Good Ole Spurge

For those of you that have had the pleasure of spending your hot and humid summers on top of roofs pulling weeds, I would like to know how everyone manages SPURGE.  We routinely encounter three species, the most common being prostrate spurge.  This plant thrives in extensive roof systems.  This summer seems to be a pretty rough one for this weed.  Manual removal often feels pointless and I am very hesitant to apply chemical based pre-emergents.  Are my concerns warranted?  Has anyone had success with the corn gluten based pre-emergents?  Does spurge get worse when nutrient conditions are lower or higher?  



Thanks for anyone taking time to respond!



Replies to this Topic


That's pretty funny....we were just talking about this the other day, how there has been such prolific germination of spurge and other common weed seeds this year after we received all those heavy rains in early August. Next year we are going to try another method and approach when installing some of our vegetated roofs in the summer for weed control called solarization that some gardeners have been practicing as a way to significantly reduce germination of weed seeds.

Soil Solarization:

If you have the time and funding set aside to incorporate this practice then soil solarization might be not only a good way to reduce weed pressure but also to help kill off some soil borne pathogens like crown and root rot. Before planting an area with sedum and other succulents, you might want to follow the management method known as soil solarization. The general premise is that after you install the growing media you saturate the soil and then cover the soil with sheets of clear or black plastic for 4 to 6 weeks. Incorporating this practice prior to planting especially during the summer can effectively reduce the number of seeds in areas where summer daytime temperatures are very hot. While, we haven't personally tried this practice in theory it makes sense and certainly beats having to put down a chemical control or pre or post-emergent like triclopyr (Turflon) and glyphosate. I have uploaded a report to the file cabinet that discusses this practice in more detail.


Most weedy spurges are summer annuals that don't like competition and depend on their prolific seed production for survival. A single plant can produce several thousand seeds, which are small and can remain dormant in the soil until conditions are suitable for germination (sprouting). Seeds produced in summer germinate immediately while those produced in late fall mostly will lie dormant and won't germinate until spring.

Spotted spurge germinates best when temperatures are 75° to 85°F, but germination can occur at temperatures as low as 60°F and as high as 100°F. When moisture is available, germination can occur from February through September in the D.C. area. Light also is a requirement for maximum germination; seeds buried deeper than 1/2 inch won't germinate well. Plants that germinate early in spring in cool conditions can remain as small seedlings until temperatures are more desirable for growth. Once the seed germinates, a small rosette of leaves develops. As growth continues, the leaves form a dense mat that can grow up to 3 feet in diameter. Reproductive growth is rapid, and the plant can produce seeds as soon as 5 weeks after germination.

 The primary method of managing spurges is prevention, since controlling these weeds is very difficult once plants have established themselves. Avoid bringing seeds into uninfested areas by using weed-free planting seed and uncontaminated planting stock. Clean work clothing and machinery such as lawn mowers to remove any seeds that might be present, and remove spurge plants as soon as you discover them.


Constantly monitor infested areas, so you can hand pull new plants before they produce seed. Take care as you weed, since plants that you hand pull often break at the stem, leaving the root and several buds or a single stem from which regrowth is possible.  Wear gloves when you hand pull, since the sap can be a skin irritant. Mowing is an ineffective method of control, since most species grow closely to the ground.

When planting new, container-grown ornamentals and ornamental beds, be sure to use sterilized or weed-free planting mix. When purchasing plants for ornamental beds, avoid those with spotted spurge infestations. Mulches can effectively limit spotted spurge if they prevent light from reaching the seed.


Is probably the most common strategy for controlling weeds in semi-intensive and intensive roof gardens and because they prevent light from reaching weed seeds and seedlings, this helps to starve them before they can start making food through photosynthesis. Bark and wooden compost at least 2 inches thick can effectively control many weed seeds including many spurge species. A large, coarse bark will require a 3- to 4-inch layer to be effective; however, larger, coarser mulches last longer than finely shredded ones. Thick mulch eventually can accumulate soil, decaying organic matter, and weed seeds that can germinate. All organic mulch needs periodic replacement.

Black, synthetic polypropylene weed barriers (fabrics or geotextiles) also block sunlight and starve weed seedlings. The fabrics are porous to allow water to drain through them. Often a synthetic barrier with bark or rock on top makes the area more aesthetically pleasing. Organic mulches such as bark and straw don't need to be as thick if you also are using the fabric. Since mulches and weed barriers reduce evaporation from the soil surface, adjust the irrigation cycle to prevent overwatering. 

