Wild Horse & Burro Population Inventory and Estimation

UNITED STATES DEPARTMENT OF THE INTERIOR
BUREAU OF LAND MANAGEMENT
WASHINGTON, D.C. 20240
http://www.blm.gov

February 1, 2010

­­­­­­­­

In Reply Refer To:

4710 (WO 260) P

EMS TRANSMISSION 02/17/2010

Instruction Memorandum No. 2010-057

Expires: 09/30/2011

To:                   All Field Officials (except Alaska)

From:               Assistant Director, Renewable Resources and Planning

Subject:           Wild Horse & Burro Population Inventory and Estimation

Program Area: Wild Horse and Burro Program

Purpose:  This Instruction Memorandum (IM) establishes program guidance and policy for inventorying and estimating wild horse population numbers.

Policy:  Two population survey methods, simultaneous double-count with sightability bias correction and mark-resight using photographs, will progressively be implemented and utilized by field offices as training for field specialists, technical support, and funding are available to facilitate implementation.  The goal of the Wild Horse and Burro Program (WH&B) is to produce a reliable estimate of each wild horse or burro population every 2 years.

Population Estimation Techniques:  Two techniques, mark-resight using photographs and simultaneous double-count with sightability bias correction, are recommended for field use as the principal methods for estimating wild horse population numbers.  These methods should be used for all future inventory flights contingent upon the availability of technical support from USGS or the National Program Office Population Survey Lead and funding.  Selection and use of a specific technique lies within the BLM District’s purview and should be based on the HMA or Complex’s topography, size, and vegetative cover.  Mark-resight using photographs should be conducted using a helicopter and simultaneous double-count with sightability bias correction can be conducted with either a helicopter or fixed-wing aircraft.  As research continues to further develop, improve, and test aerial survey techniques, modifications to the recommended techniques may be made.

1. Description of Population Estimation Techniques:

1. Simultaneous Double-Count With Sightability Bias Correction

      The simultaneous double-count uses two observers to independently observe and record data on groups of individual horses.  Sighting data are then compared using      statistical modeling to estimate sighting rates for the two observers.  The simultaneous double-count method is a form of mark-resight (i.e., animals seen by one observer are the “marked” groups, and those that are also seen by the other observer are “resighted”).

Sighting probabilities for both observers are computed from the sighting rates and Lincoln-Petersen calculations are used to generate a population estimate.  The sightability bias correction technique uses a model of the sighting probability for groups of individuals to determine which covariates (such as group size, percent tree and shrub cover, percent snow cover, observer experience, survey intensity) influence sightability.

2. Mark-resight Using Photographs

Mark-resight techniques are most effective and require fewer resight survey flights when uniquely marked animals are identified during the survey.  Horses often have sufficient distinctive natural markings to be uniquely identifiable, making artificial marking unnecessary.  Natural markings include unique coat color or face and leg patterns.  The number and makeup of individuals in a band is also used to identify the band uniquely.

This method uses photographs to record band size and natural markings of animals observed during the “mark” survey and subsequent “resight” surveys.  The observed “marked” animals are compared to animals seen during the “resight” survey(s) as a basis to estimate population size.

One important advantage of using natural markings is that the cost of marking (sighting recognizable groups/harems once) is no higher than for resight. Therefore, precision can be improved by “marking” a large portion of the population during the first survey.  Allocating roughly equal effort to the sighting and resighting surveys should produce the most cost-effective estimate.

BLM will continue working with USGS to further develop, refine, and critique these techniques over the coming years, and determine if the new techniques are appropriate for burro surveys.  The researchers will utilize the results of BLM field office aerial surveys to aid in the estimation of our wild horse populations Bureau-wide.

At a minimum, if consultation with BLM’s Population Inventory Lead and USGS indicates that neither method would be appropriate or cost-effective for specific HMAs or complexes, then wild horse and burro specialists should design their aerial surveys in accordance with the following best management practices:

• Lay out transects ahead of time -- plan for complete and systematic coverage of the whole HMA or Complex.
• Complete survey activities at or close to the same time of year for each individual HMA or Complex to reduce seasonal variability in population estimates.
• Look at the population as a whole (inventory HMAs or complexes at the same time using the same methods and observers rather than inventorying HMAs within a complex separately or at different times, using different methods and observers).
• Inter-District and State coordination is mandatory in order to have accurate population estimates on those HMAs or Complexes that cross District and State boundaries.
• Use natural or man-made barriers as opposed to administrative units to define natural inventory units.
• Use two or three observers.
• Use GIS/GPS Technology to plot flight path and animal locations.
• Take a camera – photograph all groups of twenty or more animals when possible.

Implementation Strategies: To effectively implement the proposed methodologies, WO/NPO has identified a multi-step process for implementation.

1. Recruit a population inventory lead position at the National Program Office in Reno, NV.

This position will serve as BLM’s inventory expert to assist the WH&B Program’s field staff.  This person will work closely with USGS researchers in Fort Collins, CO to develop a thorough understanding of the techniques, including analysis and interpretation of the data.  This position will be the team lead for a training cadre that will provide the necessary instruction to the field for planning and completing upcoming inventory flights. This position will be filled in FY2010.

In addition to providing hands on training and refining the inventory techniques, the National Inventory Lead and USGS will develop a field reference guide for planning and utilizing the recommended techniques.

2. Develop a training cadre.

Each State should identify one to three field specialists to serve as members of the training cadre for implementing the improved techniques. The designated personnel may be field WH&B specialists, aviation specialists, or other resource specialists (i.e., wildlife biologists, rangeland management specialists, etc.) as long as they can withstand strenuous hours of flying in helicopters or small planes, without suffering motion sickness, in order to record detailed data.  The training cadre will be organized by late FY2010.

