Predicting land cover using GIS, Bayesian and evolutionary algorithm methods

Aitkenhead, Matthew; Aalders, Inge

doi:10.1016/j.jenvman.2007.09.010

Cited by 55 publications

(36 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After testing the two most commonly used discretization techniques (equal width and equal frequency) along with the popular 'minimum description length principle' algorithm (Fayyad and Irani, 1993), this study found that equal frequency discretization provided the best results (lowest error when internally validated). This is consistent with the conclusion of Aitkenhead and Aalders (2009) that when some categories within a landscape are more prevalent than others, a frequency approach often gives a better representation.…”

Section: Bn Model Developmentsupporting

confidence: 90%

“…(1), the posterior probability P(A|B) was an unknown x, we now see that it can be calculated using our prior belief in the occurrence of event A P(A) and event B P(B) and the probability of B given that A has occurred P(B|A). This is known as Bayesian inference and to illustrate how this might work in practice for DSM applications, we adapt an example given by Aitkenhead and Aalders (2009).…”

Section: Theorymentioning

confidence: 99%

See 1 more Smart Citation

On the application of Bayesian Networks in Digital Soil Mapping

et al. 2015

View full text Add to dashboard Cite

Section: Bn Model Developmentsupporting

confidence: 90%

Section: Theorymentioning

confidence: 99%

On the application of Bayesian Networks in Digital Soil Mapping

et al. 2015

View full text Add to dashboard Cite

“…Potter et al (2000) state that forest ecosystem management implies the need to forecast the future state of complex systems, which often experience structural changes. It is by means of strategic planning that ecological integrity and sustainability (Gustafson y Rasmussen, 2002), risk management (Borchers, 2005y Heinimann, 2010 and future landscape (Aitkenhead and Aalders, 2009) are guaranteed. Naesset (1997) stresses the importance of the integration of GIS with quantitative models for long term forest management.…”

Section: Development and Current Situation Of Forest Dssmentioning

confidence: 99%

Decision Support Systems for Forestry in Galicia (Spain): SaDDriade

Pérez¹,

Franco-Vázquez²,

Álvarez-López³

2012

Sustainable Forest Management - Current Research

View full text Add to dashboard Cite

“…Bayesian Networks have also been successfully applied in numerous environmental management scenarios such as predicting land cover (Aitkenhead and Aalders 2009), forecasting change in vegetation type/cover (Liedloff and Smith 2010) and evaluating the impact of land management schemes on species habitat (Marcot et al 2001). While no research was found that specifically analysed deforestation using a Gaussian process, the algorithm's similarity with spatial techniques such as kriging (Rasmussen and Williams 2006) computational complexity, overall accuracy, sensitivity, ability to handle poor data or low prevalence rates, and interpretability.…”

Section: Albert Einsteinmentioning

confidence: 99%

“…There are several demonstrated ways that machine learning is able to assist with this. It has been used to directly analyse satellite data to estimate tree height and canopy cover (Clark et al 2010, Stojanova et al 2010 as well as predicting future land use change (Aitkenhead andAalders 2009, Liedloff andSmith 2010). Secondly, other tools that have been developed in the machine learning sector are able to help model deforestation with a view to understanding both the proximate and indirect causes of deforestation (Casse et al 2004, Pineda Jaimes et al 2010.…”

Section: Albert Einsteinmentioning

confidence: 99%

Making the most of machine learning and freely available datasets: a deforestation case study

Mayfield¹

View full text Add to dashboard Cite

There are many reasons why we study deforestation including predicting at risk areas, predicting deforestation rate and informing the development of conservation policies and programs. Each study will have its own set of objectives to meet (such as setting a deforestation baseline or advising on forest protection policies) and constraints to work within (such as time and data constraints and access to experts). This thesis develops a framework for helping to decide which of several statistical and machine learning methodologies; generalised linear models (GLMs), generalised linear mixed models (GLMMs), artificial neural networks (ANNs), Bayesian networks (BNs) and Gaussian processes (GPs) might be suitable for a given deforestation study.One common constraint on deforestation studies is data availability, as it is often not possible to acquire all the datasets that would ideally be included. High resolution demographic or socioeconomic information can be costly, and obtaining the value of dynamic variables such as road location for the correct point in time may be difficult. By using either freely available or low cost datasets to generate the variables for this thesis, it was possible to evaluate the usefulness of these data in predicting deforestation and identifying its predisposing factors. Their proven utility demonstrates that they could provide effective substitutes in those cases where the ideal datasets are not available.The main datasets used were the Conservation International land use change data for southern Mexico and north-eastern Madagascar, which are raster datasets at 30 m resolution showing forest loss for either two (Mexico) or three (Madagascar) time steps. Predictor variables were also generated from the World Database on Protected Areas, the NASA Landsat digital elevation model and several Natural Earth datasets on city and river location. Random sample points were generated across the forested areas of the study zones and models were then trained to predict whether there would be any deforestation within a 500 m x 500 m zone around each point.Models were implemented using either R or Matlab and their performance was evaluated using sensitivity, specificity, true skill statistic and the area under the receiver operating curve. The results of the best performing model designs for each methodology were mapped to examine whether the predicted high risk areas were close to where actual deforestation occurred. Separate maps were produced for the predicted results at a 50 % probability cut off, as well as across all probabilities to produce a map of predicted high risk areas. Additional maps were created showing the predictions after correcting for the rate of expected deforestation.ii When applied to complex problems such as deforestation analysis, machine learning (ML) techniques have several theoretical and practical advantages over classical statistics, primarily the ability to take into account non-linear relationships. The ML models were therefore expected to outperform the simpler...

show abstract

Predicting land cover using GIS, Bayesian and evolutionary algorithm methods

Cited by 55 publications

References 33 publications

On the application of Bayesian Networks in Digital Soil Mapping

On the application of Bayesian Networks in Digital Soil Mapping

Decision Support Systems for Forestry in Galicia (Spain): SaDDriade

Making the most of machine learning and freely available datasets: a deforestation case study

Contact Info

Product

Resources

About