November 18, 2009 – Population Estimation

Previously, I have discussed the concept of Appropriate Management Level. This is a number that describes the desired population size for different wild horse herds in different areas. When the population size exceeds AML, there are frequently management actions that occur, such as gathers and removals. Before any of this can occur, though, the population size needs to be determined. Here in the Pryors, and in other similar herds, we closely monitor the population and are able to make population size estimates that are very close to the actual population size. However, this is unusual for wild horse herds. Thus, the topic of my post today: How are wild horse population sizes determined?

The widespread censusing of wild horses dates back to the passage of the Wild Free-Roaming Horses and Burros Act of 1971. Corresponding with the 1971 Act is 43 CFR Part 4700 – Protection, Management, and Control of Wild Free-Roaming Horses and Burros. This document has some more specifics on wild horse management. In Subpart 4710 – Management Considerations is a section (4710.2) on “inventory and monitoring.” Within this section is this line: The authorized officer shall maintain a record of the herd areas that existed in 1971, and a current inventory of the numbers of animals and their areas of use. (In this document, authorized officer is defined as “any employee of the Bureau of Land Management to whom has been delegated the authority to perform the duties described herein.”) So, basically the BLM was directed in 1971 to establish and maintain records of the herd areas created in 1971; and they were to continue to maintain current records on the population sizes of herds.

Well, the BLM was required to do their best in making population estimates in the 1970’s; but it is hard to say how accurate these estimates were given the knowledge that existed on estimation techniques at the time. Accuracy became a big question to all as wild horse management became a bigger issue, and this was a subject of the National Research Council’s 1982 report “Wild and Free-Roaming Horses and Burros.” Starting on page 40 of this document is a section entitled “How Many Horses/Burros Are There In The West?” This section describes how studies done had indicated a wide range of possible error in censusing techniques employed at the time; attempts at censusing were believed to have missed 7% to 60% of the actual population. They had no way of determining which censuses were only counting 93% of the herd and which were counting only 40% of the herd. From this, they recommended some techniques that could be performed every 2 to 3 years to obtain the type of population estimates needed by the BLM. Notice that the new emphasis here was on attempting to accurately estimate the population size as the study had found that attempting to count everything had a lot of potential for error.

Today, the science of counting wild horses is still a matter of debate and research. The BLM has stated that their goal is to obtain accurate population counts for all wild horse herds at least once per four years, but many managers seem to be making annual population counts. In an effort to develop more accurate population estimation techniques, the BLM has teamed up with researchers from the United States Geological Survey and Colorado State University. This group has done a lot of research, and they have concluded that there are four techniques that are potentially good candidates for estimation wild horse populations. These can be read about by clicking here. These techniques seek to combine field observations with statistical analysis so that accurate population size estimates can be consistently and efficiently determined.

A recent publication (“Validating Aerial Photographic Mark-Recapture for Naturally Marked Feral Horses” by Bruce Lubow and Jason Ransom, Journal of Wildlife Management, 2009) describes how testing was done to determine how accurate the mark-recapture method is. This study was done in three wild horse herds that people pay a lot of attention to – The Pryors, the McCullough Peaks, and the Little Book Cliffs. Because it was done in these areas, they could estimate the population sizes for these herds and then compare their estimates to the actually known population sizes. The best estimates in the study ranged from less than 8.6% to over 2.6% of the actual size. Recall that the aforementioned National Research Council report described that censusing back then could be off by 7% to 60%. The results of the Lubow and Ransom study are thus a lot more accurate, and so these researchers feel that this particular technique has the great potential to help managers estimate the population sizes of wild horse herds whose members have many unique and visible markings.

Here in the Pryors, we work hard to maintain accurate population estimates; this is one of the reasons I spend so much time on the PMWHR observing the horses. This is something many people before me have invested many, many hours into as well; I just continue to build on their work. We have a Memorandum of Understanding with the BLM, and part of this MOU regards a relationship in which we openly share our data with the BLM. We do this because we want to do as much as we can to assist the BLM in their decision making processes. My population estimates are ranges. The low end estimate reflects my known population size. The high end estimate reflects my known population size plus the number of horses that I have listed as being missing. However, given the frequency of sightings here, the probability that missing horses are actually alive decreases over time to the point where they are counted as deceased one year after going missing. Thus, my estimate range can actually be quite small; and I can even sometimes say with some confidence what I feel the exact population size is. For example, my last population estimate before the 2009 gather here was on August 31. On that day I estimated the population size to be 189 horses ages one year and up. At the same time, I had 7 horses on my missing list. Of these, one was an old bachelor who hadn’t been seen in eleven months. Two others were middle-aged mares who were found missing this spring. The remaining four were horses born last year that went missing late last year or early this year. Thus, it was probable that these 7 missing horses were actually deceased. This is why I have stated that the pregather population in the Pryors was 189 horses. The largest the population could have possibly been was 196 horses. The largest error I could have had was to underestimate the population by about 4%.

Notice that the Lubow and Ransom study discussed above concerned a method of wild horse population estimation that could estimate the population size to within -8.6% to 2.6%. This isn’t too far off from what we can come up with here in the Pryors. Our field observations here also provide us information on births, interchange, relationships, and the like. However, if it is simply a population size estimate that is needed, then it is likely that the mark-recapture method of counting horses has a lot of potential in allow for accurate estimates to occur. I think that it should be understood that past wild horse population size estimates were likely off by a significant margin. As we now know, population size is a very big factor in the wild horse management decision making process. Thus, it is very important for managers to know herd sizes as accurately as possible; and so it is paramount that accurate techniques, such as those described above, be developed and implemented.

Published in: on November 18, 2009 at 4:44 pm  Comments (2)  

The URI to TrackBack this entry is:

RSS feed for comments on this post.

2 CommentsLeave a comment

  1. I read a recent BLM paper that stated wild horse herd would double in size every four years if no management was done. I think poor pre counting had to explain that figure as there is no way a normal herd under normal conditions doubles every four years. I tried it using their math with a herd of ten horse given that all horses were mares and the increase is 205 as the BLM says it is. By the end of 4 years if all the foals were fillies and all conceived by their third birthday there would be 17.4 horses ( if my math is right and it may not be!) In real life there are cougars and accident and about half the horses are males so it wouldn’t even be that many in reality, unless the beginning number was mis counted and suddenly there were twice as many horses as expected. One of the most valuable things I an think of for the wild horses is some one like you who follows the herds and can give an accurate count of horses and some one like Ginger Katherines who can get their story out to the public so the public cares about the individual herds and horses. While I don’t think Cloud is a more valuable hose than any of the others, his fame made the Pryor round up way more public and got a lot more attention than say the book cliff horses plight or the anonymous herds in Nevada who are being decimated by round ups all winter. I wonder if the little band that has a sheep with them is one of the targeted herds, but I can’t find much information about the Nevada herds.
    The work you and the people who came before you do is so important to the survival of the wild horses and I just wish there were advocates for the other herds as knowledgeable and dedicated as you. thank you.

    • It is my opinion that wild horse population growth rates are often overestimated, and I agree with you that many of these occasions likely resulted in part from a poor understanding of the population’s size. This is why it’s important for researchers like Jason Ransom to figure out the ins and outs of population estimation for the herds. It really is important for the public to follow the horses at this level and to educate the public on what is observed.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: