99]). Finally, results from the Fuzzy Kappa Simulation analysis (34.1%-35% for moving windows of 2-, 4-, 8-, 16-pixel radius) were also in conformance with the results reported by Altartouri et al [100] (3%-30% for 3 × 3-to 9 × 9-pixel moving windows). Moreover, the K Transition (82.7%) and K TransLoc (28.8%) components of the Kappa Simulation indicate that our model better predicts the area of soy expansion than its allocation [90].…”

Section: Discussionsupporting

confidence: 90%

Simulated Impacts of Soy and Infrastructure Expansion in the Brazilian Amazon: A Maximum Entropy Approach

Frey

West

Hickler

et al. 2018

Forests

View full text Add to dashboard Cite

Historically, the expansion of soy plantations has been a major driver of land-use/cover change (LUCC) in Brazil. While a series of recent public actions and supply-chain commitments reportedly curbed the replacement of forests by soy, the expansion of the agricultural commodity still poses a considerable threat to the Amazonian and Cerrado biomes. Identification of areas under high risk of soy expansion is thus paramount to assist conservation efforts in the region. We mapped the areas suitable for undergoing transition to soy plantations in the Legal Amazon with a machine-learning approach adopted from the ecological modeling literature. Simulated soy expansion for the year 2014 exhibited favorable validation scores compared to other LUCC models. We then used our model to simulate how potential future infrastructure improvements would affect the 2014 probabilities of soy occurrence in the region. In addition to the 2.3 Mha of planted soy in the Legal Amazon in 2014, our model identified another 14.7 Mha with high probability of soy conversion in the region given the infrastructure conditions at that time. Out of those, pastures and forests represented 9.8 and 0.4 Mha, respectively. Under the new infrastructure scenarios simulated, the Legal Amazonian area under high risk of soy conversion increased by up to 2.1 Mha (14.6%). These changes led to up to 11.4 and 51.4% increases in the high-risk of conversion areas of pastures and forests, respectively. If conversion occurs in the identified high-risk areas, at least 4.8 Pg of CO2 could be released into the atmosphere, a value that represents 10 times the total CO2 emissions of Brazil in 2014. Our results highlight the importance of targeting conservation policies and enforcement actions, including the Soy Moratorium, to mitigate future forest cover loss associated with infrastructure improvements in the region.

show abstract

Section: Discussionsupporting

confidence: 90%

Simulated Impacts of Soy and Infrastructure Expansion in the Brazilian Amazon: A Maximum Entropy Approach

Frey

West

Hickler

et al. 2018

Forests

View full text Add to dashboard Cite

show abstract

“…Cell size, neighborhood size, and neighborhood type have highly significant (p < 0.0001) unilateral (X 1 , X 2 , andX 3 ) effects on simulation accuracy and consequently show scale sensitivity (Table 5). This has been demonstrated by previous studies (Kocabas and Dragicevic 2006, Samat 2006, Altartouri et al 2015.…”

Section: Discussionsupporting

confidence: 83%

Examining the sensitivity of spatial scale in cellular automata Markov chain simulation of land use change

Clarke

et al. 2019

International Journal of Geographical Information Science

112

View full text Add to dashboard Cite

Understanding the spatial scale sensitivity of cellular automata is crucial for improving the accuracy of land use change simulation. We propose a framework based on a response surface method to comprehensively explore spatial scale sensitivity of the cellular automata Markov chain (CA-Markov) model, and present a hybrid evaluation model for expressing simulation accuracy that merges the strengths of the Kappa coefficient and of Contagion index. Three Landsat-Thematic Mapper remote sensing images of Wuhan in 1987, 1996, and 2005 were used to extract land use information. The results demonstrate that the spatial scale sensitivity of the CA-Markov model resulting from individual components and their combinations are both worthy of attention. The utility of our proposed hybrid evaluation model and response surface method to investigate the sensitivity has proven to be more accurate than the single Kappa coefficient method and more efficient than traditional methods. The findings also show that the CA-Markov model is more sensitive to neighborhood size than to cell size or neighborhood type considering individual component effects. Particularly, the bilateral and trilateral interactions between neighborhood and cell size result in a more remarkable scale effect than that of a single cell size.

show abstract

“…SLEUTH (Clarke et al, 1997, Jafarnezhad et al, 2016, Onsted and Chowdhury, 2014, Rienow and Goetzke, 2015, Sakieh et al, 2015, Silva and Clarke, 2002 is calibrated through a two-step procedure: first, a visual calibration oriented for a broad parameter definition and debugging; second, a brute force calibration procedure where multiple runs are produced in order to generate enough model data to statistically compare reference data. Li and Yeh (2002) and Li et al (2013), Almeida et al (2008), Basse et al (2014) and Altartouri et al (2015) coupled CA models with artificial neural network formulations to calibrate transition rules and other components of the CA models. , and Li et al (2014) used basic sensitivity analysis to calibrate the weighting parameters for the spatial interactions between land uses.…”

Section: A Brief Literature Survey On Cellular Automata Models and Thmentioning

confidence: 99%

Applicability and calibration of an irregular cellular automata model for land use change

Pinto

Antunes

Roca

2017

Computers, Environment and Urban Systems

View full text Add to dashboard Cite

Cellular automata (CA) models of spatial change have been developed and applied in the context of large regional or metropolitan areas and usually use regular cells, with spatial interactions and transition rules operating within fixed-size neighbourhoods. Model calibration has also been an area of intensive research with many models still using expertbased input to ensure visual calibration of modelled land use maps. In this paper, we present an innovative CA model where irregular cells and variable neighbourhoods are used to better represent space and spatial interaction. Calibration is based on an optimisation procedure that uses particle swarm (PS) to determine the optimal set of parameters of the CA model. Hypothetical test instances are used to assess the CA model and its calibration to small urban areas. Our conclusion was that the use of PS ensures calibration results for the CA model that compare very well with results obtained through other approaches reported in the literature.

show abstract

Spatial neighborhood effect and scale issues in the calibration and validation of a dynamic model of Phragmites australis distribution – A cellular automata and machine learning approach

Cited by 18 publications

References 92 publications

Simulated Impacts of Soy and Infrastructure Expansion in the Brazilian Amazon: A Maximum Entropy Approach

Simulated Impacts of Soy and Infrastructure Expansion in the Brazilian Amazon: A Maximum Entropy Approach

Examining the sensitivity of spatial scale in cellular automata Markov chain simulation of land use change

Applicability and calibration of an irregular cellular automata model for land use change

Contact Info

Product

Resources

About