Chemical Control:

Pre-emergent herbicides can help prevent spotted spurge outbreaks if you apply them in late winter before weed seeds germinate. Time the application, so it occurs before the soil temperature exceeds 55° to 60°F at a depth of 1 inch.

Pre-emergent herbicides for turf and greenroof areas include benefin (Balan), pendimethalin (Pendulum), isoxaben (Gallery), oryzalin (Surflan), trifluralin (Treflan, Preen), and dithiopyr (Dimension). Of these, only pendimethalin, trifluralin, dithiopyr, and oryzalin are available for use by home gardeners. Combination products such as oryzalin plus benefin are available to both home gardeners and landscape professionals.

Pre-emergent chemicals are almost never used in home vegetable gardens, because chemical residues last for months after application, and product labels routinely regulate against such use.

Post emergent herbicides available to home gardeners include 2,4-D/MCPP/dicamba combination products, triclopyr (Turflon), and glyphosate (available for both commercial and home landscape use). In general, 2,4-D and its combinations don't control the larger, more mature spotted spurge plants.


Edited Sat, Aug 21, 2010 10:29 AM

Ben- Greg is right on when he mentions weed  prevention.  It is best to patrol the newly planted green roof soon after planting and remove the weeds before they get started.   Most proprietary media should be weed free and warranteed against containing weed seeds. In my experience weeds are most likely to come in with planting stock, but his comments about clean work clothing and equipment is good advice,  The plant supplier should also warrantee freedom from weed seed. Put it in the specs.

Once a green roof is established and the limited media is filled with plants the competition will keep weeds down.  Never let weeds go to seed.  Sedum on roofs is problematic, because it has a weaker and less competitive root system when compared with most plants.  It just lays there.

I am not in favor applying any chemicals to green roofs.  Yes, it can be done if one is careful, but we all know that there are idiots out there working on landscape crews that are NOT careful when applying chemicals and fertilizers even though we train and preach and constantly monitor them. Remember there is much less of a buffer between a mistaken spill or over application with the very quickly draining green roof media.  Roof drains are often connected to cisterns and storm sewers and that increased dispersion of chemicals into the environment.  

Richard Sutton, PhD, GRP

Landscape Architect

Professor Agononmy and Horticulture



Please forgive my ignorance.  Even though I have acquired the GRP designation from GRHC and am a part time gardener, my profession is roof consulting, not horticulture or landscape design.  Whats wrong with having some spurge or purslane on a roof? 

I can understand that you wouldn't want spurge in a specific landscape design because it would detract from the landscape architects concept.  However, on most extensive green vegetated roofs, just getting plants like drought resistant sedums to grow and make a root system is the challenge.  It seams that some spurge or purslane would add diversity and have some benefits.  What is spurge's negatives?

Robert- It is a biodiversity issue. Like any annual weed spurge can proliferate to the point of crowding out other desirable plants and become the only plant.   And while weedy annuals do hold the media overtime, they can sap nutrients for perennials.  



Thank for the quick reply.  I should have known, that since it is an annual, areas of spurge concentrations would also be bare spots during winter months.


I agree with Richard on the plant bioversity front but also remember that the spurge we are talking about is an annual and will offer not media stabilization from October through May, yet another problem with annual weeds.

In terms of herbicides, I would submit that we think of green roofs as an aquatic herbicide environment simply because many green roofs will drain into a storm water system and be very closely connected to a raparian system. That coupled with the fact that green roof media is freely draining compared to field soils and lack the clay particles to break down the herbicides and render them benign.

I also agree with Richard that proper care during the establishment period will prevent excess weeding and there is no substitute for understanding the life cycle of weeds in your region. This systems can be fairly simple but that doesn't mean that knowledge and experience isn't of great value.

I don't agree with the sedum root comment, the only time I have seen sedums float on top of media is when drainage is poor and they have sloughed roots because the media has no oxygen. The larger leafed sedums actually have a more persistant root system than the smaller leafed species but more importantly they compete for survival in a thin non-irrigated system by just living where little else can live.

All of this to me depends on your design intent. There are green roof in Europe that they let self generate and by defintion have no weeds, others are designed for visual enjoyment and have to be gardened and they will have many weeds.

