A GIS-based multi-criteria method for prioritizing river restoration in Portugal

River management is often driven by energy production, irrigation crops, and flood mitigation, largely enabled by widespread dam construction. However, river damming has brought severe environmental consequences for river dynamics, including morphological changes, hydrological regime, and interruption of the sediment cascade. These physical changes have profound impacts on the ecology of numerous fish species, particularly highly migratory ones, which, together with parallel drivers, led to an 85% reduction in world freshwater species populations since 1970. This work aims to provide a methodological framework for identifying priority rivers within the Portuguese strategy for improving river connectivity affected by obsolete dams and weirs under the European Nature Restoration Law.

The official fish dataset of the Portuguese Institute for Nature Conservation and Forests was first used, which included >20,000 occurrences for 61 fish species between 2010 and 2020. Each species was classified according to the following variables: species origin (native, exotic), conservation status (Critically Endangered, Endangered, Vulnerable, Near Threatened, Least Concern, Data Deficient, Not Evaluated), phenology (diadromous, potamodromous, resident), socioeconomic importance (very high, high, low significance), and endemism degree (Iberian or Lusitanian endemism). A score was attributed to each criterion, and the species index was calculated using Equation 1. Finally, to ensure functional connectivity to the ocean, we have included the lowermost segment of the main rivers up to the first insurmountable dam.

Equation 1: R = a · (0,25 · X_i + 0,40 · X_ii + 0,20 · X_iii + 0,15 · X_iv)

Where the species index (R) resulted from the multiplication of the origin coefficient (α: species origin) with the weighted mean of the variables (X_i: conservation status; X_ii: phenology; X_iii: socioeconomic importance; X_iv: endemism degree).

In parallel, hydrographic modelling was carried out using the Strahler model to ensure full representation of the Portuguese river network and a hierarchy adjusted to the sub-basin scale. The sub-basins were selected based on the 4^th, 5^th, 6^th, and 7^th Strahler hierarchies, and the excluded areas in the main river margins were included using the 8^th hierarchy. For each sub-basin, the arithmetic mean of the species index was calculated within the Geographical Information System environment. The prioritization for river restoration was based on the upper quartile means for each hydrological region independently.

All rivers within the prioritized sub-basins were divided into segments according to the 3^rd cycle of River Basin Management Plans and the Water Framework Directive, as provided by the Portuguese Environmental Agency. Each segment was classified according to water quality (chemical status and ecological quality), and segments were excluded if these criteria were simultaneously negative. At this stage, the Strahler 3^rd-order streams connected to selected segments were included to guarantee ecological coherence in fluvial connectivity. The presence of a protected area (Natura 2000) and the density of transversal barriers were also evaluated. Finally, the national barrier dataset was updated using satellite imagery to identify new barriers.

In total, 77 rivers, encompassing ca. 6500 km at the sub-basin scale, were prioritized within the Portuguese strategy to improve river connectivity affected by obsolete dams and weirs.