77 results on '"SPDE"'
Search Results
2. Bayesian Geostatistics Modeling of Maritime Surveillance Data
- Author
-
Miguel, Belchior, Simões, Paula, de Deus, Rui Gonçalves, Natário, Isabel, Goos, Gerhard, Series Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Gervasi, Osvaldo, editor, Murgante, Beniamino, editor, Garau, Chiara, editor, Taniar, David, editor, C. Rocha, Ana Maria A., editor, and Faginas Lago, Maria Noelia, editor more...
- Published
- 2024
- Full Text
- View/download PDF
Catalog
3. Bayesian spatio-temporal analysis of malaria prevalence in children between 2 and 10 years of age in Gabon
- Author
-
Mougeni, Fabrice, Lell, Bertrand, Kandala, Ngianga-Bakwin, and Chirwa, Tobias
- Published
- 2024
- Full Text
- View/download PDF
4. The effects of gas extraction under intertidal mudflats on sediment and macrozoobenthic communities.
- Author
-
de la Barra, Paula, Aarts, Geert, and Bijleveld, Allert
- Subjects
- *
TIDAL flats , *GAS well drilling , *GAS extraction , *LAND subsidence , *COMPOSITION of sediments , *WORLD Heritage Sites - Abstract
Land subsidence in intertidal environments may change the flooding regime and sediment composition, two drivers of the macrozoobenthic community. In the Dutch Wadden Sea, a UNESCO world heritage site, gas extraction has resulted in an average subsidence of intertidal mudflats of 2 mm year−1. These mudflats support an abundant macrozoobenthic community that offers important resources for birds and fish. The area is managed through the 'hand on the tap' principle, meaning that human activities should be halted if they affect the natural values. To what extent land subsidence affects sediment and macrozoobenthos remains unknown and is increasingly important given sea level rise.Taking advantage of a large‐scale monitoring program, we evaluated the effect of anthropogenically caused subsidence on sediment composition and macrozoobenthos. Nearly 4600 points were sampled yearly (2008–2020) across the Dutch Wadden Sea, allowing us to compare sediment composition and macrozoobenthos biomass within and outside the subsidence area while controlling for the main drivers of these variables. We also compared population trends within and outside the subsidence area for 31 species with different habitat use.Mud fraction was 3% higher within the subsided area and median grain size decreased at 1 μm year−1 while remaining constant in other mudflats. This had no effect on the total biomass of macrozoobenthos. Within the subsidence area, however, the biomass of species that use deeper areas increased compared to outside, and the opposite was true for species using shallower habitat.Policy implications: Land subsidence is related to changes in median grain size and macrozoobenthic community composition. However, because thresholds have not been defined, it is not clear if this requires management actions. For a successful implementation of the 'hand on the tap' principle in the Wadden Sea, it is necessary to define beforehand the relevant variables that represent the natural values, implement proper monitoring and define thresholds above which effects are not acceptable. We propose median grain size, mud fraction and macrozoobenthic composition as good measures of the natural values of the Wadden Sea, and the methods used here as a way for identifying anthropogenic effects on them. [ABSTRACT FROM AUTHOR] more...
- Published
- 2024
- Full Text
- View/download PDF
5. Spatio-temporal modeling of traffic accidents incidence on urban road networks based on an explicit network triangulation.
- Author
-
Chaudhuri, Somnath, Juan, Pablo, and Mateu, Jorge
- Subjects
- *
TRAFFIC accidents , *STOCHASTIC partial differential equations , *TRIANGULATION , *ROAD safety measures , *TRAFFIC fatalities , *ACCIDENT prevention - Abstract
Traffic deaths and injuries are one of the major global public health concerns. The present study considers accident records in an urban environment to explore and analyze spatial and temporal in the incidence of road traffic accidents. We propose a spatio-temporal model to provide predictions of the number of traffic collisions on any given road segment, to further generate a risk map of the entire road network. A Bayesian methodology using Integrated nested Laplace approximations with stochastic partial differential equations (SPDE) has been applied in the modeling process. As a novelty, we have introduced SPDE network triangulation to estimate the spatial autocorrelation restricted to the linear network. The resulting risk maps provide information to identify safe routes between source and destination points, and can be useful for accident prevention and multi-disciplinary road safety measures. [ABSTRACT FROM AUTHOR] more...
- Published
- 2023
- Full Text
- View/download PDF
6. Black Scabbardfish Species Distribution: Geostatistical Inference Under Preferential Sampling
- Author
-
Simões, Paula, Carvalho, M. Lucília, Figueiredo, Ivone, Monteiro, Andreia, Natário, Isabel, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Gervasi, Osvaldo, editor, Murgante, Beniamino, editor, Rocha, Ana Maria A. C., editor, Garau, Chiara, editor, Scorza, Francesco, editor, Karaca, Yeliz, editor, and Torre, Carmelo M., editor more...
- Published
- 2023
- Full Text
- View/download PDF
7. Interpolating climate variables by using INLA and the SPDE approach.
- Author
-
Fioravanti, Guido, Martino, Sara, Cameletti, Michela, and Toreti, Andrea
- Subjects
- *
STOCHASTIC partial differential equations , *CLIMATOLOGY - Abstract
Gridded observational products of the main climate parameters are essential in climate science. Current interpolation approaches, implemented to derive such products, often lack of a proper uncertainty propagation and representation. In this study, we introduce a Bayesian spatiotemporal approach based on the integrated nested Laplace approximation (INLA) and the stochastic partial differential equation (SPDE). The method is described and discussed by using a real case study based on high‐resolution monthly 2‐m maximum (Tmax) and minimum (Tmin) air temperature over Italy in 1961–2020. The INLA‐SPDE based approach is able to properly take into account uncertainties in the final gridded products and offers interesting promising advantages to deal with nonstationary and non‐Gaussian multisource data. [ABSTRACT FROM AUTHOR] more...
- Published
- 2023
- Full Text
- View/download PDF
8. Modeling spatial dependencies of natural hazards in coastal regions: a nonstationary approach with barriers.
- Author
-
Chaudhuri, Somnath, Juan, Pablo, Saurina, Laura Serra, Varga, Diego, and Saez, Marc
- Subjects
- *
TSUNAMI damage , *HAZARD mitigation , *STOCHASTIC partial differential equations , *TSUNAMI warning systems , *NATURAL disasters , *TSUNAMIS - Abstract
Natural hazards like floods, cyclones, earthquakes, or, tsunamis have deep impacts on the environment and society causing damage to both life and property. These events can cause widespread destruction and can lead to long-term socio-economic disruption often affecting the most vulnerable populations in society. Computational modeling provides an essential tool to estimate the damage by incorporating spatial uncertainties and examining global risk assessments. Classical stationary models in spatial statistics often assume isotropy and stationarity. It causes inappropriate smoothing over features having boundaries, holes, or physical barriers. Despite this, nonstationary models like barrier model have been little explored in the context of natural disasters in complex land structures. The principal objective of the current study is to evaluate the influence of barrier models compared to classical stationary models by analysing the incidence of natural disasters in complex spatial regions like islands and coastal areas. In the current study, we have used tsunami records from the island nation of Maldives. For seven atoll groups considered in our study, we have implemented three distinct categories of stochastic partial differential equation meshes, two for stationary models and one that corresponds to the barrier model concept. The results show that when assessing the spatial variance of tsunami incidence at the atoll scale, the barrier model outperforms the other two models while maintaining the same computational cost as the stationary models. In the broader picture, this research work contributes to the relatively new field of nonstationary barrier models and intends to establish a robust modeling framework to explore spatial phenomena, particularly natural hazards, in complex spatial regions having physical barriers. [ABSTRACT FROM AUTHOR] more...
- Published
- 2023
- Full Text
- View/download PDF
9. Understanding wildfire occurrence and size in Jalisco, Mexico: A spatio-temporal analysis.
- Author
-
Toledo-Jaime, Camila, Díaz-Avalos, Carlos, Chaudhuri, Somnath, Serra, Laura, and Juan, Pablo
- Subjects
STOCHASTIC partial differential equations ,WILDFIRES ,GROUND vegetation cover ,CLIMATE change ,WILDFIRE prevention - Abstract
In recent years, the growing frequency and severity of wildfires, influenced by both human activities and climate change, have posed significant challenges worldwide. Among the regions most affected by wildfires in Mexico is the state of Jalisco, which has the largest accumulated burned area in the last five decades. In this paper, we present an in-depth analysis of the spatio-temporal patterns of wildfire occurrence and size in the state of Jalisco, spanning the period from 2001 to 2020. Our approach included modeling the spatial distribution of the area burned by wildfires, employing Bayesian methodology with Integrated Nested Laplace Approximation (INLA) and Stochastic Partial Differential Equations (SPDE). Our findings highlight the critical roles of vegetation, temperature, and human activities in shaping wildfire behavior. Additionally, our model suggests four distinct wildfire-prone regions within the state. The insights gained from this study can serve as a foundation for future research and localized studies, aiding in the development of more targeted and effective wildfire management strategies in Jalisco. • The use of INLA-SPDE methodology is an innovation to wildfire research in Mexico. • Environmental factors exhibit varying impacts on wildfire occurrence and size. • Wildfires in Jalisco are associated with vegetation cover and human activities. [ABSTRACT FROM AUTHOR] more...
