Jamison Conley

Does spatial resolution matter? A multi-scale comparison of object-based and pixel-based methods for detecting change associated with gas well drilling operations

An implicit assumption of the geographic object-based image analysis (GEOBIA) literature is that GEOBIA is more accurate than pixel-based methods for high spatial resolution image classification, but that the benefits of using GEOBIA are likely to be lower when moderate resolution data are employed. This study investigates this assumption within the context of a case study of mapping forest clearings associated with drilling for natural gas. The forest clearings varied from 0.2 to 9.2 ha, with an average size of 0.9 ha. National Aerial Imagery Program data from 2004 to 2010, with 1 m pixel size, were resampled through pixel aggregation to generate imagery with 2, 5, 15, and 30 m pixel sizes. The imagery for each date and at each of the five spatial resolutions was classified into Forest and Non-forest classes, using both maximum likelihood and GEOBIA. Change maps were generated through overlay of the classified images. Accuracy evaluation was carried out using a random sampling approach. The 1 m GEOBIA classification was found to be significantly more accurate than the GEOBIA and per-pixel classifications with either 15 or 30 m resolution. However, at any one particular pixel size (e.g. 1 m), the pixel-based classification was not statistically different from the GEOBIA classification. In addition, for the specific class of forest clearings, accuracy varied with the spatial resolution of the imagery. As the pixel size coarsened from 1 to 30 m, accuracy for the per-pixel method increased from 59% to 80%, but decreased from 71% to 58% for the GEOBIA classification. In summary, for studying the impact of forest clearing associated with gas extraction, GEOBIA is more accurate than pixel-based methods, but only at the very finest resolution of 1 m. For coarser spatial resolutions, per-pixel methods are not statistically different from GEOBIA.

Permitted water pollution discharges and population cancer and non-cancer mortality: toxicity weights and upstream discharge effects in US rural-urban areas

BackgroundThe study conducts statistical and spatial analyses to investigate amounts and types of permitted surface water pollution discharges in relation to population mortality rates for cancer and non-cancer causes nationwide and by urban-rural setting. Data from the Environmental Protection Agencys (EPA) Discharge Monitoring Report (DMR) were used to measure the location, type, and quantity of a selected set of 38 discharge chemicals for 10,395 facilities across the contiguous US. Exposures were refined by weighting amounts of chemical discharges by their estimated toxicity to human health, and by estimating the discharges that occur not only in a local county, but area-weighted discharges occurring upstream in the same watershed. Centers for Disease Control and Prevention (CDC) mortality files were used to measure age-adjusted population mortality rates for cancer, kidney disease, and total non-cancer causes. Analysis included multiple linear regressions to adjust for population health risk covariates. Spatial analyses were conducted by applying geographically weighted regression to examine the geographic relationships between releases and mortality.ResultsGreater non-carcinogenic chemical discharge quantities were associated with significantly higher non-cancer mortality rates, regardless of toxicity weighting or upstream discharge weighting. Cancer mortality was higher in association with carcinogenic discharges only after applying toxicity weights. Kidney disease mortality was related to higher non-carcinogenic discharges only when both applying toxicity weights and including upstream discharges. Effects for kidney mortality and total non-cancer mortality were stronger in rural areas than urban areas. Spatial results show correlations between non-carcinogenic discharges and cancer mortality for much of the contiguous United States, suggesting that chemicals not currently recognized as carcinogens may contribute to cancer mortality risk. The geographically weighted regression results suggest spatial variability in effects, and also indicate that some rural communities may be impacted by upstream urban discharges.ConclusionsThere is evidence that permitted surface water chemical discharges are related to population mortality. Toxicity weights and upstream discharges are important for understanding some mortality effects. Chemicals not currently recognized as carcinogens may nevertheless play a role in contributing to cancer mortality risk. Spatial models allow for the examination of geographic variability not captured through the regression models.

