Fish diversity of Colombian Andes-Amazon streams at the end of conflict is a reference for conservation before increased land use

Abstract

Reference conditions are difficult to find in the Anthropocene but essential for effective biodiversity conservation. Aquatic ecosystems in the Andes-Amazon transition zone of Colombia are now at high risk due to expanded human activities after peace agreements in 2016 ended armed conflict because lands formerly controlled by FARC and other armed groups are now prone to agricultural and urban expansion. Particularly, expanding human land use may reduce fish diversity across the altitudinal gradient, especially in the premontane streams (i.e., <500 m a.s.l.) because lands are more amenable to human use than at greater altitudes. We evaluated fish α-diversity (measured as species richness, total abundance, and effective species number) and β-diversity (spatial and temporal) in 12 sites over 8 years bracketing the end of armed conflict. All α-diversity and β-diversity analyses were evaluated relative to categorical altitude (< or >500 m) and continuous altitude. Strong differences in fish community structure among sites occurred as a function of altitude. Fish communities exhibit altitudinal biodiversity gradients that are consistent in space and time, and that need to be accounted for conservation and management considerations. Our results provide a reference to identify short- and long-term changes due to impending human land use at a critical moment for the conservation of tropical fish diversity. Similar studies in other areas of the upper Amazon Basin are needed to evaluate effects of subsequent human activities on diversity patterns and our study area to compare to reference conditions reported here.

Keywords: Amazon piedmont; abundance; altitudinal gradient; beta diversity; effective diversity; land use; multivariate analysis; species richness.

FIGURE 1

Hypothetical effects of potential land use on fish α‐diversity across elevation gradients. The solid curve represents a decay of diversity with altitude. The lower dashed curve represents potential diversity after natural lands are converted to agriculture and other human uses, where greatest diversity loss (arrows) occurs at lower altitude.

FIGURE 2

(a) Geographical location of the sample sites. (b) Examples of the aquatic ecosystems within the altitudinal levels, which are the ones recognized by van der Hammen and dos Santos (1995).

FIGURE 3

Comparisons of premontane and basimontane fish communities. (a) Rarefaction curves approach asymptotes, indicating representative sampling. Premontane sites were inhabited by more species, as supported by evidence in (b) for total abundance, richness, and effective diversity. (c) Taxonomic composition differed between premontane and basimontane sites, where Siluriformes were more prominent in basimontane sites than in premontane sites. Characiformes were dominant in both site types, but in different proportions. (d) NMDS of fish communities from the sampled sites. Dots represent pooled temporal data (2013–2022) for sites. Polygons represent fish communities based on altitudinal levels recognized by van der Hammen and dos Santos (1995).

FIGURE 4

Regression lines for species richness (a), total abundance (b), and effective diversity (c) ~ altitude, where regression lines represent GLMM predicted values and error bars. Note log axes; regressions represent power law functions.

FIGURE 5

Temporal β‐diversity (Jaccard dissimilarity) of premontane (dark green) and basimontane (light green) sites. Temporal β‐diversity compares time t to time t + 1 for each site. Linear regressions are shown with 95% confidence intervals. Data represent Jaccard index for each site between consecutive samples, and adjusted R ² values are low because trends are essentially flat through time, as indicated by nonsignificant slope coefficients for Temporal Interval (premontane p = .96, basimontane p = 17).