PubMed Health⌕ Search

PubMed · 15862335

Indicator bacteria at five swimming beaches-analysis using random forests.

Abstract

"Random forests," an extension of tree regression, provide a relatively new technique for exploring relationships of a response variable like the density of indicator bacteria in water to numerous potential explanatory variables. We used this tool to study relationships of indicator density at five beaches to numerous other variables and found that day of the week, indicator density 24h earlier, water depth at the sampling point, cloud cover, and others were related to density at one or more of the beaches. Using data from the first 52 days of measurement allowed predicting indicator densities in the following 10 days to order of magnitude at some of the beaches. Our analyses served to demonstrate the potential usefulness of this analytic tool for large data sets with many variables.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

David F Parkhurst, Kristen P Brenner, Alfred P Dufour, Larry J Wymer. 2005-02-08. Indicator bacteria at five swimming beaches-analysis using random forests.. https://doi.org/10.1016/j.watres.2005.01.001

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

Some statistical issues related to multiple linear regression modeling of beach bacteria concentrations.

As a fast and effective technique, the multiple linear regression (MLR) method has been widely used in modeling and prediction of beach bacteria concentrations. Among previous works on this subject, however, several issues were insufficiently or inconsistently addressed. Those issues include the value and use of interaction terms, the serial correlation, the criteria for model selection, and model assessment. The present work shows that serial correlations, as often present in sequentially observed data records, deserve full attention from the modeler. The testing and adjustment for the time-series effect should be implemented in a statistically rigorous framework. The R(2) and Cp-statistic as joint criteria are recommended for the model selection process, while using the t-statistics associated with the full model is erroneous. During model selection, using interaction terms can often help to decrease the bias in reduced models, although the resulting improvement in the numerical performance may be limited. For the assessment of the model predictive capacity, which is different from testing the goodness of fit, a comprehensive set of statistics are advocated to allow for an objective evaluation of different models. Results obtained from the data at Huntington Beach, OH, show that erroneous conclusions could be drawn if only the model R(2) and the count of type I and type II errors are considered. In this sense, several previous works deserve further investigation.

Bathing Beaches↗

Surveillance for waterborne disease and outbreaks associated with recreational water--United States, 2003-2004.

PROBLEM/CONDITION: Since 1971, CDC, the U.S. Environmental Protection Agency, and the Council of State and Territorial Epidemiologists have collaboratively maintained the Waterborne Disease and Outbreak Surveillance System for collecting and reporting waterborne disease and outbreak (WBDO)-related data. In 1978, WBDOs associated with recreational water (natural and treated water) were added. This system is the primary source of data regarding the scope and effects of WBDOs in the United States. REPORTING PERIOD: Data presented summarize WBDOs associated with recreational water that occurred during January 2003-December 2004 and one previously unreported outbreak from 2002. DESCRIPTION OF THE SYSTEM: Public health departments in the states, territories, localities, and the Freely Associated States (i.e., the Republic of the Marshall Islands, the Federated States of Micronesia, and the Republic of Palau, formerly parts of the U.S.-administered Trust Territory of the Pacific Islands) have primary responsibility for detecting, investigating, and voluntarily reporting WBDOs to CDC. Although the surveillance system includes data for WBDOs associated with drinking water, recreational water, and water not intended for drinking, only cases and outbreaks associated with recreational water are summarized in this report. RESULTS: During 2003-2004, a total 62 WBDOs associated with recreational water were reported by 26 states and Guam. Illness occurred in 2,698 persons, resulting in 58 hospitalizations and one death. The median outbreak size was 14 persons (range: 1-617 persons). Of the 62 WBDOs, 30 (48.4%) were outbreaks of gastroenteritis that resulted from infectious agents, chemicals, or toxins; 13 (21.0%) were outbreaks of dermatitis; and seven (11.3%) were outbreaks of acute respiratory illness (ARI). The remaining 12 WBDOs resulted in primary amebic meningoencephalitis (n = one), meningitis (n = one), leptospirosis (n = one), otitis externa (n = one), and mixed illnesses (n = eight). WBDOs associated with gastroenteritis resulted in 1,945 (72.1%) of 2,698 illnesses. Forty-three (69.4%) WBDOs occurred at treated water venues, resulting in 2,446 (90.7%) cases of illness. The etiologic agent was confirmed in 44 (71.0%) of the 62 WBDOs, suspected in 15 (24.2%), and unidentified in three (4.8%). Twenty (32.3%) WBDOs had a bacterial etiology; 15 (24.2%), parasitic; six (9.7%), viral; and three (4.8%), chemical or toxin. Among the 30 gastroenteritis outbreaks, Cryptosporidium was confirmed as the causal agent in 11 (36.7%), and all except one of these outbreaks occurred in treated water venues where Cryptosporidium caused 55.6% (10/18) of the gastroenteritis outbreaks. In this report, 142 Vibrio illnesses (reported to the Cholera and Other Vibrio Illness Surveillance System) that were associated with recreational water exposure were analyzed separately. The most commonly reported species were Vibrio vulnificus, V. alginolyticus, and V. parahaemolyticus. V. vulnificus illnesses associated with recreational water exposure had the highest Vibrio illness hospitalization (87.2%) and mortality (12.8%) rates. INTERPRETATION: The number of WBDOs summarized in this report and the trends in recreational water-associated disease and outbreaks are consistent with previous years. Outbreaks, especially the largest ones, are most likely to be associated with summer months, treated water venues, and gastrointestinal illness. Approximately 60% of illnesses reported for 2003-2004 were associated with the seven largest outbreaks (>100 cases). Deficiencies leading to WBDOs included problems with water quality, venue design, usage, and maintenance. PUBLIC HEALTH ACTIONS: CDC uses WBDO surveillance data to 1) identify the etiologic agents, types of aquatic venues, water-treatment systems, and deficiencies associated with outbreaks; 2) evaluate the adequacy of efforts (i.e., regulations and public awareness activities) to provide safe recreational water; and 3) establish public health prevention priorities that might lead to improved regulations and prevention measures at the local, state, and federal levels.

Bathing Beaches↗

Influence of sampling depth on Escherichia coli concentrations in beach monitoring.

While the US Environmental Protection Agency's (EPA) Beaches Environmental Assessment and Coastal Health (BEACH) Act requires coastal and Great Lakes' states to implement plans for monitoring bacterial contamination of recreational beach water, exactly how this monitoring should occur has not been regulated. This study examined differences in concentration of Escherichia coli in water collected from different depths and from different horizontal locations across the beach. E. coli concentrations were significantly different (p<0.05), when water from different depths was compared. Sampling water at depths of 30, 60, and 120 cm resulted in significantly lower E. coli concentrations as depth increased. Had the State of Wisconsin chosen to collect beach water monitoring samples at a shallower or deeper depth, numbers of beach closures and the potential risk to public health would have changed substantially. These data imply that a revised and standardized protocol for monitoring beach water should be adopted by all states of a monitoring region to better compare microbial contamination of beaches and protect public health.

Bathing Beaches↗