- Published
- 2024
- Full Text
- View/download PDF
10. Spatial Modelling of Black Scabbardfish Fishery Off the Portuguese Coast
- Author
-
André, Lídia Maria, Figueiredo, Ivone, Carvalho, M. Lucília, Simões, Paula, Natário, Isabel, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Woeginger, Gerhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Gervasi, Osvaldo, editor, Murgante, Beniamino, editor, Misra, Sanjay, editor, Garau, Chiara, editor, Blečić, Ivan, editor, Taniar, David, editor, Apduhan, Bernady O., editor, Rocha, Ana Maria A.C., editor, Tarantino, Eufemia, editor, Torre, Carmelo Maria, editor, and Karaca, Yeliz, editor more...
- Published
- 2020
- Full Text
- View/download PDF
11. The Spread of the Japanese Beetle in a European Human-Dominated Landscape: High Anthropization Favors Colonization of Popillia japonica.
- Author
-
Della Rocca, Francesca and Milanesi, Pietro
- Subjects
- *
STOCHASTIC partial differential equations , *COLONIZATION , *BEETLES , *BROADLEAF forests , *HUMAN settlements - Abstract
The impact of invasive species is not limited to the loss of biodiversity; it also represents significant threats to agriculture on a global scale. The Japanese beetle Popillia japonica (native to Japan but an invasive agricultural pest in North America) recently occurred in the Po plain (Italy), one of the most cultivated areas in southern Europe. Thus, our aims were to identify (i) the main landscape predictors related to the occurrence of the Japanese beetle and (ii) the areas of potential invasion of the Japanese beetle in the two Northern Italian regions in which this invasive species currently occurs, Piedmont and Lombardy. Specifically, we combined Japanese beetle occurrences available in the citizen science online platform iNaturalist with high-resolution landscape predictors in an ensemble approach and averaged the results of Bayesian generalized linear and additive models developed with the integrated nested Laplace approximation (with stochastic partial differential equation). We found that the occurrence of the Japanese beetle was negatively related to the percentage of broadleaf forests and pastures, while it was positively related to sparse and dense human settlements as well as intensive crops. Moreover, the occurrence of the Japanese beetle increased in relation to the percentage of rice fields until a peak at around 50%. The Japanese beetle was likely to occur in 32.49% of our study area, corresponding to 16,000.02 km2, mainly located in the Po plain, low hills, and mountain valleys. We stress that the Japanese beetle is a high-risk invasive species in human-dominated landscapes. Thus, we strongly recommend that local administrations quickly enact pest management in order to reduce further spread. [ABSTRACT FROM AUTHOR] more...
- Published
- 2022
- Full Text
- View/download PDF
12. The New Dominator of the World: Modeling the Global Distribution of the Japanese Beetle under Land Use and Climate Change Scenarios.
- Author
-
Della Rocca, Francesca and Milanesi, Pietro
- Subjects
STOCHASTIC partial differential equations ,SPECIES distribution ,INTRODUCED insects ,CLIMATE change ,BEETLES ,SPECIES ,POPULATION density ,HUMAN comfort - Abstract
The spread of invasive species is a threat to global biodiversity. The Japanese beetle is native to Japan, but alien populations of this insect occur in North America, and recently, also in southern Europe. This beetle was recently included on the list of priority species of European concern, as it is a highly invasive agricultural pest. Thus, in this study, we aimed at (i) assessing its current distribution range, and identifying areas of potential invasion, and (ii) predicting its distribution using future climatic and land-use change scenarios for 2050. We collected species occurrences available on the citizen science platform iNaturalist, and we combined species data with climatic and land-use predictors using a Bayesian framework, specifically the integrated nested Laplace approximation, with a stochastic partial differential equation. We found that the current distribution of the Japanese beetle was mainly, and positively, driven by the percentage of croplands, the annual range of temperature, habitat diversity, percentage of human settlements, and human population density; it was negatively related to the distance to airports, elevation, mean temperature diurnal range, wetlands, and waters. As a result, based on current conditions, the Japanese beetle is likely to occur in 47,970,200 km
2 , while its distribution will range from between 53,418,200 and 59,126,825 km2 , according to the 2050 climatic and land-use change scenarios. We concluded that the Japanese beetle is a high-risk invasive species, able to find suitable conditions for its colonization in several regions around the globe, especially in light of ongoing climatic change. Thus, we strongly recommend strict biosecurity checks and quarantines, as well as regular pest management surveys, in order to reduce its spread. [ABSTRACT FROM AUTHOR] more...- Published
- 2022
- Full Text
- View/download PDF
13. Variation in use of Caesarean section in Norway: An application of spatio-temporal Gaussian random fields.
- Author
-
Mannseth, Janne, Berentsen, Geir D., Skaug, Hans J., Lie, Rolv T., and Moster, Dag
- Subjects
- *
CESAREAN section - Abstract
Aims: Caesarean section (CS) is a medical intervention performed in Norway when a surgical delivery is considered more beneficial than a vaginal. Because deliveries with higher risk are centralized to larger hospitals, use of CS varies considerably between hospitals. We describe how the use of CS varies geographically by municipality. Since indications for CS should have little variation across the relatively homogenous population of Norway, we expect fair use of CS to be evenly distributed across the municipalities. Methods: Data from the Medical Birth Registry of Norway were used in our analyses (810,914 total deliveries, 133,746 CSs, 440 municipalities). We propose a spatial correlation model that takes the location into account to describe the variation in use of CS across the municipalities. The R packages R-INLA and TMB are used to estimate the yearly municipal CS rate and the spatial correlation between the municipalities. We also apply stratified models for different categories of delivering women (Robson groups). Estimated rates are displayed in maps and model parameters are shown in tables. Results: The CS rate varies substantially between the different municipalities. As expected, there was strong correlation between neighbouring municipalities. Similar results were found for different Robson groups. Conclusions: The substantial difference in CS use across municipalities in Norway is not likely to be due to specific medical reasons, but rather to hospitals' different policies towards the use of CS. The policy to be either more or less restrictive to CS was not specific to any category of deliveries. [ABSTRACT FROM AUTHOR] more...
- Published
- 2021
- Full Text
- View/download PDF
14. Enhancing the SPDE modeling of spatial point processes with INLA, applied to wildfires. Choosing the best mesh for each database.
- Author
-
Juan Verdoy, Pablo
- Subjects
- *
POINT processes , *STOCHASTIC partial differential equations , *MARKOV random fields , *WILDFIRE prevention , *MARKOV processes , *WILDFIRES - Abstract
Wildfires play an important role in shaping landscapes and as a source of CO2 and particulate matter, and are a typical spatial point process studied in many papers. Modeling the spatial variability of a wildfire could be performed in different ways and an important issue is the computational facilities that the new techniques afford us. The most common approaches have been through point pattern analysis or by Markov random fields. These methods have made it possible to build risk maps, but for many forest managers it is very useful to know the size of the fire as well as its location. In this work, we use Stochastic Partial Differential Equation (SPDE) with Integrated Nested Laplace Approximation (INLA) to model the size of the forest fires observed in the Valencian Community, Spain. But the most important element in this paper is the process that needs to be carried out prior to simulating and analyzing the different point patterns, namely, the choice of the most suitable mesh for the database. We describe and take advantage of the Bayesian methodology by including INLA and SPDE in the modeling process in all the scenarios. [ABSTRACT FROM AUTHOR] more...
- Published
- 2021
- Full Text
- View/download PDF
15. Bayesian prediction of spatial data with non-ignorable missingness.
- Author
-
Zahmatkesh, Samira and Mohammadzadeh, Mohsen
- Subjects
MISSING data (Statistics) ,STOCHASTIC partial differential equations ,GEOLOGICAL statistics ,RANDOM fields ,WATER temperature ,BAYESIAN field theory - Abstract
In spatial data, especially in geostatistics data where measurements are often provided by satellite scanning, some parts of data may get missed. Due to spatial dependence in the data, these missing values probably are caused by some latent spatial random fields. In this case, ignoring missingness is not logical and may lead to invalid inferences. Thus incorporating the missingness process model into the inferences could improve the results. There are several approaches to take into account the non-ignorable missingness, one of them is the shared parameter model method. In this paper, we extend it for spatial data so that we will have a joint spatial Bayesian shared parameter model. Then the missingness process will be jointly modeled with the measurement process and one or more latent spatial random fields as shared parameters would describe their association. Bayesian inference is implemented by Integrated nested Laplace approximation. A computationally effective approach is applied via a stochastic partial differential equation for approximating latent Gaussian random field. In a simulation study, the proposed spatial joint model is compared with a model that assumes data are missing at random. Based on these two models, the lake surface water temperature data for lake Vänern in Sweden are analyzed. The results of estimation and prediction confirm the efficiency of the spatial joint model. [ABSTRACT FROM AUTHOR] more...
