Counting Cranes – Debate Continues

Aransas Wildlife Refuge biologist Tom Stehn conducted Whooping Crane census flights for 29 years at Aransas during which he tried to find every crane. When Tom Stehn retired from U.S. Fish and Wildlife Service in 2011, the agency changed from doing a weekly Whooping Crane census to conducting a survey that takes place for approximately one week each December. 

Unlike a census, a survey incorporates a technique called “distance sampling” where not every crane is counted but estimates of the cranes not seen are based on how far observed cranes were from the aircraft when sighted. Unfortunately, the margin of error for this survey is quite large, equaling plus or minus 39 cranes for the estimated wild flock of 338 during the winter of 2016.

In January 2017, Dr. Bruce Pugesek, Montana State University and Tom Stehn, published an article in the Proceedings of the 13th North American Crane Workshop entitled “THE UTILITY OF CENSUS OR SURVEY FOR MONITORING WHOOPING CRANES IN WINTER”. The article compares the survey and census methods of counting Whooping Cranes.

The abstract is provided below. To download the entire text of the paper, click this link: WHOOPING-CRANE-CENSUS-OR-SURVEY

1 Department of Ecology, Montana State University, Bozeman, MT 59717, USA
THOMAS V. STEHN, 1613 South Saunders Street, Aransas Pass, TX 78336, USA

Abstract: We discuss recent changes in the monitoring program for endangered whooping cranes (Grus americana) on their winter habitat in Texas. A 61-year annual census was replaced in the winter of 2011-2012 with a distance sampling procedure. Justification for the change was, in part, based on criticism of the previous methods of counting cranes and the assessment of crane mortality on the wintering grounds. We argue here that the arguments, methods, and analyses employed to discount the census procedure and mortality estimates were applied incorrectly or with flawed logic and assertions. We provide analysis and logical arguments to show that the census and mortality counts were scientifically valid estimates. The distance sampling protocol currently employed does not provide the accuracy needed to show small annual changes in population size, nor does it provide any estimate of winter mortality. Implications of the relative merit of census and mortality counts versus distance sampling surveys are discussed in the context of management of the whooping crane.

Share Button


  1. Catherine Wohlfeil March 13, 2017 9:22 pm

    “…without the critically important mortality estimates obtained on census flights, the connection between reduced inflows and increased whooping crane mortality would never have been proven in federal court (The Aransas Project vs. B. Shaw et al. 2011).”

    If, in fact, we allow commerce, under the pseudonym of expediency, to rule with the potential result of permitting an endangered species to disappear from the face of this planet, then the cost to us, and future generations, will not be calculable in dollars, but only in tears.

  2. Bob Stewart March 13, 2017 12:36 pm

    This paper is a damning report on the US Fish and Wildlife Dept.’s(USF&WD) current population sampling methods of the wintering whooping cranes in Texas. It makes me wonder what prompts the USF&WD to continue their inaccurate sampling techniques and criticism of the old census methods under Stein’s leadership. Is it lack of money (budgets) or the simple lack of appropriately trained people to continue the old methods or some other politically inspired rational?

    I was a wildlife biologist with the responsibility of tracking moose population levels over a vast area of boreal forest in northern Ontario. We used a combination of survey and sampling techniques to arrive at population estimates. We randomly selected 10 by 5 kilometer permanent sample plots in the vast wildlife management unit(WMU) areas and flew areal surveys over these plots in winter to get accurate estimates of moose by sex and age. There were many variables that effected the counts (snow levels, flying weather, habitat vegetation density etc.) but we also used survey techniques (track following, constant circling, low flying excursions, repeating transects within the plots etc.) that insured a high level of confidence in accurate counts for the actual sample plots flown. Where the estimates declined in accuracy was when the samples were expanded to get a total WMU population estimate.

    I agree with Stein that his survey methods confidence levels would not decline until whooping cranes become much more numerous (we all hope for that!) and occupy winter habitat over a much large area. If and when this occurs population level accuracy becomes much less important. Hopefully the USF&WD will consider the error in their ways.

  3. Madeline March 13, 2017 11:08 am

    Hi Heather and thanks for reporting on Tom Stehn’s paper. Is there a problem with the link? I can’t get it to work!

    • Heather Ray March 13, 2017 5:12 pm

      It should open in your browser as a pdf file Madeline… It just worked for me and others have accessed it as well.

      • Madeline March 14, 2017 11:48 am

        Got it Heather! Thanks!