Long‐term (37 years) impacts of low‐head dams on freshwater shrimp habitat connectivity in northeastern Puerto Rico

• Published in: River research and applications Vol. 35; no. 7; pp. 1034 - 1043
• Main Authors: Chappell, Jessica, McKay, S. Kyle, Freeman, Mary C., Pringle, Catherine M.
• Format: Journal Article
• Language: English
• Published: Bognor Regis Wiley Subscription Services, Inc 01.09.2019
• Subjects: Animal migration; connectivity index; Dams; Decapoda; Drainage; Drought; Freshwater; Freshwater crustaceans; freshwater shrimp; Habitats; Headwaters; Inland water environment; Life cycle; Life cycles; Locating; long term; Longitude; low‐head dams; Marine crustaceans; metapopulation; Metapopulations; partial barriers; Predators; Refugia; Shellfish; Stream discharge; Stream flow; Streams; Water intakes; Waterfalls; Watersheds; United States > US; Puerto Rico
• ISSN: 1535-1459; 1535-1467
• DOI: 10.1002/rra.3499

Freshwater migratory shrimp in Puerto Rico depend on watershed connectivity, from stream headwaters to the ocean, to complete their life cycle. Moreover, shrimp populations in different watersheds are known to be connected in an island‐wide metapopulation. However, low‐head dams paired with water intakes on streams draining the El Yunque National Forest (EYNF) reduce streamflow. Here, we examine the cumulative effects of low‐head dams on shrimp habitat connectivity over 37 years across seven EYNF watersheds. We calculate total and refugia habitat connectivity (where refugia habitat is defined as predator‐free upstream reaches above waterfalls >5 m in height) at a monthly time step using a habitat‐weighted index of longitudinal riverine connectivity, which incorporates location and operation of water intakes and streamflow variability. Findings indicate total and refugia habitat connectivity declined over 37 years (by 27% and 16%, respectively) as additional water intakes have been placed in lower reaches of watersheds. On a monthly time step, the proportion of streamflow withdrawn has the largest effect on habitat connectivity, with the result that connectivity is ~17% lower during drought years than in nondrought years and ~7% lower in dry compared with wet seasons. Our analysis of this long‐term dataset highlights how cumulative effects of low‐head dams paired with water intakes have reduced shrimp habitat connectivity. These results underscore the importance of reducing existing withdrawal rates in EYNF, and locating intakes where effects on connectivity are minimal, if conserving shrimp habitat is a management objective.