Significant differences this year between the Energy Information Administration’s (EIA) weekly gas storage estimates and the actual monthly data it received from all of the nation’s underground storage operators have prompted the agency to plan sweeping changes next Thursday to its weekly survey. EIA also has decided to revise 16 weeks of prior data from July 4 to Oct. 17.

The agency would not say what the net change in current working gas levels will be. Although it reported an 84 Bcf weekly injection and 3,028 Bcf of working gas in storage as of Oct. 17, those numbers most likely will change next week. And judging from the impact of the new methodology on the data from the first six months of this year, working gas levels probably are higher than current estimates.

“We did the public comment with respect to our revision policy and folks expressed a preference for not having significant changes, in the case of revisions announced, without prior notice,” said EIA’s Bill Trapmann, who manages the weekly survey. “Comments that we’ve gotten indicated that before major changes in methodology they also wanted advance notice. This is the dark side of that advance notice. We’ve put out as much information as we can without implementing it.”

The revised data will be released next Thursday along with the implementation of the new methodology. Among the more significant changes to the methodology are the inclusion of 11 additional companies to the weekly survey sample for a total of 55, which reduces the amount of non-sampled/estimated volumes in the weekly survey.

In addition, EIA will group the companies in the weekly survey sample into those whose data can be used to more reliably predict the working gas levels of non-sampled companies and those whose data do not correlate as well with the data from nonsampled companies. Those changes will help EIA estimate more accurately the working gas volumes that are non-sampled.

“Recent differences between the monthly and weekly storage series have received significant attention in the industry and are a major motivation for consideration of revisions in the weekly methodology,” EIA said in a discussion of the changes on Thursday.

Over the first six months of this year, there was an average monthly difference of 50 Bcf between the storage data collected from all underground storage operators through EIA form 191 and the storage data collected from the sample of storage operators through form 912. The difference was greatest in April and May at 83 Bcf/month and smallest in January at 13 Bcf/month.

The differences were greatest in the producing region, EIA said, where weekly working gas estimates exceeded monthly totals by an average of 28.7 Bcf/month. The difference between the weekly estimates in the East Consuming region and the actual monthly totals was about 22.3 Bcf/month, and the difference in the West was only 1.5 Bcf/month.

EIA found that a number of factors contributed to the differences:

With the new methodology, the differences were almost cut in half. Although it only plans to revise its weekly data from July 4 forward, EIA provided revised estimates for the first six months of the year to illustrate the impact of the changes in methodology. What it shows is that the discrepancies between the two sets of data were cut down to about 26 Bcf/month from 50 Bcf/month.

The new methodology led to a net change of 32.3 Bcf/month to the prior weekly working gas estimates, and there was a net increase in working gas estimates in five of the first six months of the year. While there was a 10 Bcf/month drop in January estimates, there were 40 Bcf, 16 Bcf, 40 Bcf, 51 Bcf and 37 Bcf increases for each of the following months through June, respectively.

Trapmann indicated that the largest changes in current working gas levels will be seen in the Producing Region, where there were the largest discrepancies between the old weekly data and the monthly data during the first six months of the year.

EIA said the factors contributing to improvements include an updated reference month (June 2003), the inclusion of the 11 new sample companies, the reduction in non-sampled volumes that must be estimated and the removal of data from the estimation process that are not predictive or representative of the pattern of working gas volumes held by non-sampled companies.

“Because the monthly-weekly differences are the result of a mixture of one-time events (reclassification and data reporting differences) and ongoing issues (fixed ratios and sample), EIA found it difficult to find a single estimation method that could improve the match to monthly data for each of the past six months,” the agency said.

EIA intends to continue an investigation of its estimation approaches, and will make additional changes when revisions exceed a certain threshold. It also will continue an examination of its survey sampling and may further enlarge the sample size or make other sample modifications.

Trapmann said the agency will release a more detailed document on the new methodology next Wednesday, and then on Thursday the weekly estimates and the revisions will be released.

A summary discussion of the changes to the weekly survey is available at

©Copyright 2003 Intelligence Press Inc. All rights reserved. The preceding news report may not be republished or redistributed, in whole or in part, in any form, without prior written consent of Intelligence Press, Inc.