Spatial Modeling of Tropical Deforestation Using Socioeconomic and Biophysical Data

Rivera, Samuel; Anguita, Pablo Martínez de; Ramsey, R. Douglas; Crowl, Todd A.

doi:10.1007/s11842-012-9214-2

Cited by 10 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A logistic model was developed based on the binary response variable (one -deforestation; zero -nondeforestation) (Rivera et al 2012) and the explanatory variables (elevation, aspect, slope, distance to roads and residential areas). Before developing the model, the explanatory variables were standardized by dividing values by their root-mean-square because of the easier comparison of the relative effect of each variable (Etter et al 2006).…”

Section: Methodsmentioning

confidence: 99%

Deforestation modelling using logistic regression and GIS

Bavaghar¹

2015

J. For. Sci.

View full text Add to dashboard Cite

ABSTRACT:A methodology has been used by means of which modellers and planners can quantify the certainty in predicting the location of deforestation. Geographic information system and logistic regression analyses were employed to predict the spatial distribution of deforestation and detects factors influencing forest degradation of Hyrcanian forests of western Gilan, Iran. The logistic regression model proposed that deforestation is a function of slope, distance to roads and residential areas. The coefficients for the explanatory variables indicated that the probability of deforestation is negatively related to slope, distance from roads and residential areas. Although the distance factor was found to be a contributor to deforestation, its effect is lower than that of slope. The correlates of deforestation may change over time, and so the spatial model should be periodically updated to reflect these changes. Like in any model, the quality may be improved by introducing the new variables that may contribute to explaining the spatial distribution of deforestation.

show abstract

Section: Methodsmentioning

confidence: 99%

Deforestation modelling using logistic regression and GIS

Bavaghar¹

2015

J. For. Sci.

View full text Add to dashboard Cite

show abstract

“…This means that there is a substantial challenge in managing land for both conservation and livelihoods. Local livelihoods, which are often carried out unsustainably, and conservation objectives are thus often in conflict (Rivera et al 2013, Ericksen et al 2002, Southworth et al 2004). …”

Section: Study Sitementioning

confidence: 99%

Flow and rent-based opportunity costs of water ecosystem service provision in a complex farming system

Rendón¹,

Dallimer²,

Paavola³

2016

E&S

View full text Add to dashboard Cite

rent-based opportunity costs of water ecosystem service provision in a complex farming system. Ecology and Society 21 (4) ABSTRACT. Unsustainable land uses present many challenges for securing ecosystem service provision. It is also difficult to estimate the cost of a transition to more sustainable land-management practices for individual landholders. The main cost to landholders is the opportunity costs, the income foregone when changing land use for continued or enhanced ecosystem service provision. Thus accurate estimation of opportunity costs and understanding their distribution are crucial starting points for determining the economic viability and design of any payment for ecosystem services (PES) scheme. We compare two opportunity cost approaches and examine the distribution of these costs for improving drinking water quality in a complex farming system in a Honduran forest catchment. Data for both approaches was collected through a survey applied to upstream catchment landholders. Our results indicate that the direct flow approach and the proxy rent approach provide comparable and consistent opportunity cost estimates. The mean net flow return ha -1 was US$1410, but this estimate was skewed, mainly by exceptionally high coffee returns and negative returns of land uses making a loss. This estimate would imply spending over US$2 million per annum for water conservation, but a revised estimate comes to US$257,057 per annum. Opportunity costs were found to vary according to differences in land use and landholder characteristics. High value cash crops upholding the local economy, such as coffee, entail much higher opportunity costs than for example cattle grazing. These results suggest that discriminate PES payments, that vary according to opportunity costs and thus discriminate between land uses and landholders, are essential. Water quality at our case study site could be managed sustainably by a scheme focusing on highimpact land uses with lower opportunity costs and closer to water sources.

show abstract

“…A less common approach to region selection is to use the watershed (Rivera et al 2013). This approach is relevant when variables such as distance to roads are considered as a measure of accessibility (Deng et al 2011) as it takes into account that a patch of forest close to a road can still be inaccessible if there is a mountain in between them.…”

Section: Region Selection and Scalementioning

confidence: 99%

“…For example, with a spatial resolution of 0.25 km 2 per pixel and a study area of less than 300 km 2 , the proximity to the nearest house or shelter was included (Ludeke et al 1990). Proximity to nearest house was also included for a study within a watershed of 42 ha (Rivera et al 2013). For a larger study area of around 12,400 km 2 using a grid size of 120 km 2 , the input variable used for representing human habitation was the distance to the nearest settlement and was calculated from manually digitised maps (Mas et al 2004).…”

Section: Effects Of Human Settlementmentioning

confidence: 99%

“…When analysing the connections between deforestation and its predictors, researchers have often relied on classical statistical approaches, such as linear or logistic regression (Ludeke et al 1990, Mertens and Lambin 1997, Ewers 2008, Rivera et al 2013). However, the last decade or so has seen the emergence of studies branching out into the machine learning field (Mas et al 2004, Basse et al 2014.…”

Section: Statistical and Machine Learning Approachesmentioning

confidence: 99%

See 1 more Smart Citation

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

Spatial Modeling of Tropical Deforestation Using Socioeconomic and Biophysical Data

Cited by 10 publications

References 10 publications

Deforestation modelling using logistic regression and GIS

Deforestation modelling using logistic regression and GIS

Flow and rent-based opportunity costs of water ecosystem service provision in a complex farming system

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

Contact Info

Product

Resources

About