My WebLink
|
Help
|
About
|
Sign Out
Home
Browse
Search
REP17182
DRMS
>
Back File Migration
>
Report
>
REP17182
Metadata
Thumbnails
Annotations
Entry Properties
Last modified
8/24/2016 11:46:16 PM
Creation date
11/27/2007 2:03:57 AM
Metadata
Fields
Template:
DRMS Permit Index
Permit No
M1993041
IBM Index Class Name
Report
Doc Date
3/1/1994
Doc Name
PREHISTORIC HISTORIC & GEOLOGIC PROPERTIES PRESERVATION PLAN DOW FLAT BOULDER CNTY COLO
Media Type
D
Archive
No
There are no annotations on this page.
Document management portal powered by Laserfiche WebLink 9 © 1998-2015
Laserfiche.
All rights reserved.
/
187
PDF
Print
Pages to print
Enter page numbers and/or page ranges separated by commas. For example, 1,3,5-12.
After downloading, print the document using a PDF reader (e.g. Adobe Reader).
View images
View plain text
<br />J <br /> <br />1 <br />1 <br /> <br /> <br /> <br />LJ <br /> <br />1 <br /> <br /> <br /> <br />r <br /> <br />1 <br /> <br />Discussion <br />The above results suggest that a practical and useful <br />predictive model can be constructed for the Dowe Flats study area. <br />Archaeological sites are distributed across the study area <br />nonrandomly. Their locations are constrained significantly by <br />slope, distance to water, land form and vegetative cover. These <br />variables may be used to predict site locations with a degree of <br />accuracy that significantly exceeds chance estimates. <br />Exploratory use of discriminant analysis indicates that <br />different linear combinations of predictor variables may yield <br />slightly different results; some equations will overestimate and <br />others will underestimate site frequency. The goals of the <br />cultural resource management plan are best served by a model that <br />minimizes the number of false negatives. Reliance on a model that <br />overestimates site frequencies (false positives) will not adversely <br />impact the management of cultural properties while reliance on one <br />that underestimates site frequencies (false negatives) may. <br />Acceptable rates of false negatives versus false positives are <br />not easily determined. Similar decisions are often made in <br />personnel management contexts (e.g., in determining objective <br />hiring criteria that will insure the highest rate of employee <br />success) and are based on a number of financial and logistical <br />considerations. For the present purposes, the combination of water <br />and slope as predictor variables appears optimum for several <br />reasons. Although this prediction equation may generate a high <br />rate of false positives (95$), its accuracy in correctly <br />classifying sites (true positives) outweighs this drawback. It is <br />more cost effective to verify site absence in misclassified <br />locations than to contend with the delays and expenses that can <br />stem from inadvertently impacting a site that was not predicted. <br />While the combination of land form, slope, and vegetation also <br />yielded a reliable estimate of site locations, a model based on two <br />rather than three variables will probably yield a more stable <br />89 <br /> <br />
The URL can be used to link to this page
Your browser does not support the video tag.