- Published
- 2021
- Full Text
- View/download PDF
16. Bayesian space-time gap filling for inference on extreme hot-spots: an application to Red Sea surface temperatures.
- Author
-
Castro-Camilo, Daniela, Mhalla, Linda, and Opitz, Thomas
- Subjects
OCEAN temperature ,STOCHASTIC partial differential equations ,EXTREME value theory ,MARGINAL distributions ,SPACETIME ,LAPLACE distribution - Abstract
We develop a method for probabilistic prediction of extreme value hot-spots in a spatio-temporal framework, tailored to big datasets containing important gaps. In this setting, direct calculation of summaries from data, such as the minimum over a space-time domain, is not possible. To obtain predictive distributions for such cluster summaries, we propose a two-step approach. We first model marginal distributions with a focus on accurate modeling of the right tail and then, after transforming the data to a standard Gaussian scale, we estimate a Gaussian space-time dependence model defined locally in the time domain for the space-time subregions where we want to predict. In the first step, we detrend the mean and standard deviation of the data and fit a spatially resolved generalized Pareto distribution to apply a correction of the upper tail. To ensure spatial smoothness of the estimated trends, we either pool data using nearest-neighbor techniques, or apply generalized additive regression modeling. To cope with high space-time resolution of data, the local Gaussian models use a Markov representation of the Matérn correlation function based on the stochastic partial differential equations (SPDE) approach. In the second step, they are fitted in a Bayesian framework through the integrated nested Laplace approximation implemented in R-INLA. Finally, posterior samples are generated to provide statistical inferences through Monte-Carlo estimation. Motivated by the 2019 Extreme Value Analysis data challenge, we illustrate our approach to predict the distribution of local space-time minima in anomalies of Red Sea surface temperatures, using a gridded dataset (11315 days, 16703 pixels) with artificially generated gaps. In particular, we show the improved performance of our two-step approach over a purely Gaussian model without tail transformations. [ABSTRACT FROM AUTHOR] more...
- Published
- 2021
- Full Text
- View/download PDF
17. Spatio-temporal hierarchical Bayesian analysis of wildfires with Stochastic Partial Differential Equations. A case study from Valencian Community (Spain).
- Author
-
Verdoy, Pablo Juan
- Subjects
- *
STOCHASTIC partial differential equations , *STOCHASTIC analysis , *WILDFIRES , *BAYESIAN analysis , *MARKOV random fields , *FOREST fires - Abstract
The spatio-temporal study of wildfires has two complex elements that are the computational efficiency and longtime processing. Modelling the spatial variability of a wildfire could be performed in different ways, and an important issue is the computational facilities that the new methodological techniques afford us. The Markov random fields methods have made possible to build risk maps, but for many forest managers, it is more advantageous to know the size of the fire and its location. In the first part of this work, Stochastic Partial Differential Equation with Integrated Nested Laplace Approximation is utilised to model the size of the forest fires observed in the Valencian Community (Spain) and so it does the inclusion of the time effect, and the study of the emergency calls. The most crucial element in this paper is the inclusion of the improved meshes for the spatial effect and the time, these are, 2d (locations) and 1d (time) respectively. The advantage of the use of spatio-temporal meshes is described with the inclusion of Bayesian methodology in all the scenarios. [ABSTRACT FROM AUTHOR] more...
- Published
- 2020
- Full Text
- View/download PDF
18. A spatial–temporal study of dengue in Peninsular Malaysia for the year 2017 in two different space–time model.
- Author
-
Abd Naeeim, Nurul Syafiah, Abdul Rahman, Nuzlinda, and Muhammad Fahimi, Fatin Afiqah
- Subjects
- *
STOCHASTIC partial differential equations , *ARBOVIRUS diseases , *DENGUE , *DENGUE hemorrhagic fever , *DISEASE mapping , *AUTOREGRESSIVE models , *DISEASE incidence - Abstract
Spatio-temporal disease mapping models give a great worth in epidemiology, especially in describing the pattern of disease incidence across geographical space and time. This paper analyses the spatial and temporal variability of dengue disease rates based on generalized linear mixed models. For spatio-temporal study, the models incorporate spatially correlated random effects as well as temporal effects. In this study, two different spatial random effects are applied and compared. The first model is based on Leroux spatial model, while the second model is based on the stochastic partial differential equation approach. For the temporal effects, both models follow an autoregressive model of first-order model. The models are fitted within a hierarchical Bayesian framework with integrated nested Laplace approximation methodology. The main objective of this study is to compare both spatio-temporal models in terms of their ability in representing the disease phenomenon. The models are applied to weekly dengue fever data in Peninsular Malaysia reported to the Ministry of Health Malaysia in the year 2017 according to the district level. [ABSTRACT FROM AUTHOR] more...
- Published
- 2020
- Full Text
- View/download PDF
19. Prediction of High Resolution Spatial-Temporal Air Pollutant Map from Big Data Sources
- Author
-
Li, Yingyu, Zhu, Yifang, Yin, Wotao, Liu, Yang, Shi, Guangming, Han, Zhu, Hutchison, David, Series editor, Kanade, Takeo, Series editor, Kittler, Josef, Series editor, Kleinberg, Jon M., Series editor, Mattern, Friedemann, Series editor, Mitchell, John C., Series editor, Naor, Moni, Series editor, Pandu Rangan, C., Series editor, Steffen, Bernhard, Series editor, Terzopoulos, Demetri, Series editor, Tygar, Doug, Series editor, Weikum, Gerhard, Series editor, Wang, Yu, editor, Xiong, Hui, editor, Argamon, Shlomo, editor, Li, XiangYang, editor, and Li, JianZhong, editor more...
- Published
- 2015
- Full Text
- View/download PDF
20. A Spliced Gamma-Generalized Pareto Model for Short-Term Extreme Wind Speed Probabilistic Forecasting.
- Author
-
Castro-Camilo, Daniela, Huser, Raphaël, and Rue, Håvard
- Subjects
- *
WIND speed , *WIND power , *RENEWABLE energy sources , *FOSSIL fuels , *LOAD forecasting (Electric power systems) , *WIND forecasting , *WIND turbines , *LATENT structure analysis - Abstract
Renewable sources of energy such as wind power have become a sustainable alternative to fossil fuel-based energy. However, the uncertainty and fluctuation of the wind speed derived from its intermittent nature bring a great threat to the wind power production stability, and to the wind turbines themselves. Lately, much work has been done on developing models to forecast average wind speed values, yet surprisingly little has focused on proposing models to accurately forecast extreme wind speeds, which can damage the turbines. In this work, we develop a flexible spliced Gamma-Generalized Pareto model to forecast extreme and non-extreme wind speeds simultaneously. Our model belongs to the class of latent Gaussian models, for which inference is conveniently performed based on the integrated nested Laplace approximation method. Considering a flexible additive regression structure, we propose two models for the latent linear predictor to capture the spatio-temporal dynamics of wind speeds. Our models are fast to fit and can describe both the bulk and the tail of the wind speed distribution while producing short-term extreme and non-extreme wind speed probabilistic forecasts. Supplementary materials accompanying this paper appear online. [ABSTRACT FROM AUTHOR] more...
- Published
- 2019
- Full Text
- View/download PDF
21. Dealing with physical barriers in bottlenose dolphin (Tursiops truncatus) distribution.
- Author
-
Martínez-Minaya, Joaquín, Conesa, David, Bakka, Haakon, and Pennino, Maria Grazia
- Subjects
- *
BOTTLENOSE dolphin , *SPECIES distribution , *CETACEA , *DECISION making , *ARCHIPELAGOES - Abstract
• A hierarchical Bayesian SDMs that account for physical barriers was employed. • Dolphin occurrence in Archipelago La Maddalena is influenced by a seasonal effect. • Winter is the season with the highest estimated dolphin occurrence. • Spatial component was also important to explain the Dolphin occurrence. Worldwide, cetacean species have started to be protected, but they are still very vulnerable to accidental damage from an expanding range of human activities at sea. To properly manage these potential threats we need a detailed understanding of the seasonal distributions of these highly mobile populations. To achieve this goal, a growing effort has been underway to develop species distribution models (SDMs) that correctly describe and predict preferred species areas. However, accuracy is not always easy to achieve when physical barriers, such as islands, are present. Indeed, SDMs assume, if only implicitly, that the spatial effect is stationary, and that correlation is only dependent on the distance between observations and not on the direction or a spatial coordinates. The application of stationary SDMs in these cases could lead to incorrect predictions and, consequently, to uninformed decision making. In this study, we identify vulnerable habitats for the bottlenose dolphin in the Archipelago de La Maddalena, Northern Sardinia (Italy) using Bayesian hierarchical SDMs that account for the physical barriers issue and provide a full specification of the associated uncertainty. The approach we propose constitutes a major step forward in the understanding of cetacean species in many ecosystems where physical, geographical and topographical barriers are present. [ABSTRACT FROM AUTHOR] more...
- Published
- 2019
- Full Text
- View/download PDF
22. Incorporating Biotic Information in Species Distribution Models: A Coregionalized Approach
- Author
-
Xavier Barber, David Conesa, Antonio López-Quílez, Joaquín Martínez-Minaya, Iosu Paradinas, and Maria Grazia Pennino
- Subjects
Bayesian hierarchical models ,coregionalized models ,fisheries ,INLA ,predation ,SPDE ,Mathematics ,QA1-939 - Abstract
In this work, we discuss the use of a methodological approach for modelling spatial relationships among species by means of a Bayesian spatial coregionalized model. Inference and prediction is performed using the integrated nested Laplace approximation methodology to reduce the computational burden. We illustrate the performance of the coregionalized model in species interaction scenarios using both simulated and real data. The simulation demonstrates the better predictive performance of the coregionalized model with respect to the univariate models. The case study focus on the spatial distribution of a prey species, the European anchovy (Engraulis encrasicolus), and one of its predator species, the European hake (Merluccius merluccius), in the Mediterranean sea. The results indicate that European hake and anchovy are positively associated, resulting in improved model predictions using the coregionalized model. more...
