Effect of woody riparian patches in croplands on stream macroinvertebrates

Woody riparian vegetation plays important roles in stream ecosystems and its presence can have strong impacts on stream fauna. Agricultural practices have led to the removal and fragmentation of woody riparian vegetation in many watersheds. It is unclear whether small, isolated patches of woody riparian vegetation continue to exert important influences on stream fauna. The impact of these types of patches and the influence of the size of the patches on stream macroinvertebrates was examined in a watershed in northeastern Oregon that is dominated by dryland wheat production. Half of the study reaches flowed through patches of woody riparian vegetation and half flowed through areas in which there was no woody vegetation and wheat fields were found within 3 m of the stream. Stream reaches flowing through patches of woody riparian vegetation had higher taxa richness than those flowing through areas lacking woody vegetation. Size of woody patches, as measured by patch length, was positively correlated with shredder abundance, scraper abundance and diversity, and was also correlated with overall community composition as defined by ordination scores. An analysis of individual taxa revealed that patch length was positively associated with nine taxa that are known to be sensitive to human disturbances and negatively associated with one taxon which is considered tolerant of human disturbances. Patch length was also negatively associated with the proportion of sediment on the substrate and it appears that this environmental factor might drive the relationships between patch length and the abundance of the ten taxa. Woody riparian patch width was also examined as a measure of size. In contrast to length, patch width was not correlated with any metric of macroinvertebrate community structure. The results of this study also revealed that macroinvertebrate community structure was influenced by the drainage the study sites were found in. This was unexpected because the study was conducted in a small watershed and the sites within each drainage were specifically chosen to be similar in land use and geology. These results indicate that patch length may be as important, if not more important, than patch width in influencing stream ecosystems. However, the literature generally considers only buffer width when examining the relationship between patch size and stream systems. This study highlights the need to consider patch length, as well as width, in studies examining the effect of patches of riparian vegetation on stream fauna, in the design of buffer projects, and in the monitoring of existing project effectiveness.