The surprising similarity between rivers and trees

Rethinking rivers mini series, Post 1 of 3: Rivers are much more than channels of flowing water, they're networks.

It's been World Water Week this week. It may be wrapping up, but the conversations it sparked are just beginning — and they matter more than ever. Over the next few days, I’ll be sharing a three-part series on some unique aspects of rivers and why they’re so challenging to manage. We’ll explore how rivers function as networks, their unique rhythms, and what that means for conservation and restoration. Don't worry, these won't be long-form deep dives. Just short, sharp pieces to keep you engaged and thinking.

Freshwater needs all the help it can get. By sharing these posts, you’re helping raise the visibility of ecosystems that are often overlooked — and increasingly at risk. Let’s get more people thinking about rivers not just as scenery, but as living systems worth protecting.

When most people think about a river, they picture a channel that carries water out to sea. Often, people fear rivers because of their destructive force. If you're lucky some people value them for the food they provide. But far too many rivers are undervalued.

Even river ecologists historically thought of rivers as linear systems that change from the headwaters to the sea. To be fair, they are and they do. But this view overlooks one key element. Something that shapes how biodiversity — the bugs, the fish, the microbes, the birds — is organised in rivers:

Rivers are networks.

More specifically, they're hierarchical dendritic networks. Who the what now? Let's try that again. They're organised like trees (dendritic) that go from small headwaters and branch into large mainstems in a way that's highly connected.

The River Continuum Concept, one of the most famous papers in freshwater ecology, posed that rivers followed a predictable sequence from the headwaters downstream. Here, small headwaters are dominated by energy entering the stream from outside (leaf fall) and little light entering the stream bed. Middle reaches switch to production instream via photosynthesis by algae and diatoms due to more light entering the stream. Finally, large rivers downstream are again dominated by the processing of organic matter, this time in fine particles passed from upstream, with little light entering the river, preventing photosynthesis. Much of the ideas behind this paper from 1980 still largely hold after all these years as a general concept, but many exceptions have been found since and there's more to rivers are now thought of as networks rather than longitudinal continua.

What's very cool here is how biodiversity is organised. The obvious assumption would be that there's more diversity in larger downstream habitats. And that's correct — if you're thinking solely about how many species are at a particular location. But if you zoom out and ask where biodiversity sits more generally across whole river networks, it might just be the opposite. Sure, each of those tiny little branches might host fewer species, but they're different from each other. Add all that branch-to-branch variation up, and you get a lot more diversity¹.

And it's not just because the conditions in each of these small headwaters differ — they might actually be the same. But it's also that species can't necessarily get to each of them. So differences in dispersal rates of species mean some species can get to some headwaters and not others, leading to variation in who gets where. As a result, you get greater biodiversity among all the different branches. This is less so for downstream habitats. So they tend to be more homogeneous.

Even creating river-like networks leads to characteristic patterns of biodiversity compared to simplistic landscapes. Figure modified from Carrara et al. (2012) PNAS. The authors used microcosms to setup experimental river networks and found simply connecting microcosms in a way that mimics a river network (right) led to completely different patterns in biodiversity compared to a uniformly connected system (left) where all species were able to get everywhere.

The design of the experiment run by Carrara et al. (2012) PNAS. I'll save you the details of what's in each panel, but B shows the protists, C-E show the microcosms and how they were connected for each design.

So yes, rivers are connected longitudinally — along their length — and form into networks.

Physical landscapes channel species movements, so biodiversity patterns emerge from both who can get where and how the network is structured. How species disperse within, along, and among these different streams is dictated by their physical form. Some flying species get trapped at headwaters and cannot pass mountain ranges (e.g. c), others are highly isolated and connectivity is limited to only the best flying dispersers (e.g. a), and others are limited to dispersal along the channel due to gorges blocking lateral dispersal pathways (e.g. b, f). Figure taken from Tonkin et al. (2018) Freshwater Biology.

But they're also connected laterally to their margins — if they are functioning naturally. Floods pulse into their floodplains, replenishing soils (there’s a reason we farm on floodplains!), allowing fish to forage on the rich terrestrial life, transporting seeds of riparian trees into their floodplains. These flood pulses are fundamental to many tropical rivers because they're so predictable. And when rhythms are predictable, life evolves to sync with them.

One of my favourite figures of all time from Hauer et al. (2016) Science Advances showing how gravel-bed river floodplains are the ecological nexus of regional biodiversity. I won’t explain what each sub-panel represents, but you can see that rivers are more than just 1-D conveyors of water to the sea.

And this lateral connectivity isn't all one-way — it's reciprocal. Aquatic insects emerge from rivers and feed riparian bats, birds, and beetles. In reverse, terrestrial insects fall into the water, feeding fish. Leaves drop, fuelling microbes and shredding insects. Energy flows out onto the land, and then flows back in. These exchanges form the basis of food webs that connect the land with water. They're seasonal, essential, and fascinating.

Energy pathways between rivers and the land are reciprocal. Copied from paper: "A generalised diagram showing reciprocal flows of invertebrate prey and inputs of plant material (dark arrows) that have direct and indirect effects in stream and riparian food webs." Source: Baxter et al. (2005) Freshwater Biology.

And let’s not forget vertical connectivity. Water upwells from below, bringing nutrients and cooler temperatures. Groundwater and surface water interact in ways that shape habitat quality, thermal regimes, and ecological processes. And specialist life occupies these places, whether in the springs that provide respite from temperature extremes, or species occupying the deep, dark places several metres under the surface.

Where groundwater enters or leaves a riverbed shapes temperature and productivity. Here, upwelling zones support more algae (right). Source: Hauer et al. (2016) Science Advances.

Rivers are not just water flowing to the sea — they’re living networks that connect land, air, and underground worlds. But they have been and continue to be fragmented, channelised, and stripped of their connections — often in the name of progress. When we break those connections, we don’t just lose species. We lose resilience. We lose the ability of rivers to recover, adapt, and support life. And we continue to do it at pace. So let’s share the love for healthy, functioning rivers! And let's keep these connections for the sake of the life they support.

Coming up in the next post

Rivers aren’t just connected in three dimensions. Stay tuned for the next post that talks about the fourth dimension, time!

If this post resonated, consider subscribing to follow the rest of the series.

1. And biodiversity is fundamental for a variety of reasons:

Diversity is the spice of life