Assessing machine-learning algorithms and image-and lidar-derived variables for GEOBIA classification of mining and mine reclamation

This study investigates machine-learning algorithms and measures derived from RapidEye satellite imagery and light detection and ranging (lidar) data for geographic object-based image analysis classification of mining and mine reclamation. Support vector machines, random forests, and boosted classification and regression trees classification algorithms were assessed and compared with the k-nearest neighbour (k-NN) classifier. For geographic object-based image analysis classification of mine landscapes, the use of disparate data (i.e. lidar data) improved overall accuracy, whereas the use of complex, object-oriented variables such as object geometry measures, first-order texture, and second-order texture from the grey-level co-occurrence matrix decreased or did not improve the classification accuracy. Support vector machines generally outperformed k-NN and the ensemble tree classifiers when only using the band means. With the incorporation of lidar-descriptive statistics, all four algorithms provided statistically comparable accuracies. K-NN suffered reduced classification accuracy with high-dimensional feature spaces, suggesting that a more complex machine-learning algorithm may be more appropriate when a large number of predictor variables are used.

Estimation of exposure to toxic releases using spatial interaction modeling

BackgroundThe United States Environmental Protection Agencys Toxic Release Inventory (TRI) data are frequently used to estimate a communitys exposure to pollution. However, this estimation process often uses underdeveloped geographic theory. Spatial interaction modeling provides a more realistic approach to this estimation process. This paper uses four sets of data: lung cancer age-adjusted mortality rates from the years 1990 through 2006 inclusive from the National Cancer Institutes Surveillance Epidemiology and End Results (SEER) database, TRI releases of carcinogens from 1987 to 1996, covariates associated with lung cancer, and the EPAs Risk-Screening Environmental Indicators (RSEI) model.ResultsThe impact of the volume of carcinogenic TRI releases on each countys lung cancer mortality rates was calculated using six spatial interaction functions (containment, buffer, power decay, exponential decay, quadratic decay, and RSEI estimates) and evaluated with four multivariate regression methods (linear, generalized linear, spatial lag, and spatial error). Akaike Information Criterion values and P values of spatial interaction terms were computed. The impacts calculated from the interaction models were also mapped. Buffer and quadratic interaction functions had the lowest AIC values (22298 and 22525 respectively), although the gains from including the spatial interaction terms were diminished with spatial error and spatial lag regression.ConclusionsThe use of different methods for estimating the spatial risk posed by pollution from TRI sites can give different results about the impact of those sites on health outcomes. The most reliable estimates did not always come from the most complex methods.

Evaluating the utility of companion animal tick surveillance practices for monitoring spread and occurrence of human Lyme disease in West Virginia, 2014-2016

Domestic dogs and cats are potentially effective sentinel populations for monitoring occurrence and spread of Lyme disease. Few studies have evaluated the public health utility of sentinel programmes using geo-analytic approaches. Confirmed Lyme disease cases diagnosed by physicians and ticks submitted by veterinarians to the West Virginia State Health Department were obtained for 2014-2016. Ticks were identified to species, and only Ixodes scapularis were incorporated in the analysis. Separate ordinary least squares (OLS) and spatial lag regression models were conducted to estimate the association between average numbers of Ix. scapularis collected on pets and human Lyme disease incidence. Regression residuals were visualised using Local Morans I as a diagnostic tool to identify spatial dependence. Statistically significant associations were identified between average numbers of Ix. scapularis collected from dogs and human Lyme disease in the OLS (β=20.7, P<0.001) and spatial lag (β=12.0, P=0.002) regression. No significant associations were identified for cats in either regression model. Statistically significant (P≤0.05) spatial dependence was identified in all regression models. Local Morans I maps produced for spatial lag regression residuals indicated a decrease in model over- and under-estimation, but identified a higher number of statistically significant outliers than OLS regression. Results support previous conclusions that dogs are effective sentinel populations for monitoring risk of human exposure to Lyme disease. Findings reinforce the utility of spatial analysis of surveillance data, and highlight West Virginias unique position within the eastern United States in regards to Lyme disease occurrence.