The cadre will be trained by the National Inventory Lead and USGS to provide consistent reference observers and tutors for the field staff.  The cadre will also participate in any inventory flights in their States until field personnel are proficient in using the techniques.

3. Deployment

Bureau-wide deployment of the improved techniques should begin no later than early FY2011 once the training cadre has been organized and training has commenced.  The use of the new inventory techniques in FY2010 will be limited by the availability of technical support to field offices that can be arranged on a State or District level through coordination with the National Inventory Lead or the National Research Coordinator at the NPO.

Data Management: Upon completion of each inventory flight, the field specialist will work with the National Inventory Lead to ensure that the data are organized in the appropriate format for analysis and storage. Until data storage protocols can be developed, primary storage of the inventory data will be at the field office, with copies of the data provided to the National Program Office and USGS.

Data Analysis and Population Estimates:  After the inventory data have been properly organized, either the National Inventory Lead or USGS will perform the required analysis and interpretation.  The National Inventory Lead will work closely with a USGS’ contract statistician until he or she has demonstrated sufficient proficiency in analyzing the data and other resource information necessary to produce scientifically supportable and defensible population estimates of wild horse and burro populations.

Inter-District and State Inventory Coordination:  To further refine population estimates, inter-District and inter-State coordination of inventory operations is required.  Many of BLM’s HMAs share a common boundary with HMAs in a neighboring District or State.  This coordination will reduce the potential to double-count or miss animals due to movement between HMAs.  Inventory flights should run consecutively, to the extent possible, so that horses that routinely travel between neighboring HMAs have the highest probability of being counted and to assure inventory accuracy.

Timeframe:  This Instruction Memorandum is effective upon issuance.

Budget Impact:  Implementation of the new inventory methodologies will include increased costs for labor (inventory personnel as well as analysis) and aviation.  In most cases, the overall costs of completing population inventories will rise when compared to current inventory methods due to increased frequency of inventory (every 2 years), a slight increase in the number of flight hours needed depending on how the previous surveys were conducted, and the need for larger fixed-wing aircraft and helicopters to carry two or more passengers.  It is anticipated that fixed-wing plane costs (Cessna 210) will average $350 per hour while helicopter costs (B206 Ranger) will average $900 per hour.  Labor costs will average $75-$125 per hour for the required two or more observers.  Other associated costs for fuel trucks and support mileage should be comparable to current inventory methods.

Background:  The Bureau of Land Management (BLM) is responsible for management of 180 herd management areas (HMAs) across 34 million acres in ten western States. 

Most BLM offices currently base their WH&B population estimates on direct counts from either a helicopter or a fixed-wing airplane.  However, research reviewed by the National Research Council (1982) indicated that this practice can undercount the actual number of wild horses by 7-60% depending on topography, vegetation, observer experience, weather, type of aircraft, etc.  More recently, Lubow and Ransom (2009) found an undercount bias as large as 32% before making any statistical corrections.

Underestimating the numbers of wild horses and burros by hundreds of animals in large HMAs has led to unplanned gather, removal, adoption and holding costs; delays in achieving and maintaining AML; and public controversy.

The accuracy and precision of the current direct count inventory method have proven to be unreliable, especially in large areas with tree cover and broken topography.  Statistically valid estimation techniques are needed to account for the animals not observed during inventories and to assure accurate and credible population estimates.

More than ever before, Field Managers and WH&B specialists are challenged to base WH&B management decisions on accurate and credible population estimates.  Standardized, tested, defensible, cost-effective, and easy-to-use population estimation techniques for wild horses are needed.

BLM’s strategic research plan for the Wild Horse and Burro Program identified aerial survey techniques for wild horses and burros as a high priority. USGS and BLM assembled an Aerial Survey Team that began work in 2003.  Team members include USGS researchers and field technicians, BLM field specialists, and a Colorado State University (CSU) research statistician.

The Aerial Survey Team selected the following five most promising aerial techniques for research: 1) simultaneous double-count, 2) sightability bias correction, 3) distance sampling, 4) photographic mark-resight, and 5) combinations of various techniques.  Since 2003, the Aerial Survey Team has tested and evaluated three techniques (simultaneous double-count, sightability bias correction, and photographic mark-resight) in HMAs where population sizes were known and unknown.  The three techniques have been tested with favorable results on three HMAs (Little Book Cliffs, Pryor Mountains, and McCullough Peaks) that had known population sizes.  In areas where the population size was unknown, inventory flights were conducted just before and just after a gather/removal to provide estimates for comparison with the known removal.

To date, the Survey Team has used one or more of these techniques or a combination of techniques to estimate populations in Nevada, Wyoming, Arizona, Colorado, and Montana HMAs.  Based on this field work, no single technique has emerged as being applicable in all cases, due to the highly variable conditions (e.g., terrain, vegetation, snow cover) across HMAs.  Two techniques, mark-resight using photographs and simultaneous double-count with sightability bias correction, produced the best results in the variety of conditions that existed.

Manual/Handbook Sections Affected:  The monitoring requirements do not change or affect any manual or handbook. This policy is consistent with the Strategic Research Plan – Wild Horse and Burro Management.

Coordination:  The requirements outlined in this policy have been evaluated by the Wild Horse and Burro Research Advisory Team, coordinated with the National Wild Horse and Burro Advisory Board, and reviewed by Field Specialists.

Contact:  Questions concerning this policy should be directed to Alan Shepherd at the Nevada State Office in Reno, Nevada at (775) 861-6469.

Signed by:                                                                     Authenticated by:

Edwin L. Roberson                                                      Bridget D. Williams

Assistant Director                                                         Division of IRM Governance, WO-560

Renewable Resources and Planning