Green roofs are engineered ecosystems with unique nutrient cycling characteristics compared to natural systems. This uniqueness is due to the designed and selected components that have not co-evolved over long periods of time. Hence, direct comparisons with the nutrient dynamics of natural systems can be flawed, which is why green roofs deserve, and require, further dedicated research attention to develop best practices for our current and future cities.
What do we know about green roof nutrient runoff?
Green roof substrates generally have not been designed with runoff quality in mind but have been optimized based on engineering parameters such as load and drainage rates. Nutrient runoff from green roofs is a valid concern due to nitrogen and phosphorus leaching, leading to eutrophication of local water bodies.
What is the current state of research on this topic?
Nutrients in vegetated roofs
Nitrogen and phosphorus are commonly limiting to plant growth in terrestrial ecosystems and key nutrients on vegetated roofs. Green roofs provide a unique ecosystem in that high nutrient substrates are coupled with plant species that generally are adapted to low-nutrient environments (so-called extremophiles) [1]. Furthermore, green roofs are hydrologic catchments but with unique properties compared with a natural watershed. The green roof hydrologic system is mainly built to drain fast, but others have added layers with the aim to increase retention (e.g., mineral wool). This variation in hydrologic behavior coupled with variations in climate and roof design leads to a highly variable hydrologic residence time between roofs, which will affect nutrient cycling immensely and sometimes make between-roof comparisons difficult.
There is also a substantial seasonal variation in nutrient runoff, making snap measurements difficult. To get a good view of the actual nutrient runoff, frequent monitoring over several seasons is required.
Load vs. concentration
It is important to remember the difference between load and concentration when discussing nutrient runoff from green- vs. bare roofs. Hypothetical example for the sake of the argument: Two roofs, bare Roof A and green Roof B, have the same nitrogen (N) concentrations in the runoff. However, Roof B has an annual 50% reduction in runoff due to plant evapotranspiration, compared with Roof A. Hence, the yearly nutrient load will be much lower for that Roof B despite the runoff having the same nutrient concentrations in question.
Nutrient limitations
To understand nutrient dynamics and fertilization decisions, it is critical to learn a bit about nutrient ratios and limitations. A nutrient is limiting if adding more of that nutrient increases the system's productivity. It is a bit like building cars and running out of a particular critical bolt. No matter how many extra tires and engines you have, there will not be any more cars built until you get more of that specific bolt. That bolt is the limiting factor.
Nitrogen and/or phosphorus availability are generally the limiting factors for primary production in most ecosystems. Nitrogen is the most commonly limiting nutrient in terrestrial systems [2]. When we look at nutrients, it is critical to look at ratios. It is about getting a balance between the nutrients to optimize growth. N:P co-limitation is generally at a mass ratio of 15:1. Higher ratios lead to P limitation and lower ratios to N limitation [3]. It might be worth mentioning that secondary production can also be limited by other factors such as the supply of labile organic carbon. A use case would be microbial decomposition, where the microbes must have something to break down (labile organic carbon) or their activity decreases.
The cycling of N and P in unmanaged ecosystems is exceptionally efficient. More than 90% of the N and P uptake annually stems from "recycled" nutrients from, e.g., last year's leaves [4] This is currently not achieved in green roof systems.
Nutrient ratios on green roofs
Green roof substrates are often very nutrient-rich to support the plant community. The organic fraction of the substrate is commonly P-rich compost that has also been supplemented with fertilizers containing P, N, and potassium (K). This leads to a situation where there are equal amounts of P and N despite the 15-fold demand of plants for N compared to P [3].
The importance of mycorrhizae
As the plant acquires the nutrients through the mass flow of water from the roots to the shoots, the high nutrient hotspots around the roots are essential for the plant's health. The need for the plant to access nutrients is also the base of the common mycorrhizal symbioses that often form between these types of fungi and the plant root systems. The fungi effectively increase the roof surface area and thus increase the plant's access to nutrients. In exchange, the fungi receive carbohydrates from the plant roots.
The plant-mycorrhizae interactions on green roofs are still very under-studied and future research into this field could lead to new and exciting insights into the improvement of green roof nutrient cycling to achieve lower fertilization needs and reduced nutrient leaching.
Nitrogen
Nitrogen is often the limiting nutrient in terrestrial ecosystems. Microbial nitrogen fixation of atmospheric nitrogen, e.g., cyanobacteria or other nitrogen-fixing organisms, supplies the central part of the reactive nitrogen species in an unpolluted terrestrial ecosystem [3]. NO3 and NH4 are added through atmospheric deposition.
NO3, NH4, and NO2 are accessible to plants and microbes and can be assimilated into different pools. Microbes then mineralize these pools to form NH4 through the decomposition of organic matter. Nitrogen runoff is often reduced with the age of the green roof and then varies with fertilization practices and atmospheric deposition. Nitrogen runoff from a green roof also varies with plant species composition and changes in the microbial community that affects the balance between the different nitrogen species and thus nitrogen mobility and leachability of the soil.
Nitrogen mobility in soil
NH4 binds on cation exchange surfaces and experience slow diffusion. NO3 diffuses rapidly through the soil media and can leach out.