Spatial Technology Applications

This chapter presents an overview of the various spatial technologies that are utilized by law enforcement agencies to document evidence for the preservation of crime scenes which can be used for further investigations or as evidence during trial. The first technology presented is remote sensing . Four of the more common types of remote sensing are discussed: aerial photography , satellite imagery , ground-penetrating radar , and thermal imaging . The chapter continues with a discussion of geographic information systems and their application for crime mapping and analysis as well as geographic profiling . A review of the innovative uses of laser scanning technologies to document crime and accident scenes concludes the discussion.

Geospatial Technologies in the Courtroom

The function of a court is to resolve disputes through a legal process. With few exceptions, the progression of a legal case will follow the strict guidelines of rules and codes developed from numerous court decisions to fairly and efficiently securing a just determination. All federal courts adhere to a flexible set of rules published in the Federal Rules of Evidence (FRE ). The FRE provides rules and definitions governing general provisions, judicial notice, presumptions, relevance , privileges, witnesses , expert witnesses , hearsay , and authentication . However, there are as yet no special rules governing the use of geospatial technologies or spatial data . From a pragmatic legal perspective, spatial data differs immensely from the traditional form of evidence. However, the power of spatial information is extremely persuasive and compelling in litigation. While the acceptance of spatial data and methods has increased in litigation, there are also several issues that merit careful consideration when using spatial data. This chapter examines key rules and court decisions that impact the potential admissibility of spatial data and technologies in a modern courtroom.

Spatial Analysis of Fear of Crime and Police Calls for Service: An Example and Implications for Community Policing

Individuals’ fear of crime exhibits a complex spatial relationship with not just actual crime incidents , but a mix of actual crime, perceptions of crime, neighborhood disorder , and collective efficacy . If people have a high fear of crime, they may be more likely to report suspicious or criminal activities to the police . The fear of crime individuals maintain is most often directly linked to the fear of violent crime ; however, a spatially explicit examination of the impact of violent crime calls for service to police officers, neighborhood disorder and collective efficacy on the fear of crime is still needed. In the current study, we examine the relationship among all of these factors using measures of spatial correlation and spatial regression . While the reactive policing strategy of responding to calls for service is more cost-effective than community policing , targeted proactive strategies might be more useful for long-term crime prevention . Our findings illustrate the potential of spatial analysis in informing policing strategies, by highlighting variation in the spatial relationships between fear of crime, violent crime incidents, collective efficacy, and neighborhood disorder. Using the results of this type of analysis can lead to a better use of police resources to avert crime.

Spatial Tracking Applications

In this chapter, various technologies that permit law enforcement agencies to track people and objects through time and space are discussed. Innovative applications of spatial technologies have been adopted by law enforcement agencies to reduce crime , apprehend offenders , and keep officers safe. A brief review of each technology is given followed by examples in which the technologies have been successfully utilized. Specifically, this chapter examines the following: (1) Global Positioning Systems , (2) cellular phone tracking, (3) unmanned aerial vehicles , (4) automated license plate recognition technology, and (5) radio frequency identification technology.

Concepts, Principles, and Definitions

Forensics is the application of science to solve crime . Geographic Information Science , encompassing geospatial information and technology (GIT), has become established within the criminology and forensic fields in the last decade. Law enforcement agencies and forensic investigators embrace geospatial science and technologies for collecting, storing, manipulating, analyzing, and displaying spatial data , resulting in new information, procedures, and models for investigation , policy, and decision making. Applications, acceptability, relevance , and procedural legality of geospatial technologies vary substantially, leading to the assessment of their roles in law enforcement, rules of evidence , protection of privacy , and constitutional liberties . This chapter discusses the context and principles of geospatial technologies and the integration of geospatial tools, principles, and methods into a five-stage model of crime analysis and investigation.