Assessing the Geographic Representativeness of Genebank Collections: the Case of Bolivian Wild Potatoes

Abstract: Genebank collection databases can be used for ecogeographical studies under the assumption that the accessions are a geographically unbiased sample. We evaluated the representativeness of a collection of wild potatoes from Bolivia and defined and assessed four types of bias: species, species-area, hotspot, and infrastructure. Species bias is the sampling of some species more often than others. Species-area bias is a sampling that is disproportionate to the total area in which a species is found. Hotspot bias is the disproportionate sampling of areas with high levels of diversity. Infrastructure bias is the disproportionate sampling of areas near roads and towns. Each of these biases is present in the Bolivian wild potato collection. The infrastructure bias was strong: 60% of all wild potato accessions were collected within 2 km of a road, as opposed to 22%, if collections had been made randomly. This analysis can serve as a guide for future collecting trips. It can also provide baseline information for the application of genebank data in studies based on geographic information systems.     

Simplicity versus efficiency in sampling designs and estimation

Simplicity and efficiency in design and estimation are all important in deciding on sampling strategies. A simple model is given and illustrated for four practical situations to show how a good sampling strategy should be selected.The U.S. Government right to retain a non-exclusive, royalty free licence in and to any copyright is acknowledged.     

Combining mapped and statistical data in forest ecological inventory and monitoring – supplementing an existing system

A forest ecological inventory and monitoring system combining information derived from maps and samples is proposed based on ecosystem regions (Bailey, 1994). The system extends the design of the USDA Forest Service Region 6 Inventory and Monitoring System (R6IMS) in the Pacific Northwest of the United States. The key uses of the information are briefly discussed and expected results are illustrated with examples. The system is flexible, allowing regions based on ecological considerations to be modified. Sampling intensities that are affordable are likely to be insufficient to provide meaningful estimates for key parameters relating to rare and endangered species, watersheds, and other ecological units. Methods are proposed for collecting additional information in follow-up surveys and combining it with relevant information obtained in R6IMS. Near-continuous information on weather and possible pollution variables recorded by instruments at sampling sites is needed to develop meaningful models to understand what is happening in the ecoregions. R6IMS and the proposed additions constitute a dynamic system which will be modified further as data are analyzed.     

Prototyping a Vision for Inter-Agency Terrestrial Inventory and Monitoring: a Statistical Perspective

A demonstration project in Oregon examined the feasibility of combining Federal environmental monitoring surveys. An integrated approach should remove duplication of effort and reduce the possibility of providing apparently conflicting information to policy makers and the public. Data collection teams made photo interpretation measurements and on-site soil/vegetation/animal observations at locations that were selected from the Forest Inventory and Analysis (FIA), National Forest System (NFS) Region 6, and National Resource Inventory (NRI) surveys in a six-county area in Northern Oregon. The project demonstrated the feasibility of conducting a combined FIA/NFS/NRI survey and suggests an approach that will preserve the utility of the critical historical information from these surveys. We suggest a framework for estimating the extent of forest and range land that explains FIA/NRI differences and provides a common basis for both surveys. We suggest indicator and protocol criteria that will allow consistent national and regional estimates over all vegetation types, and stress the importance of including measurement repeatability in the design of the combined survey.     

For What Applications Can Probability and Non-Probability Sampling Be Used?

Almost any type of sample has some utilitywhen estimating population quantities. The focus inthis paper is to indicate what type or combination oftypes of sampling can be used in various situationsranging from a sample designed to establishcause-effect or legal challenge to one involving asimple subjective judgment. Several of these methodshave little or no utility in the scientific area buteven in the best of circumstances, particularlycomplex ones, both probabilistic and non-probabilisticprocedures have to be used because of lack ofknowledge and cost. We illustrate this with a marbledmurrelet example.     

Estimating the Number of Tree Species in a Forest Community using Survey Data

Using some large samples as actual populations we study a series of promising estimators for number of species in a population. We recommend a nonparametric estimator, CM2f = cs + fc1/(2c2) where cs is the number of species observed, c1 and c2 are the numbers of species occurring once and twice respectively on the sample plots, and f is the finite population correction for sampling, from such data as collected by Forest Inventory and Analysis (FIA) of the US Forest Service. For FIA samples, there are estimated to be 0, 4, 3, 1 and 6 rare tree species not sampled in Minnesota, Missouri, Pennsylvania, New, York, and Ohio respectively.     

LOCATION OF MUNICIPAL WELL FIELDS IN AN ENVIRONMENTALLY SENSITIVE REGION1

ABSTRACT: A visual selection technique can be utilized to locate potential well field sites prior to on-site testing. The technique is based upon an understanding of the regional hydrology and an appreciation for development goals such as the identification of high-yield and low-impact locations. Although the uniqueness of Pasco, Pinellas, and Hillsborough Counties in Western Florida has been incorporated into the current application, the methodology, with modifications, is transferable to other geographic regions. Overlays of each criterion used in site selection are combined to form a regional composite showing favorable site locations and further serve a useful communications role as visual aids during public presentation. Implications of direct and indirect cost savings to the public are obtained when the methodology is applied to the location of municipal well fields.     

Relative Costs and Benefits of a Continuous and a Periodic Forest Inventory in Minnesota

We show that the Annual Forest Inventory System (AFIS) being developed by the USDA Forest Service and tested in the state of Minnesota is no more costly than the periodic Minnesota forest inventory (12 year measurement cycle). Since very useful results could be generated with AFIS every 4 years in addition to the results from the periodic survey, a continuous inventory would be cost effective. Periodic surveys yield more precise estimates within the first few years of the full survey but afterwards the annual survey would be more precise. Where the changeover in precision occurs still needs to be determined. This cost-neutral advantage of AFIS is quite important to note since the USFS has decided to go to annualized inventories such as AFIS this year.