Ginal, P. (2017): <b>Acoustic indices do not reflect avian diversity in high-diverse tropical rainforests in Ecuador</b> Philipps-Universität Marburg, <i>master thesis</i>
Resource Description
Title:
Acoustic indices do not reflect avian diversity in high-diverse tropical rainforests in Ecuador
FOR816dw ID:
1556
Publication Date:
2017-02-12
License and Usage Rights:
PAK 823-825 data user agreement. (www.tropicalmountainforest.org/dataagreementp3.do)
Resource Owner(s):
Individual:
Philipp Ginal
Contact:
email:
philipp.ginal <at> gmx.de
Abstract:
Biodiversity monitoring is a global need due to environmental degradation and climate change. Birds are<br/>
often monitored indicators for environmental health because they are easy assessable, fast reacting species,<br/>
which supply important ecosystem services. But conventional biodiversity monitoring can be timeconsuming,<br/>
invasive<br/>
and<br/>
it<br/>
often<br/>
needs<br/>
trained<br/>
specialists,<br/>
thus<br/>
appropriate<br/>
alternatives,<br/>
especially<br/>
for<br/>
largescale<br/>
assessments,<br/>
like<br/>
acoustic<br/>
monitoring<br/>
are<br/>
needed.<br/>
While<br/>
numerous<br/>
acoustic<br/>
indices<br/>
were<br/>
generated<br/>
no<br/>
studies<br/>
occur<br/>
about<br/>
their<br/>
ability<br/>
to<br/>
reflect<br/>
conventional<br/>
alpha-diversity<br/>
measures<br/>
such<br/>
as<br/>
species<br/>
richness,<br/>
abundance,<br/>
diversity or evenness under complex conditions like a gradient of elevation and degradation.<br/>
We choose bird communities along an elevation gradient in natural and degraded rainforests in a highdiverse<br/>
region in the Ecuadorian Andes to investigate the association among five conventional (total<br/>
abundance, species richness, Shannon Index, evenness, Gini Index) and three acoustic (Acoustic Diversity<br/>
Index [ADI], Bioacoustic Index [BI], Acoustic Evenness Index [AEI]) alpha-diversity measures. Furthermore,<br/>
we investigated the influence of higher resolutions (FFT window size, frequency band size) on the acoustic<br/>
indices. Further, we tested a new acoustic beta-diversity measure. With help of multivariate analyzes we<br/>
found acoustic diversity measures were not significantly correlated with conventional diversity measures in<br/>
high-diverse rainforests. The acoustic indices were correlated between another. The values of the acoustic<br/>
indices increased with higher resolutions and were strongly positively correlated with their next higher<br/>
resolution. The NMDS of the conventionally assessed bird communities resulted an elevation gradient and a<br/>
distinct separation between degraded and natural forest communities. The NMDS of the acoustically<br/>
assessed bird communities resulted no distinct gradients. The conventional and acoustic alpha-diversity<br/>
measures were not significantly associated with the elevation or habitat type. ADI, BI and AEI seem not<br/>
appropriate to reflect conventional diversity measures in high-diverse ecosystems, but BI can reflect avian<br/>
abundance in less diverse ecosystems. Avian acoustic activity was driven by noisy species and not by<br/>
species richness or abundance. Higher resolutions of the acoustic indices did not resolute acoustic activity<br/>
more exactly in this case, thus for communities where birds with long or complex calls (relevant for BI) or<br/>
narrow frequency ranges (relevant for ADI/AEI) are not dominating the acoustic activity the default resolutions of the acoustic indices seem sufficient. Acoustic beta-diversity assessment seems problematic<br/>
because the analyzes of frequencies does not reflect species turnover among communities due to the<br/>
reason that several species can occupy the same frequencies with their calls.
Additional Infos:
Author: Philipp Ginal
Keywords:
| beta diversity | species richness | abundance | Alpha diversity | Shannon diversity | soundscape | Acoustic Diversity Index | Bioacustic Index | Acoustic Evenness Index | Gini Index | resolution |
