DEVELOPING A COMPREHENSIVE ILT MONITORING PLAN

ERDC TN-DOER-E17

September 2004

Distance sampling.

Removal modeling.

One potential approach to consider in the development of an ILT monitoring plan is a double-

observer method to estimate and control for variations in detectability among individuals in the

population. The basic idea of this method is to have a primary observer who detects birds and

conveys this information to a secondary observer. The secondary observer records all birds

detected by the primary observer, plus any additional detections. Both observers sample many

different areas, reversing the primary and secondary roles. Although birds will be counted at sites

throughout the range, p can be estimated at a subset of sites stratified according to basic habitat

types used by breeding ILT's (e.g., small versus large sandbars, reservoir shorelines, fly-ash

deposits). This method will account for variation in p attributable to differences among habitat

types. Detectability can be estimated at approximately 10-20 percent of the sites sampled, and

the estimated variation in detectability can be used to estimate population size. Data from this

monitoring effort could then be used to estimate parameters for demographic modeling and to

evaluate the impacts of management activities. The type of monitoring plan could potentially

provide a better estimate of population size, yield scientifically sound protocols and data sets,

and permit robust hypothesis testing of issues concerning population distribution and trends.

DEVELOPING A COMPREHENSIVE ILT MONITORING PLAN: A series of facilitated

discussions followed formal presentations and these discussions focused on survey objectives

and identifying the specific parameters that should be components of a standardized ILT moni-

toring effort.

What should we count?

Count adult birds--change in numbers over years might provide insights into productiv-

ity. Problems associated with this include mobility of birds and the potential movement of indi-

viduals between coastal and inland areas.

Count nests and multiply by 2--assumes that all adults are mated. Most researchers in

attendance suggested that approximately 95 percent of adults are breeding.

Count both nests and adults (and use the higher of the two values).

Regardless of what method is ultimately used, counts of adults should include estimates of

detectability so that final breeding population numbers can be reliably estimated from counts and

accompanied by standard errors. There needs to be long-term consistency in data collection

among years at individual sites.

Where should monitoring be conducted?

It will be almost impossible to count every reach of every river where ILT are known or sus-

pected to breed. Efforts may have to focus on monitoring a representative subset of suitable ILT

breeding habitat. Moreover, the sampling methodology may need to be somewhat flexible

depending on size of area or colonies being sampled. For example, in the lower Mississippi

River, colonies are too large to count all adults or nests. Methods on small colonies may be more

easily standardized. A focus on area sampling may provide flexibility in the use of different

sampling methods to count individuals, but timing of the surveys must be coordinated, and dif-

ference in accuracy of counts between different areas must be quantified. The final monitoring