I got some great advise from Ed Snodgrass to use a small propane torch on these pesky weeds.  It not only kills the weed but also kills the seeds that would normally fall off during hand weeding!  We tried it last week and it worked great!  Just be careful around the sedum plants (since sedums hold water in their leaves and stems, they are less likely to burn quickly) This is also a great back-saver!   Good luck!


 A couple of additional comments:

Euphorbia maculata, E. supina can produce viable seed in as little as two weeks after emergence (Weeds of the Northeast, p. 226), and I sometimes suspect  it can be shorter.  This makes weed control critical in the early stages.  Only when inspection shows little recurrence should one risk extending the inspection rate beyond two weeks.  

The nature of the sticky sap means it is very possible to pick up spurge seeds on one's shoes in a parking lot or especially in disturbed construction soil.  Then up on a hot, dry roof, the mud and sap can quickly be shed, inoculating the roof with seeds.  So cleanliness and awareness are important.  On a green roof, look for a different weed density near the points of access.  you might see more weeds near an access hatch or doorway, especially if the roof is frequented by 'visitors', who step out just a few feet to admire the roof.

I do not agree with the comment about a warranty against weeds.  A warranty is no substitute for proper maintenance.  With such a short life cycle; it is impossible to tell if weed seed originated at the nursery, in transportation, during installation, or subsequently.  Weeds tend to germinate in the area around the plugs because that material is generally more moist and therefore more favorable to germination.  A concerned installer should inspect the nursery of origin - if the plants and open areas are weed free, then it is unlikely that major infestations originated in the nursery.

Finally, the torches that Janie mentioned are available at Lowes, Grainger, and other places.  They are great on the back and effective on the spurge.


Edited Wed, Sep 29, 2010 2:52 PM

I absolutely agree with John. There are uncountable ways in getting living organisms introduce to a green roof. This can be seeds, spores, roots or other plant parts but also viruses, bacteria or insects. Each component can contain a certain contamination and suppliers can't full control since components are exposed to the elements. I don't support the term "weed" (not because of the memories of my teenage time) but the term weeds triggers a different picture in any person. A lot of so called "weeds" for a farmer are wanted plants for green roofs. A nursery has a different picture of weeds too than what I would tolerate on a roof.  Using the term weed requires that you know who you are talking to.

Unwanted plants is the right term and any unwanted plants on green roof are most likely indicators and tell you something about the conditions, quality of the materials or even about the qualification of the installer. Example:  excessive watering (i.e. temporary irrigation systems) promote a whole variety of unwanted plants that are sitting in the system and suddenly find perfect conditions. Compaction, high organic content in the soil or around plugs triggers other unwanted plants and in combination with any "maintenance in good faith" we promote the unwanted and forget that the wanted might have other needs. There is an endless list about indicators.

The most efficient way to control unwanted plants: 

Maintenance starts with the design i.e. Plant selection continues with installation (gardener  101: prevention) i.e. break plugs carefully apart on a roof remote location, get rid of this soil and don't spread that on the roof. Green thumb i.e learn about unwanted plants and what they typically prefer.  Promote the wanted plants i.e watering might not a good promotion for that. Train your eye on unwanted plants and develop a tolerances for them - it is biodiversity, nature and some unwanted actually help the wanted. Train your back bone - only manual removal with a trained eye and good judgment works. Chemicals, lights, torches or machines might help under certain conditions - rarely on extensive green roof - . The key is simply your back bone and your eyes - fast, efficient, environmental friendly and reliable. That's waht installers in Europe do who build over 150 million sft every year. I guess there is still much we could learn from the old world?  Cool jbi

I forgot to mention: prevention doesn't include any wired technologies or chemicals. Everything needs to be affordable, efficient, environmental friendly and without harm for the investment (roofing, building).

Nobody in Europe is utilizing any of these wired ideas because of the reason I just mentioned. Over the last 30 years there were countless movements (including the mentioned) but non was successful. There is no machine better than a living machine. jbi  

This thread is relevant to the recent conversation about weed prevention and green roof maintenance.  I'd be interested whether anyone has any new thoughts?

Post Reply

You must be logged in and a member of this Groupsite in order to post a reply to this topic.
To post a reply, contact your group manager(s) Join this Groupsite


Capitol Greenroofs
Powered by

Visibility Public Membership By Invitation or Approved Request Default Profile Professional

Your Status Not Logged-In