- Published
- 2021
- Full Text
- View/download PDF
23. Species distribution modeling: a statistical review with focus in spatio-temporal issues.
- Author
-
Martínez-Minaya, Joaquín, Cameletti, Michela, Conesa, David, and Pennino, Maria Grazia
- Subjects
- *
GEOLOGICAL statistics , *SPECIES , *SPATIO-temporal variation , *BIOLOGICAL variation , *BAYESIAN analysis - Abstract
The use of complex statistical models has recently increased substantially in the context of species distribution behavior. This complexity has made the inferential and predictive processes challenging to perform. The Bayesian approach has become a good option to deal with these models due to the ease with which prior information can be incorporated along with the fact that it provides a more realistic and accurate estimation of uncertainty. In this paper, we first review the sources of information and different approaches (frequentist and Bayesian) to model the distribution of a species. We also discuss the Integrated Nested Laplace approximation as a tool with which to obtain marginal posterior distributions of the parameters involved in these models. We finally discuss some important statistical issues that arise when researchers use species data: the presence of a temporal effect (presenting different spatial and spatio-temporal structures), preferential sampling, spatial misalignment, non-stationarity, imperfect detection, and the excess of zeros. [ABSTRACT FROM AUTHOR] more...
- Published
- 2018
- Full Text
- View/download PDF
24. Using additive and coupled spatiotemporal SPDE models: a flexible illustration for predicting occurrence of Culicoides species.
- Author
-
Kifle, Yimer Wasihun, Hens, Niel, and Faes, Christel
- Abstract
This paper formulates and compares a general class of spatiotemporal models for univariate space-time geostatistical data. The implementation of stochastic partial differential equation (SPDE) approach combined with integrated nested Laplace approximation into the R-INLA package makes it computationally feasible to use spatiotemporal models. However, the impact of specifying models with and without space-time interaction is unclear. We formulate an extensive class of additive and coupled spatiotemporal SPDE models and investigate the distinction between them by (1) Extending their temporal effect, allowing a random walk process in time, (2) varying the spatial correlation function and (3) running a simulation study to assess the effect of misspecifying the spatial and temporal models, and to assess the generalizability of our results to a higher number of locations. Our methods are illustrated with Culicoides data from Belgium. The Bayesian spatial predictions showed that the highest prevalence of Culicoides species was found in the Northeastern and central parts of Belgium during summer. [ABSTRACT FROM AUTHOR] more...
- Published
- 2017
- Full Text
- View/download PDF
25. A geostatistical model for combined analysis of point-level and area-level data using INLA and SPDE.
- Author
-
Moraga, Paula, Cramb, Susanna M., Mengersen, Kerrie L., and Pagano, Marcello
- Abstract
In this paper a Bayesian geostatistical model is presented for fusion of data obtained at point and areal resolutions. The model is fitted using the INLA and SPDE approaches. In the SPDE approach, a continuously indexed Gaussian random field is represented as a discretely indexed Gaussian Markov random field (GMRF) by means of a finite basis function defined on a triangulation of the region of study. In order to allow the combination of point and areal data, a new projection matrix for mapping the GMRF from the observation locations to the triangulation nodes is proposed which takes into account the types of data to be combined. The performance of the model is examined and compared with the performance of the method RAMPS via simulation when it is fitted to (i) point, (ii) areal, and (iii) point and areal data to predict several simulated surfaces that can appear in real settings. The model is applied to predict the concentration of fine particulate matter (PM 2.5 ), in Los Angeles and Ventura counties, United States, during 2011. [ABSTRACT FROM AUTHOR] more...
- Published
- 2017
- Full Text
- View/download PDF
26. The New Dominator of the World: Modeling the Global Distribution of the Japanese Beetle under Land Use and Climate Change Scenarios
- Author
-
Francesca Della Rocca, Pietro Milanesi, Della Rocca F., and Milanesi P.
- Subjects
Global and Planetary Change ,climate change ,biodiversity platform ,Ecology ,invasive specie ,citizen science ,INLA ,pest specie ,biodiversity platforms ,invasive species ,pest species ,species distribution models ,SPDE ,Nature and Landscape Conservation - Abstract
The spread of invasive species is a threat to global biodiversity. The Japanese beetle is native to Japan, but alien populations of this insect occur in North America, and recently, also in southern Europe. This beetle was recently included on the list of priority species of European concern, as it is a highly invasive agricultural pest. Thus, in this study, we aimed at (i) assessing its current distribution range, and identifying areas of potential invasion, and (ii) predicting its distribution using future climatic and land-use change scenarios for 2050. We collected species occurrences available on the citizen science platform iNaturalist, and we combined species data with climatic and land-use predictors using a Bayesian framework, specifically the integrated nested Laplace approximation, with a stochastic partial differential equation. We found that the current distribution of the Japanese beetle was mainly, and positively, driven by the percentage of croplands, the annual range of temperature, habitat diversity, percentage of human settlements, and human population density; it was negatively related to the distance to airports, elevation, mean temperature diurnal range, wetlands, and waters. As a result, based on current conditions, the Japanese beetle is likely to occur in 47,970,200 km2, while its distribution will range from between 53,418,200 and 59,126,825 km2, according to the 2050 climatic and land-use change scenarios. We concluded that the Japanese beetle is a high-risk invasive species, able to find suitable conditions for its colonization in several regions around the globe, especially in light of ongoing climatic change. Thus, we strongly recommend strict biosecurity checks and quarantines, as well as regular pest management surveys, in order to reduce its spread. more...
- Published
- 2022
- Full Text
- View/download PDF
27. Spatio-temporal modeling of traffic accidents incidence on urban road networks based on an explicit network triangulation
- Author
-
Pablo Juan, Somnath Chaudhuri, and Jorge Mateu
- Subjects
Statistics and Probability ,network triangulation ,traffic risk mapping ,INLA ,Statistics, Probability and Uncertainty ,Poisson hurdle model ,SPDE - Abstract
Traffic deaths and injuries are one of the major global public health concerns. The present study considers accident records in an urban environment to explore and analyze spatial and temporal in the incidence of road traffic accidents. We propose a spatio-temporal model to provide predictions of the number of traffic collisions on any given road segment, to further generate a risk map of the entire road network. A Bayesian methodology using Integrated nested Laplace approximations with stochastic partial differential equations (SPDE) has been applied in the modeling process. As a novelty, we have introduced SPDE network triangulation to estimate the spatial autocorrelation restricted to the linear network. The resulting risk maps provide information to identify safe routes between source and destination points, and can be useful for accident prevention and multi-disciplinary road safety measures. more...
- Published
- 2022
- Full Text
- View/download PDF
28. The Spread of the Japanese Beetle in a European Human-Dominated Landscape: High Anthropization Favors Colonization of Popillia japonica
- Author
-
Francesca Della Rocca, Pietro Milanesi, Della Rocca F., and Milanesi P.
- Subjects
Ecology ,GAM ,GLM ,iNaturalist ,INLA ,invasive species ,pest ,species distribution models ,SPDE ,Ecological Modeling ,invasive specie ,Agricultural and Biological Sciences (miscellaneous) ,Nature and Landscape Conservation - Abstract
The impact of invasive species is not limited to the loss of biodiversity; it also represents significant threats to agriculture on a global scale. The Japanese beetle Popillia japonica (native to Japan but an invasive agricultural pest in North America) recently occurred in the Po plain (Italy), one of the most cultivated areas in southern Europe. Thus, our aims were to identify (i) the main landscape predictors related to the occurrence of the Japanese beetle and (ii) the areas of potential invasion of the Japanese beetle in the two Northern Italian regions in which this invasive species currently occurs, Piedmont and Lombardy. Specifically, we combined Japanese beetle occurrences available in the citizen science online platform iNaturalist with high-resolution landscape predictors in an ensemble approach and averaged the results of Bayesian generalized linear and additive models developed with the integrated nested Laplace approximation (with stochastic partial differential equation). We found that the occurrence of the Japanese beetle was negatively related to the percentage of broadleaf forests and pastures, while it was positively related to sparse and dense human settlements as well as intensive crops. Moreover, the occurrence of the Japanese beetle increased in relation to the percentage of rice fields until a peak at around 50%. The Japanese beetle was likely to occur in 32.49% of our study area, corresponding to 16,000.02 km2, mainly located in the Po plain, low hills, and mountain valleys. We stress that the Japanese beetle is a high-risk invasive species in human-dominated landscapes. Thus, we strongly recommend that local administrations quickly enact pest management in order to reduce further spread. more...
- Published
- 2022
29. Spatiotemporal modeling of traffic risk mapping: A study of urban road networks in Barcelona, Spain.
- Author
-
Chaudhuri, Somnath, Saez, Marc, Varga, Diego, and Juan, Pablo
- Abstract
Accidents on the road have always been a major concern in modern society. According to the World Health Organization, globally road traffic collisions are one of the leading and fastest growing causes of disability and death. The present research work is conducted on ten years of traffic accident data in an urban environment to explore and analyze spatial and temporal variation in the accidents and related injuries. The proposed spatiotemporal model can make predictions regarding the number of injuries incurred on individual road segments. Bayesian methodology using Integrated Nested Laplace Approximation (INLA) with Stochastic Partial Differential Equations (SPDE) has been applied to generate a predicted risk map for the entire road network. The current study introduces INLA- SPDE modeling to perform spatiotemporal predictive analysis on selected areas, precisely on road networks instead of traditional continuous regions. Additionally, the result risk maps act as a baseline to identify the safe routes in a spatiotemporal context. The methodology can be adapted and applied to enhanced INLA-SPDE modeling of spatial point processes precisely on road networks. [ABSTRACT FROM AUTHOR] more...
