ABSTRACT 2017-284:
Over the past decade, three realizations have evolved from our collection and analysis of oil spill data. First is that more response data are being collected than ever before, including field and laboratory measurements in addition to observational data. To process this diverse information, we use sophisticated computer-based systems that allow us to integrate, analyze, and visualize satellite imagery, real-time weather and ship locations, field notes (e.g., shoreline cleanup assessment technique [SCAT] data), chemistry data, and photos. The second is the increased political and social interest in spills. Increased use of social media and the impact of these information pathways on the public’s perception of the spill response can drive real political decisions; making spill communications based on timely and high data quality critical. Lastly is the growing linkages between the collection, management, and uses of environmental data, not only for spill response, but also for NRD assessment (NRDA), determination of civil penalties (e.g., the Clean Water Act [CWA]), and third party legal claims. For example, observational and remote sensing data collected for response actions will ultimately be used to understand questions about contaminant pathways and exposures inherent to NRDA. Similarly, data collected as part of response mitigation and cleanup needs often provides our earliest understanding of the potential and actual natural resource damage issues, which are important for NRDA, third party claims, and CWA penalty mitigation.
Historically, the inherent differences in temporal and spatial scales over which oil spill data are collected and used, coupled with the requirements of data quality, usability, and/or provenance, diminishes the ability to effectively optimize collection and uses of these data. Data optimization recognizes that data can/will have multiple uses and thus requires all data, whether response or NRDA-related to be of high and equivalent quality and be based on compatible, if not identical, data quality objectives (DQOs). In this paper, we review several examples that underscore the need for data optimization in environmental data collection. Specifically, we will explore how a focus on the long view and the need for data optimization can drive the collection of appropriate and multipurpose data, as well as inform the structure of data management systems. Using specific examples, we will demonstrate the value of embracing a data optimization framework in developing a common sample/data collection imperative that facilitates multiple uses.
Area Detector Systems: Setting Higher Standards for Data Quality