Impacts of Global River Delta Modification on Ecosystem Services

Human modification of natural systems has typically enhanced provisioning ecosystem services (ES), such as agriculture, at the expense of biodiversity and other types of services. Long-term sustainability requires a balance of ES flows to both maintain human wellbeing, while preserving biodiversity. In river deltas this balance is critical – fertile flat land, well stocked fisheries, and water for use and navigation have promoted rapid population growth and development. Yet the same development degrades the key ES protecting these deltas from hazards and pollution. Many deltas face a critical juncture, at risk or already ‘locked-in’ to the need for engineered solutions to global change problems, unsustainable globally and in the long-term.

We created a global dataset of 237 deltas and collected indicators of the extent each was modified from its natural state, and its ES supply. Several types of modification to the most important aspects of delta systems were considered – overall human impact (human footprint), pressure and demand on ES (population density), modification of water systems (flow disruption) and modification of natural productivity (human appropriation of net primary productivity); grouping deltas by modification state. The impacts of this human modification on over 50 robust biodiversity and ES indicators were then analysed.

Firstly, we attempted to create bundles of commonly associated ES in delta areas. Hierarchical clustering highlighted several logical clusters related to crops, fisheries, water, species richness, biodiversity intactness and NPP. Secondly, we examined synergies and trade-offs between different ES. Provisioning services all showed clear correlations with one another, but clear trade-offs with supporting services or biodiversity. There were weaker synergies within and between regulating and supporting ES. Finally, the relationship between each type of modification and the ES was classified using six typologies fitted by applying a decision tree to their LOESS regression curves. Crop indicators typically had an inverted-U relationship, increasing in moderately modified deltas, but decreasing in the most modified, presumably pushed out by other land uses. While many other ES declined with modification, interestingly, species richness and intactness both began to increase again in the most modified deltas. In summary, this global analysis is the first to illustrate how ES vary along a gradient of development in deltas, and highlights the need to balance further modification against these critical services.