Research Article: A stream classification system to explore the physical habitat diversity and anthropogenic impacts in riverscapes of the eastern United States

Date Published: June 20, 2018

Publisher: Public Library of Science

Author(s): Ryan A. McManamay, Matthew J. Troia, Christopher R. DeRolph, Arlene Olivero Sheldon, Analie R. Barnett, Shih-Chieh Kao, Mark G. Anderson, Steven Arthur Loiselle.

http://doi.org/10.1371/journal.pone.0198439

Abstract

Describing the physical habitat diversity of stream types is important for understanding stream ecosystem complexity, but also prioritizing management of stream ecosystems, especially those that are rare. We developed a stream classification system of six physical habitat layers (size, gradient, hydrology, temperature, valley confinement, and substrate) for approximately 1 million stream reaches within the Eastern United States in order to conduct an inventory of different types of streams and examine stream diversity. Additionally, we compare stream diversity to patterns of anthropogenic disturbances to evaluate associations between stream types and human disturbances, but also to prioritize rare stream types that may lack natural representation in the landscape. Based on combinations of different layers, we estimate there are anywhere from 1,521 to 5,577 different physical types of stream reaches within the Eastern US. By accounting for uncertainty in class membership, these estimates could range from 1,434 to 6,856 stream types. However, 95% of total stream distance is represented by only 30% of the total stream habitat types, which suggests that most stream types are rare. Unfortunately, as much as one third of stream physical diversity within the region has been compromised by anthropogenic disturbances. To provide an example of the stream classification’s utility in management of these ecosystems, we isolated 5% of stream length in the entire region that represented 87% of the total physical diversity of streams to prioritize streams for conservation protection, restoration, and biological monitoring. We suggest that our stream classification framework could be important for exploring the diversity of stream ecosystems and is flexible in that it can be combined with other stream classification frameworks developed at higher resolutions (meso- and micro-habitat scales). Additionally, the exploration of physical diversity helps to estimate the rarity and patchiness of riverscapes over large region and assist in conservation and management.

Partial Text

Classification systems have a long history in stream ecology [1]. Stream classifications serve many fundamental purposes, including understanding similarities and differences among different types of streams, making inferences regarding stream ecosystem behavior, and communicating the complexities of ecosystem function [2]. However, they also provide many applied outcomes, such as grouping sites with similar characteristics [3], stratifying analyses for monitoring and/or experimentation [4], prioritizing aquatic conservation actions [5], and generalizing ecological responses to disturbances [6].

Out of 1.5 million km of stream reaches in the Eastern US, approximately 125,000 km of streams (8.3%) were unavailable to classification due to inability to summarize geospatial variables in stream networks (i.e., braided or artificial channels). The remaining 1.38 million km of stream reaches (91.7%) had sufficient geospatial characteristics to assign physical classes deterministically or through predictive modeling.

Understanding the diversity of streams across large spatial extents highlights commonalities and uniqueness in stream ecosystems [78]. Heterogeneity in stream physical diversity provides a spatial template [79] or landscape filter [80] to examine the influence of physiochemical variation and anthropogenic disturbance regimes on ecological strategies. These patterns are of practical significance because they can be used to assess stream rarity and prioritize conservation efforts. Additionally, generalizing streams into classes has relevance to increasing the representation of finer-scale stream processes in large-scale models of Earth’s ecosystems [81].

 

Source:

http://doi.org/10.1371/journal.pone.0198439

 

0 0 vote
Article Rating
Subscribe
Notify of
guest
0 Comments
Inline Feedbacks
View all comments