- Published
- 2023
- Full Text
- View/download PDF
30. Geographical variation in a fatal outcome of acute myocardial infarction and association with contact to a general practitioner.
- Author
-
Ersbøll, Annette Kjær, Kjærulff, Thora Majlund, Bihrmann, Kristine, Schipperijn, Jasper, Gislason, Gunnar, and Larsen, Mogens Lytken
- Abstract
Background Geographical variation in incidence and mortality of acute myocardial infarction (AMI) is present in Denmark. We aimed at examining the association between contact to a general practitioner (GP) the year before AMI and a fatal outcome of AMI. Methods Register-based data and individual-level addresses including 69,608 individuals with AMI in 2006-2011. A Bayesian hierarchical logistic regression model was used to examine the association. Results A fatal outcome of AMI was seen among 12.0% (78%) of individuals with (without) contact to a GP the year before AMI. A significant association was estimated. Conclusions A fatal outcome of AMI was significantly associated with contact to a GP. A high population to GP ratio and long distance to GP could not explain the increased odds of a fatal outcome of AMI for individuals with no contact to a GP. [ABSTRACT FROM AUTHOR] more...
- Published
- 2016
- Full Text
- View/download PDF
31. Arsenic and chromium topsoil levels and cancer mortality in Spain.
- Author
-
Núñez, Olivier, Fernández-Navarro, Pablo, Martín-Méndez, Iván, Bel-Lan, Alejandro, Locutura, Juan, and López-Abente, Gonzalo
- Subjects
ARSENIC poisoning ,SOIL pollution risk assessment ,TUMOR risk factors ,TOXICOLOGY of chromium ,SPATIO-temporal variation - Abstract
Spatio-temporal cancer mortality studies in Spain have revealed patterns for some tumours which display a distribution that is similar across the sexes and persists over time. Such characteristics would be common to tumours that shared risk factors, including the chemical soil composition. The objective of the present study is to assess the association between levels of chromium and arsenic in soil and the cancer mortality. This is an ecological cancer mortality study at municipal level, covering 861,440 cancer deaths in 7917 Spanish mainland towns from 1999 to 2008. Chromium and arsenic topsoil levels (partial extraction) were determined by ICP-MS at 13,317 sampling points. To estimate the effect of these concentrations on mortality, we fitted Besag, York and Mollié models, which included, as explanatory variables, each town's chromium and arsenic soil levels, estimated by kriging. In addition, we also fitted geostatistical-spatial models including sample locations and town centroids (non-aligned data), using the integrated nested Laplace approximation (INLA) and stochastic partial differential equations (SPDE). All results were adjusted for socio-demographic variables and proximity to industrial emissions. The results showed a statistical association in men and women alike, between arsenic soil levels and mortality due to cancers of the stomach, pancreas, lung and brain and non-Hodgkin's lymphomas (NHL). Among men, an association was observed with cancers of the prostate, buccal cavity and pharynx, oesophagus, colorectal and kidney. Chromium topsoil levels were associated with mortality among women alone, in cancers of the upper gastrointestinal tract, breast and NHL. Our results suggest that chronic exposure arising from low levels of arsenic and chromium in topsoil could be a potential risk factor for developing cancer. [ABSTRACT FROM AUTHOR] more...
- Published
- 2016
- Full Text
- View/download PDF
32. Bayesian space-time gap filling for inference on extreme hot-spots: an application to Red Sea surface temperatures
- Author
-
Linda Mhalla, Thomas Opitz, Daniela Castro-Camilo, University of Glascow, HEC Montréal (HEC Montréal), Biostatistique et Processus Spatiaux (BioSP), and Institut National de Recherche pour l’Agriculture, l’Alimentation et l’Environnement (INRAE) more...
- Subjects
FOS: Computer and information sciences ,Statistics and Probability ,Extreme-value theory ,Gaussian ,Economics, Econometrics and Finance (miscellaneous) ,Sea surface temperature ,01 natural sciences ,Standard deviation ,Methodology (stat.ME) ,010104 statistics & probability ,symbols.namesake ,0502 economics and business ,Statistical inference ,INLA ,050207 economics ,0101 mathematics ,[MATH]Mathematics [math] ,Extreme value theory ,Engineering (miscellaneous) ,Statistics - Methodology ,Mathematics ,SPDE ,Markov chain ,05 social sciences ,Red Sea ,Maxima and minima ,Generalized additive modeling ,Laplace's method ,symbols ,Algorithm ,Gaussian network model - Abstract
International audience; We develop a method for probabilistic prediction of extreme value hot-spots in a spatio-temporal framework, tailored to big datasets containing important gaps. In this setting, direct calculation of summaries from data, such as the minimum over a space-time domain, is not possible. To obtain predictive distributions for such cluster summaries, we propose a two-step approach. We first model marginal distributions with a focus on accurate modeling of the right tail and then, after transforming the data to a standard Gaussian scale, we estimate a Gaussian space-time dependence model defined locally in the time domain for the space-time subregions where we want to predict. In the first step, we detrend the mean and standard deviation of the data and fit a spatially resolved generalized Pareto distribution to apply a correction of the upper tail. To ensure spatial smoothness of the estimated trends, we either pool data using nearest-neighbor techniques, or apply generalized additive regression modeling. To cope with high space-time resolution of data, the local Gaussian models use a Markov representation of the Matern correlation function based on the stochastic partial differential equations (SPDE) approach. In the second step, they are fitted in a Bayesian framework through the integrated nested Laplace approximation implemented in R-INLA. Finally, posterior samples are generated to provide statistical inferences through Monte-Carlo estimation. Motivated by the 2019 Extreme Value Analysis data challenge, we illustrate our approach to predict the distribution of local space-time minima in anomalies of Red Sea surface temperatures, using a gridded dataset (11315 days, 16703 pixels) with artificially generated gaps. In particular, we show the improved performance of our two-step approach over a purely Gaussian model without tail transformations. more...
- Published
- 2021
- Full Text
- View/download PDF
33. A new avenue for Bayesian inference with INLA.
- Author
-
Van Niekerk, Janet, Krainski, Elias, Rustand, Denis, and Rue, Håvard
- Subjects
- *
BAYESIAN field theory , *BIG data , *PROPORTIONAL hazards models - Abstract
Integrated Nested Laplace Approximations (INLA) has been a successful approximate Bayesian inference framework since its proposal by Rue et al. (2009). The increased computational efficiency and accuracy when compared with sampling-based methods for Bayesian inference like MCMC methods, are some contributors to its success. Ongoing research in the INLA methodology and implementation thereof in the R package R-INLA , ensures continued relevance for practitioners and improved performance and applicability of INLA. The era of big data and some recent research developments, presents an opportunity to reformulate some aspects of the classic INLA formulation, to achieve even faster inference, improved numerical stability and scalability. The improvement is especially noticeable for data-rich models. Various examples of data-rich models, like Cox's proportional hazards model, an item-response theory model, a spatial model including prediction, and a three-dimensional model for fMRI data are used to illustrate the efficiency gains in a tangible manner. [ABSTRACT FROM AUTHOR] more...
- Published
- 2023
- Full Text
- View/download PDF
34. Bayesian and network models with covariate effects for predicting heating energy demand.
- Author
-
Juan, Pablo, Braulio-Gonzalo, Marta, Díaz-Ávalos, Carlos, Bovea, María D., and Serra, Laura
- Abstract
The spatial effect is an element presented in many geostatistical works and it should be incorporated into studies regarding the heating energy demand of residential building stocks. The most common approaches have been made by simple descriptive statistics or using analyses by Markov random fields. In this work, we propose two different methods. First, the Stochastic Partial Differential Equation with the Integrated Nested Laplace Approximation to model the variable heating energy demand in Castellón de la Plana, Spain also considering covariates and the spatial effect. Second, simulated street networks for analysing data. We describe and take advantage of the Bayesian methodology in the modelling process in all the scenarios, including covariates and the possibility of creating a simulated street network with the data for the modelling issue. Our results show that the spatial location of the building is a crucial element to study the heating energy demand using both methodologies. [ABSTRACT FROM AUTHOR] more...
- Published
- 2022
- Full Text
- View/download PDF
35. Incorporating Biotic Information in Species Distribution Models: A Coregionalized Approach
- Author
-
Joaquín Martínez-Minaya, Antonio López-Quílez, Xavier Barber, David Conesa, Maria Grazia Pennino, and Iosu Paradinas
- Subjects
0106 biological sciences ,General Mathematics ,Species distribution ,Bayesian probability ,species ,coregionalized models ,Bayesian hierarchical models ,010603 evolutionary biology ,01 natural sciences ,010104 statistics & probability ,models ,Engraulis ,Hake ,Anchovy ,Statistics ,Computer Science (miscellaneous) ,INLA ,distribution ,European anchovy ,Pesquerías ,Centro Oceanográfico de Murcia ,0101 mathematics ,Engineering (miscellaneous) ,SPDE ,fish ,species interaction ,biology ,mathematics ,lcsh:Mathematics ,Univariate ,Merluccius merluccius ,biology.organism_classification ,lcsh:QA1-939 ,fisheries ,Environmental science ,predation - Abstract
In this work, we discuss the use of a methodological approach for modelling spatial relationships among species by means of a Bayesian spatial coregionalized model. Inference and prediction is performed using the integrated nested Laplace approximation methodology to reduce the computational burden. We illustrate the performance of the coregionalized model in species interaction scenarios using both simulated and real data. The simulation demonstrates the better predictive performance of the coregionalized model with respect to the univariate models. The case study focus on the spatial distribution of a prey species, the European anchovy (Engraulis encrasicolus), and one of its predator species, the European hake (Merluccius merluccius), in the Mediterranean sea. The results indicate that European hake and anchovy are positively associated, resulting in improved model predictions using the coregionalized model., SI more...
