A SEVEN-YEAR COMPARISON OF RELATIVE-ABUNDANCE AND DISTANCE-SAMPLING METHODS

Abstract

W e u se d data f r om statew i d e s urve y s of r i p a r ian b i rds in Utah , 1992–1998, to c omp a r e r e l ative - a bun d anc e and d istanc e - sam pl in g method s . By g ene r atin g re l ative - a bun -d anc e in d i c es with p oint -c o u nt method s and density w ith p oint - t r ansec t sampling methods ,we e x amine d w hether the assu m p tions u n d e rlyin g ea c h method w e r e met dur in g fi e ld s ur-ve y s fo r fo ur b i rd s p e c ies (Br o w n - head e d C o wbi rd [ Molothrus ater], Bull o ck’s Or io l e [ Icterus bullockii ], W a rblin g V i r eo [ Vireo gilvus], and Ye ll o w Wa rble r [Dendroica petechia ]). P oint -c o u nt method s fail e d to r easonably meet the f u n d amental assu m p tion of c onstant pr o p o r tional it y, w ith estimate d detec ta b i l it y va ry in g 3- to 5-fo ld d es p ite the u se of wi d e ly a cc e p te d and w e ll-stan d a rdi z e d method s . P o pulation t r end s based on r e l ative a bun d anc e we r e s ubse quently u nsta bl e , o t en v a ry in g in b oth mag nitude and di r e c tion with the s urve y pl ot r a d i u s u se d (25 m , 50 m , o r un l imite d distanc e ). D istanc e - sam pl in g method s a ppea r e d to meet criti c a l assu m p-tions, w e r e r o bust to assu m p tion v io l ations , a ll o we d method o l o g i c a l se l f - assessment, and w e r e d emonstr a bly e ffi cient in a l a rg e - s c a l e , m ul tisp e c ies s urve y se t in g. Our d ata sho w s urve y s of b i rds witho u t estimations of d etec ta b i l it y a r e l i k e ly b iase d be c a u se the assu m p tion of c onstant pro p o r tional it y is v io l ated to a d e gr ee that pr e clud es st r i c t infe r enc e and may confou n d t r end ana ly ses.