Search Results

You are looking at 1 - 4 of 4 items for

  • Author or Editor: Wendy S. Parker x
  • All content x
Clear All Modify Search
Wendy S. Parker
Open access
Wendy S. Parker

Abstract

Are there important differences between reanalysis data and familiar observations and measurements? If so, what are they? This essay evaluates four possible answers that relate to: the role of inference, reliance on forecasts, the need to solve an ill-posed inverse problem, and understanding of errors and uncertainties. The last of these is argued to be most significant. The importance of characterizing uncertainties associated with results—whether those results are observations or measurements, analyses or reanalyses, or forecasts—is emphasized.

Full access
Wendy S. Parker and Greg Lusk

Abstract

Increasingly there are calls for climate services to be “co-produced” with users, taking into account not only the basic information needs of users but also their value systems and decision contexts. What does this mean in practice? One way that user values can be incorporated into climate services is in the management of inductive risk. This involves understanding which errors in climate service products would have particularly negative consequences from the users’ perspective (e.g., underestimating rather than overestimating the change in an impact variable) and then prioritizing the avoidance of those errors. This essay shows how inductive risk could be managed in climate services in ways that serve user values and argues that there are both ethical and practical reasons in favor of doing so.

Open access
Sandra E. Yuter and Wendy S. Parker

Abstract

Fifteen rain measurement instruments were deployed on the National Oceanic and Atmospheric Administration Ship Ronald H. Brown during the 1997 Pan American Climate Studies (PACS) Tropical Eastern Pacific Process Study (TEPPS). To examine differences in rainfall catchment related to instrument design, three types of disdrometers, an optical rain gauge, a ship rain gauge, and a siphon gauge were clustered in one location to ensure similar exposure. To address exposure effects, eight siphon rain gauges were deployed on different sides of the ship and on several different levels.

Cross-ship differences in hourly rainfall accumulation were negligible when relative wind speeds were less than 3 m s−1 and became significant at greater than 5 m s−1, especially when the relative wind direction was 20° or greater from the bow. Instruments with both horizontal and vertical catchment surfaces yielded a measurable collection advantage over instruments with only horizontal catchment surfaces.

Analysis of data collected during TEPPS using a multiple-instrument, multiple-location approach yields the following recommendations for reducing uncertainty in rain measurement at sea. The first two of the four recommendations apply to rain measurements on buoys as well as on ships. 1) Deploy experimental rain measurement instrumentation paired with a baseline minimum siphon gauge or other trusted instrument. Comparison of the rain-rate time series between the baseline gauge measurements and the experimental instrument data permits detection of erratic behavior and bias. 2) Apply an appropriate wind correction. To do this step properly, both a wind correction formula derived for the specific gauge type and a nearby measurement of relative wind are needed. These features are already incorporated into the ship rain gauge. 3) Locate gauges where distortion of the airflow by the ship is locally minimized and relative wind speeds are as low as possible. This analysis confirms previous recommendations for placement of rain instrumentation at lower locations as long as the location is protected against direct spray from the sea without being shadowed by higher objects. 4) Place instrumentation on both sides of ship and along centerline. Airflow distortion by the ship itself can induce significant differences between port and starboard accumulations at high wind speeds and high angle of wind attack to the bow. Multiple locations aid in constraining error, because relative wind direction and speed vary during a cruise and there is no one perfect location on ship for rain instrumentation.

Full access