- Published
- 2021
36. A spatio-temporal Poisson hurdle point process to model wildfires.
- Author
-
Serra, Laura, Saez, Marc, Juan, Pablo, Varga, Diego, and Mateu, Jorge
- Subjects
- *
POINT processes , *FOREST fires , *RELATIVE medical risk , *STATISTICAL software , *ECONOMETRICS , *MATHEMATICAL models - Abstract
Wildfires have been studied in many ways, for instance as a spatial point pattern or through modeling the size of fires or the relative risk of big fires. Lately a large variety of complex statistical models can be fitted routinely to complex data sets, in particular wildfires, as a result of widely accessible high-level statistical software, such as R. The objective in this paper is to model the occurrence of big wildfires (greater than a given extension of hectares) using an adapted two-part econometric model, specifically a hurdle model. The methodology used in this paper is useful to determine those factors that help any fire to become a big wildfire. Our proposal and methodology can be routinely used to contribute to the management of big wildfires. [ABSTRACT FROM AUTHOR] more...
- Published
- 2014
- Full Text
- View/download PDF
37. Discussing the 'big n problem'.
- Author
-
Jona Lasinio, Giovanna, Mastrantonio, Gianluca, and Pollice, Alessio
- Subjects
STOCHASTIC partial differential equations ,LAPLACE distribution ,APPROXIMATION theory ,SIMULATION methods & models ,INTERPOLATION ,COVARIANCE matrices ,DATA analysis ,SET theory - Abstract
When a large amount of spatial data is available computational and modeling challenges arise and they are often labeled as 'big n problem'. In this work we present a brief review of the literature. Then we focus on two approaches, respectively based on stochastic partial differential equations and integrated nested Laplace approximation, and on the tapering of the spatial covariance matrix. The fitting and predictive abilities of using the two methods in conjunction with Kriging interpolation are compared in a simulation study. [ABSTRACT FROM AUTHOR] more...
- Published
- 2013
- Full Text
- View/download PDF
38. Non-stationary spatial model for the distribution of Xylella fastidiosa in Alicante
- Author
-
Cendoya, Martina, Hubel, Ana, Vicent, Antonio, Conesa, David, and Irigoien, Itziar
- Subjects
Alicante ,Xylella fastidiosa ,INLA ,U10 Mathematical and statistical methods ,H20 Plant diseases ,Bayesian hierarchical models ,Barriers ,SPDE - Abstract
Describing the effect of climatic and spatial factors on the geographic distribution of the plant pathogenic bacterium Xylella fastidiosa has been the main aim since the moment that it was discovered its presence in Alicante (Spain). This work started with the analysis of the presence/absence data of the pathogen using Bayesian hierarchical models through the integrated nested Laplace approximation methodology and the stochastic partial differential equation approach. Spatial models usually assume stationarity, however, this may be not applicable when physical barriers are present in the study area. Taking into account the irregularities of the terrain and what this may entail in the spread of the disease, higher altitude areas have been considered as possible barriers in the area of interest. The results show that the spatial effect had a strong effect in the model and also that there was no great influence of the barriers due to their reduced extension. Future work will be focused in using these barriers models with theoretical phytosanitary barriers. more...
- Published
- 2020
39. Spatio-temporal hierarchical Bayesian analysis of wildfires with Stochastic Partial Differential Equations. A case study from Valencian Community (Spain)
- Author
-
Pablo Juan Verdoy
- Subjects
Statistics and Probability ,021103 operations research ,Computer science ,Application Notes ,Bayesian probability ,Bayesian inference ,0211 other engineering and technologies ,02 engineering and technology ,computer.software_genre ,01 natural sciences ,Valencian community ,spatio-temporal mesh ,wildfire ,Stochastic partial differential equation ,010104 statistics & probability ,INLA ,Spatial variability ,Data mining ,0101 mathematics ,Statistics, Probability and Uncertainty ,computer ,SPDE - Abstract
The spatio-temporal study of wildfires has two complex elements that are the computational efficiency and longtime processing. Modelling the spatial variability of a wildfire could be performed in different ways, and an important issue is the computational facilities that the new methodological techniques afford us. The Markov random fields methods have made possible to build risk maps, but for many forest managers, it is more advantageous to know the size of the fire and its location. In the first part of this work, Stochastic Partial Differential Equation with Integrated Nested Laplace Approximation is utilised to model the size of the forest fires observed in the Valencian Community (Spain) and so it does the inclusion of the time effect, and the study of the emergency calls. The most crucial element in this paper is the inclusion of the improved meshes for the spatial effect and the time, these are, 2d (locations) and 1d (time) respectively. The advantage of the use of spatio-temporal meshes is described with the inclusion of Bayesian methodology in all the scenarios. more...
- Published
- 2019
40. Spatial dynamics of armed conflict in Kenya during the electioneering periods of 1997, 2002, 2007, 2013 and 2017 general elections
- Author
-
Kimani, Peter, Athiany, Henry, and Mugo, Caroline
- Subjects
Armed conflict ,INLA ,SPDE ,Electioneering period - Abstract
Paper presented at the 5th Strathmore International Mathematics Conference (SIMC 2019), 12 - 16 August 2019, Strathmore University, Nairobi, Kenya Since Kenya became a multiparty state in 1991, most of the elections have been proceeded by election violence mostly in form of armed conflict. In addition, election violence is prone to some Kenyan parts. To understand the dynamics of armed conflict during the electioneering period, there is a need to understand the interaction between the time before and after the election and the location of the armed conflict. Furthermore, there is a need of an empirical approach that describes armed conflict spatial dynamics during the electioneering period putting into consideration spatial effects, time effects and also the interaction between time and space for the armed conflict. This study aims at mapping armed conflict relative risk during the electioneering period of 1997, 2002, 2007,2013 and 2017 general elections held. The electioneering period is defined as 180 days before and 180 days after the election. Five-time knots, with each knot at day 180 and day 90 before the election, the election date, day 90 and day 180 after elections are used. Secondary data from the Armed Conflict Location and Event Data (ACLED) is used. A Stochastic Partial Differential Equation (SPDE) is used to analyze the point level armed conflict during the electioneering period, where the continuous Gaussian field is represented as discrete indexed Gaussian Random Markov Field (GRMF). Integrated Laplace approximation (INLA) is used to estimate the marginal posterior distribution of the model parameters. In all the electioneering periods of 1997,2002,2007,2013 and 2017 there was a similar pattern of armed conflict relative risk. The relative risk was low at day 180 before the election and continuously increased at day 90 with its peak at the election date. At day 90 after the election, the relative risk is lower than at the election date and lowest at day 180 after the election. Nyanza, Central Rift Valley, Nairobi and Mombasa regions are having the highest relative of armed conflict during the electioneering period. Armed conflict during the election period follows the same pattern in all electioneering periods, with the relative risk being highest at the period near the election date and lowest at periods that are far away from the election dates. Also, some parts of Kenya have a high relative risk of armed conflict in all the electioneering periods. The study offers insights at spatial dynamics of armed conflict in Kenya during the electioneering periods which is important for policy formulation aiming at reducing armed conflict in Kenya. Jomo Kenyatta University of Agriculture and Technology, Kenya. more...
- Published
- 2019
41. Bayesian spatial modelling of geostatistical data using INLA and SPDE methods: A case study predicting malaria risk in Mozambique.
- Author
-
Moraga, Paula, Dean, Christopher, Inoue, Joshua, Morawiecki, Piotr, Noureen, Shahzeb Raja, and Wang, Fengpei
- Abstract
Bayesian spatial models are widely used to analyse data that arise in scientific disciplines such as health, ecology, and the environment. Traditionally, Markov chain Monte Carlo (MCMC) methods have been used to fit these type of models. However, these are highly computationally intensive methods that present a wide range of issues in terms of convergence and can become infeasible in big data problems. The integrated nested Laplace approximation (INLA) method is a computational less-intensive alternative to MCMC that allows us to perform approximate Bayesian inference in latent Gaussian models such as generalised linear mixed models and spatial and spatio-temporal models. This approach can be used in combination with the stochastic partial differential equation (SPDE) approach to analyse geostatistical data that have been collected at particular sites to predict the spatial process underlying the data as well as to assess the effect of covariates and model other sources of variability. Here we demonstrate how to fit a Bayesian spatial model using the INLA and SPDE approaches applied to freely available data of malaria prevalence and risk factors in Mozambique. We show how to fit and interpret the model to predict malaria risk and assess the effect of covariates using the R-INLA package, and provide the R code necessary to reproduce the results or to use it in other spatial applications. [ABSTRACT FROM AUTHOR] more...