Phosphorus
Water-soluble and biologically accessible phosphorus sources are the mineralization of organic material by microbes and weathering of rocks in a terrestrial ecosystem. Unlike nitrogen, little phosphorus is added to the ecosystems via atmospheric deposition.
Biologically accessible phosphate can be taken up by the plants and microbes, adsorbed onto surfaces, precipitated out of solution, and bio-unavailable or lost via runoff. As phosphate easily forms precipitates in the soil, it generally diffuses very slowly [3]. It should be noted that green roofs often use high-phosphorus compost as part of the organic fraction.
Extensive green roofs have frequently been shown to be a phosphorus source. However, they can be either a sink or a nitrogen source depending on roof age, substrate, climate, microbial community, and plant palette. Phosphorus can leach out over long periods, likely due to the mineralization of organic material rich in P.
P runoff concentrations from green roofs vary a lot between measurements and roofs; from 0.025mg/L [5] to a whopping 29mg/L [6], the latter is comparable to concentrations found in wastewater (3-10mg/L). In the UK, 70% of the P entering local water bodies is due to sewage discharges [7]. This P is the primary source for eutrophication, hypoxia, anoxia, and thus the death of local water bodies, which underlines the importance of a detailed understanding of the nutrient dynamics and hydrologic properties of the entire urban hydrosphere as a basis for best practices, political and legal decisions.
The natural dynamics of nutrients
Before entering outrage mode, it is vital to take a step back and ponder what the high P measured in the green roof above really means.
Firstly, we must be critical about snapshots, especially when dealing with nutrients. Nutrient measurements are extraordinarily fickle, and a snapshot in time is only one part of the story. Nutrient leaching varies immensely between seasons but can also fluctuate within an event, between events, and between years. One example is the well-known phenomenon of the spring flush in natural aquatic systems, e.g., the Baltic Sea. After the thaw, with no vegetation actively growing and taking up the nutrients, nutrients are flushed out to sea, becoming food for fast-growing plankton [8].
The below video shows the impact of high phosphorus on the Baltic Sea. These cyanobacterial blooms are extremely serious from an ecological perspective but are also costly for society due to the negative impact on tourism and fisheries.
A measurement taken during the flush cannot be applied to the whole year, which would be immensely skewed. The measurement would not be wrong; it would just be only one chapter of many. Therefore, long-term monitoring programs are critical to getting a more comprehensive view of green roof nutrient dynamics.
A well-known flush effect in green roofs is "the first flush," It describes the high nutrient loads often found in newly constructed vegetated roofs. Several studies have shown that the ability of a green roof increases its potential to retain P with time [9]. This has been attributed to plant establishment and would be an argument for ensuring the roof is covered by vegetation as fast as possible, e.g., by using mats instead of plugs or seeds. Hence, it is essential to always put a specific measurement in context before making any assumptions.
The future
There is no doubt that green roofs are an essential part of our future sustainable cities, and our goal is to improve this important technology further. We should also keep in mind that though green roofs might, at times, be nutrient sources, lawns are a much more significant source, especially in the US. Regulating lawn fertilization and irrigation would do much more for our environment than getting all green roofs to a zero-leaching state. Urban meadows are a tremendous biodiverse replacement to lawn monoculture.
Though perhaps I am making myself guilty of "whataboutism." If we aim to make green roofs a sustainable standard for most buildings, we must take green roof leaching seriously, and many already do.
There are exciting developments on the horizon, with new and exciting innovations to look forward to over the coming years. Green roofs are an integral part of our future urban lives; a necessity to cool our cities, reestablish the natural water cycle, capture pollution, and so much more. Exciting times ahead!
1. Buffam I, Mitchell ME, Mitchell ME. Nutrient Cycling in Green Roof Ecosystems. doi:10.1007/978-3-319-14983-7_5 2. Vitousek PM, Howarth RW. Nitrogen limitation on land and in the sea: how can it occur? Biogeochemistry. 1991;13(2):87-115. 3. Chapin FS, Matson PA, Vitousek PM. Fundamentals of terrestrial ecosystem ecology. 2011. 4. Likens GE. Biogeochemistry of a Forested Ecosystem. Springer Science & Business Media; 2013. 5. Gregoire BG, Clausen JC. Effect of a modular extensive green roof on stormwater runoff and water quality. Ecol Eng. 2011;37(6):963-969. doi:10.1016/j.ecoleng.2011.02.004 6. Vijayaraghavan K, Joshi UM, Balasubramanian R. A field study to evaluate runoff quality from green roofs. Water Res. 2012;46(4):1337-1345. doi:10.1016/j.watres.2011.12.050 7. Bunce JT, Ndam E, Ofiteru ID, Moore A, Graham DW. A review of phosphorus removal technologies and their applicability to small-scale domestic wastewater treatment systems. Front Environ Sci. 2018;6(FEB):1-15. doi:10.3389/fenvs.2018.00008 8. Piiparinen J, Kuosa H, Rintala JM. Winter-time ecology in the Bothnian Bay, Baltic Sea: Nutrients and algae in fast ice. Polar Biol. 2010;33(11):1445-1461. doi:10.1007/s00300-010-0771-6 9. Caldwell Logan MM, Sandra Díaz Cordoba U, Gerhard Heldmaier Marburg A, et al. Ecological Studies Analysis and Synthesis Volume 223 Series Editors. http://www.springer.com/series/86.