In the delicately tangled web of an ecosystem, snapping certain anchor lines can bring the whole thing tumbling down faster than severing other threads.

Now, researchers have found that the species we see as most valuable and worth protecting often aren't the 'threads' most critical to maintaining the complex ecological webs we rely on.

Ecologist Aislyn Keyes from the University of California and colleagues used data from three coastal food webs to simulate a dozen extinction sequences to gain a better understanding of how the connections anchor ecosystem services and the stability of the entire webs themselves.

Ecosystem services are the ways in which nature contributes to humanity's wellbeing, such as fresh water resources, pollination, or plants mitigating soil erosion. The contribution of services from insects alone is worth billions of dollars a year just in the US.

These services are interwoven with everything we do. For example, even financial institutions' investments are highly dependent on ecosystem services, and over half of all global GDP is thought to depend on goods and services provided by the natural world.

Yet, many of our activities are severing the threads that support these ecosystem services at random and at an accelerating rate. When the loss of one species cascades into the loss of other species that once depended on it (secondary extinctions), this creates more threats to the ecosystem services we rely on.

Earlier this month, a team of researchers found support for a strong connection between rapidly vanishing species diversity and ecosystem services.

"These connections need to be made more visible in assessments of nature's contributions to human wellbeing in order to fully understand how to manage and protect these benefits to humans," urged co-author and ecologist Silvia Ceaușu from University College London.

Now, in an effort towards better understanding these contributions, modeling in the study from Keyes and team has found "species providing services do not play a critical role in stabilizing food webs - whereas species playing supporting roles in services through interactions are critical to the robustness of both food webs and services."

This means that ecosystem services are under greater threat than anticipated because more species support these services than we've accounted for. And it is a huge problem when it comes to how we prioritize conservation.

"There's not nearly enough money for conservation," explains Keyes. "[The] traditional approaches that assess ecosystem services might be missing a lot of this stuff.

"A lot of ecosystem service assessments focus on only the species that directly provides the service, but we know that impacts can cascade through an ecosystem, and so we show that these secondary extinctions represent an increased vulnerability for services that hasn't necessarily been considered."

The loss of highly connected species has already been shown to cause more secondary extinctions, at least in grassland ecosystems, and the new study's findings support this.

What's more, not all levels of the ecosystem are as vulnerable to this unraveling effect. Those at the top of the food chain, like large fish, were more likely to be negatively impacted by secondary extinctions than those on lower trophic levels, like plants.

"Protecting certain species that are disproportionately contributing to services either by supporting them or directly providing them can potentially help mitigate any indirect threats," suggests Keyes.

"I think this approach could be a way to better allocate resources to protect multiple species and services."

The researchers caution their model does not yet incorporate many interacting factors - it is only a first step in understanding the impacts of secondary extinctions. There's also a lot of work to do to identify key species and their complicated tapestries of interactions in the field, especially for large-scale interactions.

A recent report from the Biodiversity and Ecosystem Services Index suggested loss of "ecosystem services could be significantly slowed down or even reversed if the role of biodiversity and its full contribution to economic production were an integrated part of decisions made by governmental entities, companies, and other stakeholders."

Keyes and team are planning to investigate if the same factors that make food webs more resilient against extinction also help maintain ecosystem services.

This research was published in Nature Communications.