- Published
- 2021
- Full Text
- View/download PDF
42. The success of the invasive macrophyte Hydrilla verticillata and its interactions with the native Egeria najas in response to environmental factors and plant abundance in a subtropical reservoir.
- Author
-
Florêncio, Fernanda Moreira, Alves, Diego Corrêa, Lansac-Tôha, Fernando Miranda, Silveira, Márcio José, and Thomaz, Sidinei Magela
- Subjects
- *
HYDRILLA , *TROPICAL plants , *MACROPHYTES , *POTAMOGETON , *WATER levels , *INTRODUCED species , *COLONIZATION (Ecology) - Abstract
• Egeria najas and Hydrilla verticillata (invasive) co-occur in the Itaipu Reservoir. • Abundance variation drivers were analyzed with a spatio-temporal Bayesian modelling. • E. najas and H. verticillata relate negatively with water level oscillations. • Correlations between E. najas and H. verticillata depend on relative abundance. • Previous species abundance is an important determinant of current abundance. The invasion success of non-native species depends on a variety of factors, and abiotic characteristics of invasive ranges and their native populations can offer resistance to non-native species establishment. Reservoirs, compared to natural lakes, can facilitate submerged macrophyte invasion because they provide favorable abiotic conditions to macrophyte establishment and development. However, previous colonization of native species can hamper the development of invasive ones through priority effects. In this investigation we used large data-set obtained in the Itaipu Reservoir (Brazil/Paraguay) to assess the relationship between the abundance of the native submerged macrophyte Egeria najas and of the invasive Hydrilla verticillata and water level, littoral slope and Secchi disk depth. We also investigated how the concomitant and previous abundance of both macrophytes correlates. A Bayesian modeling was applied, controlling the effect of spatial-temporal autocorrelation. Our results indicate an important role of water transparency, littoral slope and water level oscillations in a short period of time (30 days) for the temporal dynamics of these species' abundances. Fluctuations of water levels and littoral slope affected H. verticillata to a greater extent than E. najas. In addition, the abundance of each species was related with its own abundance in previous samplings, highlighting the possible importance of habitat suitability, vegetative propagule pressure and resistance structures for population recovery. In low abundances, there was a possible facilitation between both species, while high abundances probably increased competition. The great importance of water level oscillations on these species' abundances indicate that this strategy can be used as management tool. [ABSTRACT FROM AUTHOR] more...
- Published
- 2021
- Full Text
- View/download PDF
43. Enhancing the SPDE modeling of spatial point processes with INLA, applied to wildfires. Choosing the best mesh for each database
- Author
-
Pablo Juan Verdoy
- Subjects
Statistics and Probability ,021103 operations research ,spatial point process ,Bayesian inference ,0211 other engineering and technologies ,02 engineering and technology ,computer.software_genre ,01 natural sciences ,Point process ,010104 statistics & probability ,mesh ,Modeling and Simulation ,Statistics ,INLA ,Data mining ,0101 mathematics ,computer ,Mathematics ,SPDE - Abstract
Wildfires play an important role in shaping landscapes and as a source of CO2 and particulate matter, and are a typical spatial point process studied in many papers. Modeling the spatial variability of a wildfire could be performed in different ways and an important issue is the computational facilities that the new techniques afford us. The most common approaches have been through point pattern analysis or by Markov random fields. These methods have made it possible to build risk maps, but for many forest managers it is very useful to know the size of the fire as well as its location. In this work, we use Stochastic Partial Differential Equation (SPDE) with Integrated Nested Laplace Approximation (INLA) to model the size of the forest fires observed in the Valencian Community, Spain. But the most important element in this paper is the process that needs to be carried out prior to simulating and analyzing the different point patterns, namely, the choice of the most suitable mesh for the database. We describe and take advantage of the Bayesian methodology by including INLA and SPDE in the modeling process in all the scenarios. more...
- Published
- 2019
44. On spatial statistical methods and applications for large datasets
- Author
-
Zaida Jesus Quiroz Cornejo, Marcos Oliveira Prates, Vinicius Diniz Mayrink, Håvard Rue, Sudipto Banerjee, and Flavio Bambirra Goncalves
- Subjects
GMRF ,Análise espacial (Estatística) ,MCMC ,Geostatística ,NNGP ,INLA ,modelamento espaço-temporal ,Geologia Métodos estatísticos ,estatística espacial ,SPDE ,Estatística - Abstract
O foco deste trabalho está na aplicação de modelos inovadores para a análise espaçotemporal da biomassa de anchova em um grande banco de dados e no desenvolvimento de um novo campo aleatório Gaussiano adequado para a análise de grandes conjuntos de dados. O primeiro artigo apresenta uma aplicação avançada da modelagem espaço-temporal através da Equação Diferencial Parcial Estocástica (SPDE) para estimar e prever a biomassa de anchova na costa do Peru. Foi introduzido um modelo espaço-temporal hierárquico Bayesiano completo, levando em consideração as possíveis dependências espaciais ou espaço-temporais dos dados. Estes modelos, computacionalmente eficientes e flexíveis, são também capazes de realizar previsões tanto da presença quanto da abundância de anchovas, em particular, quando o conjunto de locais é grande (> 500) e diferente ao longo do tempo. Eles são baseados em que os campos Gaussianos Matérn podem ser vistos como soluções de uma determinada SPDE que, em combinação com o INLA (Aproximação Integrada Aninhada de Laplace), tem uma melhora na eficiência computacional. O segundo trabalho é dedicado a estender o Processo de vizinho mais próximo Gaussiano (NNGP), recentemente proposto. Uma nova classe de processos de campo aleatório Gaussiano foi construída e, também, mostrada sua aplicabilidade a dados com pequenas ou grandes dependências espaciais. A idéia-chave por trás do novo processo espacial é subdividir o domínio espacial em vários blocos, que são dependentes de alguns dos blocos passados. A redução na complexidade computacional é obtida através da dispersão das matrizes de precisão e e na paralelização de extensos cálculos através de blocos de dados. Estes modelos são úteis para grandes conjuntos de dados espaciais, no qual os métodos tradicionaissão computacionalmente intensivos, tendo um alto custo para serem utilizados. Finalmente, para realizar a inferência, foi adotado o enfoque Bayesiano, no qual utilizou-se algoritmos de Monte Carlo via cadeias de Markov (MCMC). Além de demonstradas as capacid The focus of this work is on the application of novelty models for the spatio-temporal analysis of large anchovy biomass dataset, and the development of a new Gaussian random field suitable for the analysis of large datasets. The first paper presents an advance application of spatio-temporal modeling through the Stochastic Partial Differential Equation (SPDE) for estimating and predicting anchovy biomass off the coast of Peru. We introduce a complete, and computationally efficient, flexible Bayesian hierarchical spatio-temporal modeling for zero-inflated positive continuous, accounting for spatial or spatio-temporal dependencies in the data. The models are capable of performing predictions of anchovy presence and abundance, in particular,in particular, when the set of observed sites is large (> 500) and different across the temporal domain. They are based on the fact that Gaussian Matérn field can be viewed as solutions to a certain SPDE, which combined with Integrated Nested Laplace Approximations (INLA) improves the computational efficiency. The second paper is devoted to extend the newly proposed Nearest Neighbor Gaussian Process (NNGP). A new class of Gaussian random field process is constructed and, it is showed its applicability to simulated data with small or large spatial dependences. The key idea behind this new spatial process (or random field) is to subdivide the spatial domain into several blocks which are dependent on some of the past blocks. The new spatial process recovers the NNGP and independent blocks approach. Moreover, The reduction in computational complexity is achieved through the sparsity of the precision matrices and parallelization of many computations for blocks of data. It is useful for large spatial data sets where traditional methods are too computationally intensive to be used efficiently. Finally, to perform inference we adopt a Bayesian framework, we use Markov chain Monte Carlo (MCMC) algorithms and demonstrate the full inferential capabilities of the modeling including the new more...
- Published
- 2018
45. Unemployment estimation: Spatial point referenced methods and models.
- Author
-
Pereira, Soraia, Turkman, K.F., Correia, Luís, and Rue, Håvard
- Abstract
From the fourth quarter of 2014, the Portuguese Labour Force Survey (PLFS) started geo-referencing the sampling units, namely the dwellings, in which the surveys are carried out. This opens new possibilities in analysing and estimating unemployment and its spatial distribution across any region by employing point referenced methods and models. According to a preestablished sampling criteria, the labour force survey selects a certain number of dwellings from across the nation to study, and establishes the number of unemployed people in each dwelling. Based on this survey, the National Statistical Institute of Portugal presently uses direct estimation methods to estimate the national unemployment figures. Recently however, there has been increased interest in estimating these figures in smaller areas. Due to reduced sampling sizes in small areas, direct estimation methods tend to produce fairly large sampling variations. Therefore, model based methods should be favoured as these tend to borrow strength from area to area by making use of the areal dependence. These model based methods tend to use areal counting processes as models and typically introduce spatial dependence through the model parameters using a latent random effect. In this paper, we suggest using point referenced models as an alternative to the traditional small area estimation methods for unemployment estimation. Specifically, we model the spatial distribution of residential buildings across Portugal using a log Gaussian Cox process, and the number of unemployed people per residential unit as a mark attached to these random points. Thus, the main focus of the study is to model the spatial intensity function of this marked point process. The number of unemployed people in any region can then be estimated using a proper functional of this marked point process. The principal objective of this point referenced method for unemployment estimation is to produce reliable estimates at higher spatial resolutions, and at the same time to incorporate into the model any available auxiliary information of the residential units, such as mean age or education level as compared to areal unit averages used in small area estimation. [ABSTRACT FROM AUTHOR] more...
