Plants Really Work

“You mean, plants are useful in stormwater systems?”

Many decision-makers don’t see plants as useful in stormwater systems. Sadly this leads to downplaying plants’ role in stormwater mitigation and choosing a pipe instead.

By treating water as a resource and designing water into the landscape from the start, all piped networks can be avoided and expenditure reduced by a minimum of 10%. Land values rise with this approach, leading to higher sale prices.

— this is for one layer of plants

When you have the whole web of plant life – trees, large shrubs, low shrubs plus bulbs, grasses and perennials, plant handling of stormwater is in the 70% range.

— click on ANY images below to enlarge

Plants Really Work!

Catch and hold 🙂
Muehlenbeckia​ astonii, a New Zealand native (endemic). Winter appearance

Rain shadow protection from a linden lime, Tilia x europea.
Dunedin, New Zealand

Centre-of-driveway storm garden. Kew, London UK


Snowed under. Even small plants have a role to play …


It’s a systems approach


Some site data is needed so that a plant solution can be tailored:

  • evaporation by month
  • storm frequency by month
  • largest storm
  • soil water-holding volume

Stemflow (represented in the above diagram by ‘Tree holds – 2-8%)

This image shows a hard-to-measure phenomenon called stemflow. With heavy rainfall the whole plant becomes wetted and water starts to flow down the branches and trunk – dark areas underside of branches in the image. This is very variable and ranges from 2% to 8%. Many trees absorb this water into their bark with little to no water reaching the ground. Stemflow is often either disregarded or over-estimated in stormwater calculations.

Almost all sites have the potential for some water infiltration to shallow and deep groundwater. Plant roots assist this process where roots create drainage channels down into the earth.

The ability of a plant to act as a water pump can also be increased by choosing the right plant, altering soil acidity and placing the plant where it can capture the most rainfall.

Strategic planting: these tree roots grow down into a swale assisting drainage while also reducing runoff from above. Queens Park, Invercargill, Southland, New Zealand

Santa Monica Urban Runoff Recycling Facility – SMURRF. LA, USA

This site is a technical filtration system with a natural backup comprising a dense sward of Bermudagrass (Cynodon dactylon). Bermudagrass is used in stormwater and wastewater systems as it tolerates and destroys many pollutants.

SMURRF Virtual Tour

Plants Really Work
Text, photos and graphics by Nigel, editorial support: Liz. Growplan (2020)

18 thoughts on “Plants Really Work

Add yours

  1. It is always amazing to me that civic planners often have no idea about all that is required to make some decisions. Of course plants are important in rainwater control. Otherwise hillsides would all run into streets and brooks etc. Weeping Willows-Salix babylonica are often planted in wet areas to take up water that collects in low lying or poorly draining areas here. And anyone who has sought shelter under a tree during a downpour knows that the leaves capture a lot of what would otherwise hit the ground.
    I like how your diagram shows the many ways a tree deals with water. I notice the arrows for treeholds and believe excess water often is part of groundwater recharge.

    1. Thank you for the comment Steve, re stemflow (and other ways of holding water back from draining off immediately) and groundwater recharge. Absolutely, SUDS is all about holding water back in a series of stages (although I have one project where loss to groundwater is 0% due to clay and being a low point in the landscape).

      In a natural system precipitation runoff is about 10% and approx 10% to 25% goes to recharge, with SUDS we’re seeking to replicate the function of a forest/savanna/grassland etc while still building houses and roads and towns.

  2. Excellent presentation. And I don’t recall running across any discussion of “stemflow” as a factor. I like the appearance of the center-of-the-drive planting.
    I read recently of a city in upstate New York, which found it indisputably cost effective, to begin switching to permeable pavements and rain gardens, to reduce the storm-water runoff, which was overwhelming the combined surface runoff/sanitary sewer system (which is pretty common in the older NE cities). Another recent article on phytoremediation, in the NYTimes, discussed the use of poplar plantings on superfund sites, where the trees have been inoculated with naturally-occurring microbes, whose enzymes break down carcinogenic chemicals and toxins.
    So they’re not just preventing contaminated runoff, into groundwater or lakes, but actually destroying the pollutants at the same time.

