Nonparametric estimation of the dependence of a spatial point process on spatial covariates

Baddeley, Adrian; Chang, Ya-Mei; Song, Yong; Turner, Rolf

doi:10.4310/sii.2012.v5.n2.a7

Cited by 77 publications

(76 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Bayes's theorem provided a simple solution to this estimation problem. For instance, to estimate the dependence of the deposition rate on bed slope, Bayes's theorem states that

D_{p} (θ) = (〈〉, D_{p}) f_{Θ | ↓} / f_{Θ}

, where

(〈〉, D_{p})

was the average particle deposition rate (obtained by dividing the total number of deposition events by the total time of particle motion), f Θ | ↓ is the probability density function of bed slope associated with particle deposition, and f Θ was the probability density function of all bed slopes visited by moving particles (see the proof in the supporting information) [ Baddeley et al , ]. The same formula applied to local areal entrainment rates, with the difference that the probability f Θ should not be computed from bed slopes visited by moving particle, but from all possible bed slopes.…”

Section: Methodsmentioning

confidence: 99%

Entrainment, motion, and deposition of coarse particles transported by water over a sloping mobile bed

Heyman

Bohorquez

Ancey

2016

J. Geophys. Res. Earth Surf.

View full text Add to dashboard Cite

International audienceIn gravel bed rivers, bed load transport exhibits considerable variability in time and space. Recently, stochastic bed load transport theories have been developed to address the mechanisms and effects of bed load transport fluctuations. Stochastic models involve parameters such as particle diffusivity, entrainment, and deposition rates. The lack of hard information on how these parameters vary with flow conditions is a clear impediment to their application to real-world scenarios. In this paper, we determined the closure equations for the above parameters from laboratory experiments. We focused on shallow supercritical flow on a sloping mobile bed in straight channels, a setting that was representative of flow conditions in mountain rivers. Experiments were run at low sediment transport rates under steady nonuniform flow conditions (i.e., the water discharge was kept constant, but bed forms developed and migrated upstream, making flow nonuniform). Using image processing, we reconstructed particle paths to deduce the particle velocity and its probability distribution, particle diffusivity, and rates of deposition and entrainment. We found that on average, particle acceleration, velocity, and deposition rate were responsive to local flow conditions, whereas entrainment rate depended strongly on local bed activity. Particle diffusivity varied linearly with the depth-averaged flow velocity. The empirical probability distribution of particle velocity was well approximated by a Gaussian distribution when all particle positions were considered together. In contrast, the particles located in close vicinity to the bed had exponentially distributed velocities. Our experimental results provide closure equations for stochastic or deterministic bed load transport models. ©2016. American Geophysical Union. All Rights Reserved

show abstract

“…Bayes's theorem provided a simple solution to this estimation problem. For instance, to estimate the dependence of the deposition rate on bed slope, Bayes's theorem states that

D_{p} (θ) = (〈〉, D_{p}) f_{Θ | ↓} / f_{Θ}

, where

(〈〉, D_{p})

Section: Methodsmentioning

confidence: 99%

Entrainment, motion, and deposition of coarse particles transported by water over a sloping mobile bed

Heyman

Bohorquez

Ancey

2016

J. Geophys. Res. Earth Surf.

View full text Add to dashboard Cite

show abstract

“…A second way to spatially characterize the population of captured dogs was to produce a graph of the Euclidean distance between the cap- ture site and the ZCC (Baddeley et al, 2012). This analysis contains the estimated distances to a pattern of points (in this case, the capture site) to a reference point (ZCC in this case), thus evaluating the distance with the higher point density.…”

Section: Methodsmentioning

confidence: 99%

Monitoring techniques in the capture and adoption of dogs and cats

et al. 2015

View full text Add to dashboard Cite

The continuous improvement of the information systems of organizations that work toward the control of stray dog and cat populations facilitates the implementation of programmes aimed at reducing the number of animals that roam free in public streets. This study investigated the fate of animals in a Brazilian zoonosis control centre (ZCC) with special reference to euthanasia and adoption. Techniques were applied to visualize the spatial distribution of stray dogs and cats as well as that of the people who adopt these animals. Ripley's K function was used with a Euclidean distance graph to detect the distribution patterns involved. An estimate of the kernel density was used to allow assessment of the spatial distribution of the phenomenon studied. The results show that the distribution of the captured animals, and that of the people who adopted them, form spatial clusters (P=0.01). Most of the animals were captured near the premises of the ZCC and near the downtown area. Factors such as the abandonment of animals near animal control agencies and the availability of food in these areas could play a role for this outcome. The awareness of the people who live in places where there is a greater number of stray animals and the simultaneous presence of an urban population may explain the concentration of adoptions in these areas. The results may contribute to the implementation of improved measures for the treatment of stray animals.

show abstract

“…While the form of this equation is loglinear, it can readily capture nonlinear relationships between intensity and the environment, for example, using quadratic and interaction terms, smoothed functions in generalised additive models (GAMs, Hastie & Tibshirani ) or via kernel regression (Guan ; Baddeley et al . ).…”

Section: Point Process Modelsmentioning

confidence: 97%

Point process models for presence‐only analysis

et al. 2015

Self Cite

View full text Add to dashboard Cite

Summary1. Presence-only data are widely used for species distribution modelling, and point process regression models are a flexible tool that has considerable potential for this problem, when data arise as point events. 2. In this paper, we review point process models, some of their advantages and some common methods of fitting them to presence-only data. 3. Advantages include (and are not limited to) clarification of what the response variable is that is modelled; a framework for choosing the number and location of quadrature points (commonly referred to as pseudoabsences or 'background points') objectively; clarity of model assumptions and tools for checking them; models to handle spatial dependence between points when it is present; and ways forward regarding difficult issues such as accounting for sampling bias. 4. Point process models are related to some common approaches to presence-only species distribution modelling, which means that a variety of different software tools can be used to fit these models, including MAXENT or generalised linear modelling software.

show abstract

Nonparametric estimation of the dependence of a spatial point process on spatial covariates

Cited by 77 publications

References 44 publications

Entrainment, motion, and deposition of coarse particles transported by water over a sloping mobile bed

Entrainment, motion, and deposition of coarse particles transported by water over a sloping mobile bed

Monitoring techniques in the capture and adoption of dogs and cats

Point process models for presence‐only analysis

Contact Info

Product

Resources

About