- Published
- 2021
- Full Text
- View/download PDF
46. Regressão espacial quantílica para previsão da velocidade do vento
- Author
-
Gabriel Henrique Oliveira Assunção, Marcos Oliveira Prates, Thais Paiva Galletti, and Fernanda de Bastiani
- Subjects
Kumaraswamy ,INLA ,Estatistica ,Regressão Espacial Quantílica ,Previsão da Velocidade do Vento ,Log-Logística ,SPDE - Abstract
Analisar a velocidade do vento no estado de Minas Gerais é de extrema importância para planejamentos estratégicos da geração de energia através de fontes renováveis. Regressão Quantílica permite realizar a inferência sobre uma variável resposta em diversos quantis, ao contráriodaRegressãousualemqueoestudoérealizadodiantedamédiadoprocesso. Sendo assim, entender o comportamento da velocidade do vento em diferentes quantis é de suma importância para um melhor entendimento do comportamento da velocidade do vento. Para isso, neste trabalho será proposto um modelo de Regressão Quantílica Espacial para criar mapas dos quantis, provenientes desta regressão, referentes ao potencial eólico no estado de Minas Gerais, e assim, observar pontos especícos em relação à velocidade do vento com a possibilidade de implementação de um parque eólico. more...
- Published
- 2018
47. Species distribution modeling: a statistical review with focus in spatio-temporal issues
- Author
-
David Conesa, Michela Cameletti, Maria Grazia Pennino, and Joaquín Martínez-Minaya
- Subjects
0106 biological sciences ,Hierarchical Bayesian models ,Environmental Engineering ,Computer science ,Bayesian probability ,Species distribution ,Point processes ,Context (language use) ,Computational intelligence ,computer.software_genre ,010603 evolutionary biology ,01 natural sciences ,010104 statistics & probability ,Frequentist inference ,INLA ,Environmental Chemistry ,Geostatistics ,0101 mathematics ,Safety, Risk, Reliability and Quality ,General Environmental Science ,Water Science and Technology ,Preferential sampling ,SPDE ,Statistical model ,Laplace's method ,Data mining ,Focus (optics) ,Settore SECS-S/01 - Statistica ,computer - Abstract
The use of complex statistical models has recently increased substantially in the context of species distribution behavior. This complexity has made the inferential and predictive processes challenging to perform. The Bayesian approach has become a good option to deal with these models due to the ease with which prior information can be incorporated along with the fact that it provides a more realistic and accurate estimation of uncertainty. In this paper, we first review the sources of information and different approaches (frequentist and Bayesian) to model the distribution of a species. We also discuss the Integrated Nested Laplace approximation as a tool with which to obtain marginal posterior distributions of the parameters involved in these models. We finally discuss some important statistical issues that arise when researchers use species data: the presence of a temporal effect (presenting different spatial and spatio-temporal structures), preferential sampling, spatial misalignment, non-stationarity, imperfect detection, and the excess of zeros. more...
- Published
- 2018
48. Modélisation spatiale multi-sources du carbone organique dans le sol
- Author
-
zaouche , Mounia, Bel , Liliane, Tressou , Jessica, Vaudour , Emmanuelle, Mathématiques et Informatique Appliquées (MIA-Paris), Institut National de la Recherche Agronomique (INRA)-AgroParisTech, Ecologie fonctionnelle et écotoxicologie des agroécosystèmes (ECOSYS), Université Paris-Saclay, AgroParisTech-Institut National de la Recherche Agronomique (INRA), Université Paris Saclay (COmUE), Mathématiques et Informatique Appliquées ( MIA-Paris ), Institut National de la Recherche Agronomique ( INRA ) -AgroParisTech, Ecologie fonctionnelle et écotoxicologie des agroécosystèmes ( ECOSYS ), AgroParisTech-Institut National de la Recherche Agronomique ( INRA ), and Université Paris Saclay more...
- Subjects
Statistique spatiale ,sol ,[ SDV ] Life Sciences [q-bio] ,Spatial statistics ,[SDV]Life Sciences [q-bio] ,INLA ,joint modelling ,SOC ,carbone ,modélisation jointe ,SPDE - Abstract
In order to reduce chemical fertilizer in agricultural use and to value the urban organic matter in substitution, a precise knowledge of soil properties is mandatory. Carbon is a good indicator of soil fertility and having at our disposal a precise mappingof its content is useful. In this study we aim at spatially estimate the soil carbon content (SOC) in the Versailles plain and the Alluets plateau, a 220 km2 agricultural area. The novel Bayesian inference approach called Integrated Nested Laplace Approximation with Stochastic Partial Differential Equation (INLA-SPDE) allows us to ensure consistency between the various available sources of information (soil samples and satellite image) and to produce in a short time a posteriori estimations of the parameters and the SOC field, considered as a latent field. Several models were evaluated and compared using the elevation covariate stemming from a Digital Elevation Model (DEM), including or not the data from the satellite image. Adding the image slightly improves the prediction quality in terms of RMSE (Root Mean Square Error RMSE) since the median goes from 3.17 g.kg−1 to 3.15 g.kg −1. Overall the carbon prediction map from the joint model represents more realistically the spatial structure of the carbon field.; La réduction des intrants chimiques d’usage agricole et la valorisatition des mati`eres organiques d’origine urbaine en substitution nécessitent une connaissance fine des propriétés des sols agricoles. Le carbone étant un bon indicateur de la fertilité des sols, il s’avère nécessaire de disposer d’une cartographie précise des teneurs. Dans cette étude, on cherche à produire une cartographie des teneurs en carbone dans la plaine de Versailles et le plateau des Alluets, région agricole de 220 km2. La nouvelle approche d’inférence Bayesienne Integrated Nested Laplace Approximation et Stochastic Partial Differential Equation (INLA-SPDE) nous permet de mettre en cohérence les différentes sources d’information disponibles (prélèvements au sol et image satellite) et d’obtenir en peu de temps des estimations a posteriori des paramètres et du champ de carbone considéré comme un champ latent. Nous avons évalué et comparé via une procédure bootstrap les performances plusieurs modèles utilisant comme covariable l’altitude issue d’un modèle numérique de terrain incluant ou non les données des images satellites.L’intégration de l’image améliore légèrement la qualité de prédiction en terme de RMSE (Root Mean Square Error) dont la médiane passe de 3.17 g.kg−1 `a 3.15 g.kg−1. Mais surtout la carte de prédiction du carbone issue de la modélisation jointe présente de façon beaucoup plus réaliste la structure spatiale du carbone. more...
- Published
- 2018
49. Incorporating Biotic Information in Species Distribution Models: A Coregionalized Approach.
- Author
-
Barber, Xavier, Conesa, David, López-Quílez, Antonio, Martínez-Minaya, Joaquín, Paradinas, Iosu, Pennino, Maria Grazia, and Debón, Ana
- Subjects
SPECIES distribution ,ENGRAULIS encrasicolus ,PREDICTION models - Abstract
In this work, we discuss the use of a methodological approach for modelling spatial relationships among species by means of a Bayesian spatial coregionalized model. Inference and prediction is performed using the integrated nested Laplace approximation methodology to reduce the computational burden. We illustrate the performance of the coregionalized model in species interaction scenarios using both simulated and real data. The simulation demonstrates the better predictive performance of the coregionalized model with respect to the univariate models. The case study focus on the spatial distribution of a prey species, the European anchovy (Engraulis encrasicolus), and one of its predator species, the European hake (Merluccius merluccius), in the Mediterranean sea. The results indicate that European hake and anchovy are positively associated, resulting in improved model predictions using the coregionalized model. [ABSTRACT FROM AUTHOR] more...
- Published
- 2021
- Full Text
- View/download PDF
50. A spatio-temporal Poisson hurdle point process to model wildfires
- Author
-
Laura Serra, Jorge Mateu, Diego Varga, Pablo Juan, and Marc Saez
- Subjects
Environmental Engineering ,Computer science ,Computational intelligence ,Estadística ,Hurdle model ,Wildfire ,Poisson distribution ,Point process ,symbols.namesake ,Econometrics ,INLA ,Environmental Chemistry ,Forest fires -- Prevention and control ,Safety, Risk, Reliability and Quality ,Incendis forestals -- Prevenció i control ,Statistical software ,General Environmental Science ,Water Science and Technology ,SPDE ,business.industry ,Environmental resource management ,Statistics ,Statistical model ,Econometric models ,Variety (cybernetics) ,Econometric model ,Spatio-temporal point processes ,symbols ,Spatial variability ,business ,Models economètrics - Abstract
Wildfires have been studied in many ways, for instance as a spatial point pattern or through modeling the size of fires or the relative risk of big fires. Lately a large variety of complex statistical models can be fitted routinely to complex data sets, in particular wildfires, as a result of widely accessible high-level statistical software, such as R. The objective in this paper is to model the occurrence of big wildfires (greater than a given extension of hectares) using an adapted two-part econometric model, specifically a hurdle model. The methodology used in this paper is useful to determine those factors that help any fire to become a big wildfire. Our proposal and methodology can be routinely used to contribute to the management of big wildfires. more...
- Published
- 2013
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.