    1. Thanks Robert, yes plants have many of the answers, and help avoid costly hard engineering, and in partnership my soil fungi some really amazing things are possible with pollutant breakdown.

  3. I am fascinated by the diagram, Nigel, and how well it displays so many of the different factors that come into play and how they interact with each other. It also helps folks like me who have little experience in this area to understand better what you are showing in the individual photos. Yesterday we had thunderstorms and torrential rain and there was a huge amount of runoff and some localized flooding–I suspect that our local stormwater efforts could use some additional efforts.

    1. Thank you Mike! I’m vey happy you find my graphic helpful, some thought went into distilling the info without using too many words – when I studied landscape more words meant lower marks!

      I’m interested in your local storms; do you think you’re getting more rain from single events? Have you noticed if conditions underfoot are the same as in years gone by, or wetter, or drier?

      1. The previous two years were really wet years with a lot more rain than the historical average. Yesterday’s storms were stronger than normal–according to the newspaper we had a bit over two and a half inches (64 mm) during the last 24 hours, which is almost as much as we normally get for April. It appears that there are areas in my neighborhood where water always seems to collect and that are wet for lengthy periods after a rain. I visited a pond yesterday and fishing ponds were swamped and a nearby stream was way up over its normal banks.

        1. Thanks Mike, the same here re heavy events; it all gets smoothed out by the statistics people but ask a farmer and they tell of loss of seasons and almost all rainfall occurring in intense cloudbursts.

          I have a post in the works for another time on how the rising CO2 level is making plants more efficient in their water use and leaving more water to drain off the land.

  4. Exploitation of plants for some of such purposes is more common in other cultures. Common cannas that are grown as ornmentals here are commonly used to absorb toxins from polluted water, although I sort of wonder what they do with those toxins.

    1. Toxins, Yes. Well there’s no free lunch – just as hard stormwater systems need cleaning, plant-based systems need a certain amount of ‘harvest’.

      This is especially true for the metals, and the so-called Highly persistent, Bioaccumulative and Toxic substances – PBT, some things like asbestos can probably be broken down (through mycorrhizal breaking of the iron bond), but many others break down to a point but remain toxic, e.g. flame retarders, herbicide secondary products, … plants can retain many of these products and stop them ending up in the sea and in us, but we have to look after the plants to keep them working – it still costs a lot less than a piped network.

      1. Some of elegant landscapes around where I worked in Southern California included an awkwardly placed colony of cannas. Some landscapes were quite formal and symmetrical, with only an odd herd of cannas to disrupt the symmetry, or interfere with the expanse of a lawn. They were quite random, but were installed wherever ground water appeared at the surface. With few exceptions, there was not enough water to flow away, but there was just enough to make the ground soggy. It was not toxic, but smelled badly. Some of it came up with oily residue. Cannas somehow prevented the ground from getting too swampy, and eliminated the objectionable aroma. There was concern after even a slight earthquake that such a flow might appear within one of our landscapes, but it never happened. (Those things can stop flowing in some spots, and start flowing somewhere else.) It actually seems to me that it is less of a problem than it was in the 1980s. Most people in the region are not aware that oil continues to be pumped from the Beverly Hills Oil Field.

        1. That’s very interesting re Canna; I’ve been reading about “Zero methane emission bogs” in Patagonia where Astelia (a Southern hemisphere monocot) predominate. One pathway for wetland methane out-gassing is low carbon in an oxygen-poor environment. It seems several species can put roots and thus carbon into these deep wet soils.

          Oil in the Los Angeles basin, last time we we were there I met an academic who had designed a plant mixture for taking up the toxic chemical brew at old well heads.

          He pointed out a number of buildings that initially looked like offices until you looked closer – oil production wells are everywhere e.g. Culver City (Inglewood) active wells

          1. Oh, yes! Not many people know about those. The biggest one has not changed in decades. It is well landscaped, but lacks windows and a front door. It is effective in the regard that no one notices it. Those on the islands off of Long Beach look like nice apartment towers.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Create a website or blog at

Up ↑

%d bloggers like this: