Our paper on the potential for bee movements between gardens and urban/peri-urban agriculture has been published in a special issue on Agroecology in the City, in the journal Sustainability.

Langellotto, G.A.; Melathopoulos, A.; Messer, I.; Anderson, A.; McClintock, N.; Costner, L. Garden Pollinators and the Potential for Ecosystem Service Flow to Urban and Peri-Urban Agriculture.Sustainability 2018, 10, 2047.

In this paper, we estimated how far the bees we collected from our Garden Pollinators Study could move between gardens and pollination-dependent cropland. We found that when pollination-dependent crops (commercial-scale or residential-scale) are nearby, 30–50% of the garden bee community could potentially provide pollination services to adjacent crops.

But, we currently know so little about bee movements in complex landscapes ~ if and how bees move across roads or through gardens embedded in housing developments. This question will be a focus of our future work.

Some of the bees collected from our 2017 Garden Pollinators study.
garden ecology lab

garden ecology lab

Urban agriculture has received a lot of attention over the past decade, as more folks are looking to localize their food supply, reduce food miles, and/or exert greater control over their food. Urban agriculture, however, brings a distinct set of challenges from farm systems in more rural regions. For example, urban farms tend to be relatively small and diverse (which can make it challenging to rotate crops), and are often close to neighborhoods and housing developments (which may make urban farms more prone to nuisance complaints). Urban farmers tend to be younger and to have less experience in agriculture, compared to rural farmers, and in need to high levels of technical assistance from Extension and other providers (Oberholtzer et al. 2014). However, many of the resources that Extension has to offer are focused on traditional growers, rather than new urban farmers.

Our lab group wanted to examine an issue that is specific to urban growers, and for which we could find very little information: urban agricultural soils. Soil scientists have prioritized research on urban agricultural soils as a key priority for the 21st century (Adewopo et al. 2014). Yet for his thesis work, Mykl Nelson could only find 17 academic papers that looked at urban agricultural soils in the United States. Most of these studies focused on

residential-scale or community-scale urban agriculture (in home or community gardens). Only one paper looked at soils on an urban farm.

Still, residential- and community-scale gardening is an important type of urban agriculture. In Portland, a conservative count of 3,000 home gardens collectively covers more than 20 acres of land (McClintock et al. 2013). In Chicago, residential food gardens cover 29 acres of land, and represent 89% of all urban agriculture (Taylor and Lovell 2012). In Madison, WI, more than 45,000 food gardens cover more than 121 acres of land (Smith et al. 2013).

For Mykl’s thesis, he looked at urban soils from 27 Master Gardener-tended gardens, in Portland and Corvallis, OR. Even though all gardens were tended by OSU Extension trained Master Gardeners, they were incredibly diverse: 74 different annual crops, and 58 different perennial crops were grown across these gardens. Unique crops included kalettes, papalo, thistle, savory, paw paw, quince, sea berry, and service berry, among others.

In terms of the soils, Mykl found that soils were within the recommended range for physical parameters, such as bulk density, wet aggregate stability, and soil compaction. However, home garden soils tended to be over-enriched in soil organic matter. Growers generally aim to foster soils that are between 3-6% organic matter. However, Mykl’s tested soils were on average 13% organic matter! Raised beds were on average 15% organic matter. In ground beds were a bit better: 10% organic matter, on average. So to put this another way, Master Gardener vegetable garden soils had 2-5X the recommended level of organic matter for productive agricultural soils. We suspect that Master Gardeners were annually adding organic matter to their soils, without necessarily knowing the baseline levels in their soils. Adding more organic matter, without knowing where you’re starting from, encourages over-applications.

Does that matter? Afterall, for years, we have been preaching that if you have sub-par soils, ‘just add organic matter’. Biological activity in these soils was great! But, the excess in organic matter promoted excess in several soil nutrients. Garden soils were over-enriched in phosphorus (mean phosphorus across all gardens was 2-3X recommended levels. Potassium in some gardens was 5X recommended levels! Gardens were over-enriched in magnesium and manganese, too. Nutrient excess was worse in raised beds, compared to in-ground gardens.

Unexpectedly, Mylk found the highest lead levels in raised beds. Often, we tell gardeners to grow their food in raised beds, to avoid heavy metal contaminants. Why would there be high lead in raised beds, if we weren’t finding elevated lead levels in nearby in-ground beds? We suspect that the lead might be coming in from compost waste that can be purchased on the retail market. If a compost product makes no nutritional claim, then it is exempt from analysis and contamination limits.

We can’t wait to finalize this work for publication. In the meantime, I wanted to share a brief update on this work.

Mykl will be defending his thesis on May 31st. We’re trying to arrange an online broadcast of the public portion of his thesis defense (1pm-2pm, May 31st). I will update this post, if we are able to get an online link for his presentation.

All bees have been pinned, labelled, and data-based. Now we’re (and when I say ‘we’re’, I’m mostly referring to Lucas and Isabella) are going through the painstaking process of photographing all specimens: head on, from the top, and from each side. We’ll then start sorting them by morphotype (how they look), and working to identify them. Some of the bees are very common, and fairly easy to identify (like Anthidum manicatum, Bombus vosnesenskii, Apis meliifera). Others will take a bit more time and expertise to get to species.

You can take a look at the entire album, representing about 150 of the nearly 700 collected bees. We’ll be adding the rest of the bees, as we can.

We collect and pin the bees, because most are difficult to identify, without getting them under a microscope, and without the help of a museum-level bee specialist. For those bees that are easy to identify by site (such as the ones listed above), we only collect one per garden (so that we have a record of its presence). We don’t collect multiple specimens of the same species, if we can identify it in the field. And, we don’t collect obvious queens (larger, reproductive bees).

We collect using a combination of water pan traps and hand collection. For hand collection, we use a pooter (an insect aspirator) for the smaller bees and baby food jars for the larger bees.

Water pan traps. We buy plastic bowls from the dollar store, prime them, and paint them with UV paint that is optimized for the wavelengths that bees see.
Here, I’m holding an insect aspirator, otherwise known as a pooter. You can suck insects off of flower heads without damaging blossoms, by carefully placing the metal part of the pooter, over the bee. It is then sucked into a small plastic vial, which I’m holding in my right hand.

This is such an exciting part of the research for me. I find myself obsessing over the photos, trying to organize them in my mind, and to at least get them to genus. Grouping them by genus makes it easier for an expert to sort through and identify them. And, I’m so grateful for their assistance, that I want to make it as easy as possible for them!

We’ve collected bees from gardens near Forest Park, in Portland’s city center, and in outlying suburbs. We’ll analyze the data to see if there are any patterns associated with garden location (forest, city, suburbs), or to see if there are specific bees that are only found in forest gardens, for example.

Getting ready to install plants at our field site.

The post below comes from Aaron Anderson, a M.S. student in the OSU Department of Horticulture, and a member of the Garden Ecology Lab.


This past summer, we conducted the first field season of a study screening native plants for their attractiveness to pollinators and natural enemies. We selected 23 native Willamette Valley wildflower species based on drought tolerance, as well as four exotic garden species known to be attractive to bees: Nepeta cataria ‘Catnip’; Salvia elegans ‘Pineapple Sage’; Origanum vulgare ‘Italian’; Lavandula intermedia ‘Grosso’.

Table 1.  Native plants selected for this study.

Plant Species Common Name Life History Bloom Color
Clarkia amoena Farewell-to-spring Annual Pink
Collinsia grandiflora Giant blue eyed Mary Annual Blue
Gilia capitata Globe gilia Annual Blue
Lupinus polycarpus Miniature lupine Annual Purple/Blue
Madia elegans Common madia Annual Yellow
Nemophila menziesii Baby blue eyes Annual Blue/White
Eschscholzia californica California Poppy Annual Orange
Helianthus annuus Common sunflower Annual Yellow
Phacelia heterophylla Varied-leaf phacelia Annual White
Acmispon (Lotus) parviflorus Annual White/Pink
Achillea millefolium Yarrow Perennial White
Anaphalis margaritacea Pearly everlasting Perennial White
Asclepias speciosa Showy milkweed Perennial Pink/White
Aquilegia formosa Western red columbine Perennial Red
Aster subspicatus Douglas’ aster Perennial Purple
Camassia leichtlinii Common camas Perennial Purple/White
Eriophyllum lanatum Oregon sunshine Perennial Yellow
Fragaria vesca Wild strawberry Perennial White
Iris tenax Oregon iris Perennial Purple
Sedum oregonense Cream Stonecrop Perennial Yellow
Sidalcea virgata Rose Checkermallow Perennial Pink
Sisyrinchium idahoense Blue-eyed grass Perennial Blue/Purple
Solidago canadensis Goldenrod Perennial Yellow

We planted them in meter squared plots at OSU’s North Willamette Research Center. Between April and October, we monitored floral visitation, sampled visiting insects using an “insect vacuum”, and tracked floral bloom.

With one season in the books, we have some purely anecdotal impressions of which wildflower species are the most attractive to bees. Goldenrod (Solidago canadensis) and Douglas aster (Symphyotrichum subspicatum) were both highly attractive to a wide diversity of native bees, as well as to a variety of beetles, bugs, and syrphid flies. As an added bonus, both these species had long bloom durations, providing habitat and colorful displays for significant portions of the summer. Annual flowers Clarkia amoena and Gilia capitata attracted a range of native bees; Clarkia was also visited by leafcutter bees for a different purpose – cutting circular petal slices to build nest cells with.

Bumblebee on Clarkia.
Syrphid fly on Goldenrod.

Results from this year need to be analyzed, and further research is needed to account for seasonal variability and to gather more data on floral visitors.

Additionally, w e will ask the public to rate the attractiveness of each of our study flower species in an effort to determine the best candidates for garden use. After a few more field seasons (and sorting lots of frozen insect samples!), the result of this study will be a pollinator planting list for home gardeners, as well as a pollinator and natural enemy friendly plant list for agricultural areas. These will help inform deliberate plantings that increase the habitat value of planted areas.


A soil pit is used to understand the nature of subsoil strata.
The Benton County Master Gardener demonstration garden was one of our soil test sites. This site had vegetables growing in raised beds, and in in-ground beds.
The Benton County Master Gardener demonstration garden used intercropping techniques to suppress weed growth in their beds.

This post is modified from a submission from Michael Nelson. It details lessons learned from his survey of garden soils, across Corvallis, Oregon, and the Portland Metropolitan area.  In September 2017, Michael sampled soils from about 25 gardens. These gardens used raised beds and/or in ground gardens to grow a variety of vegetables, herbs, and fruits. We wanted to study urban garden soils ~ and soils in raised beds versus in ground beds ~ for a few reasons. Specifically, we wanted to look at a few different questions:

  1. Do raised bed gardens offer greater protection from soil contaminants than in-ground gardens? In the Master Gardener Program, we recommend raised beds as a way to work around soils that may have heavy metal contaminants. However, heavy metals can become airborne, and deposited on soils from industrial emissions, traffic, and re-suspension of road dust. If this is the case, then gardening in raised beds might offer a false sense of comfort. We thus chose to sample gardens that are close to, versus further from, major roadways and traffic.
  2. Are garden soils deficient in some nutrients (such as nitrogen), but over-enriched in others (such as phosphorus)? With enthusiasm surrounding organic gardening and composting, we are wondering if repeated applications of compost might be contributing to nitrogen deficiencies, phosphorus leaching, or other soil nutrient issues.
  3. What is the general state of urban garden soils in Oregon? If we had to ‘grade’ soil health, by looking at soil structure, tilth, nutrients, and other biological, chemical, and physical characteristics of soils ~ what would that grade look like?

I asked Michael to write up a short report on his summer work. What did he observe in the gardens? What did he hear from gardeners? Are there initial findings or impressions he could share?  His report is below.


We began this project to examine differences between raised and in-ground garden beds in urban areas. We conducted a short survey of each site, where we noted weed pressure, garden area (in meters squared), and crops grown. We also noted any concerns voiced by the gardeners, about their vegetable production site. We sampled garden beds and kept samples separate depending on the type of bed (in ground versus raised-bed). We are now processing the soil samples in the Central Analytical Laboratory of OSU, so that we can determine the chemical, physical, and biological characteristics of our garden soil samples.

A few initial observations:

  • The most common complaint we heard from gardeners was a lack of space to properly rotate their crops. For example, nearly every site had tomatoes, but many did not have the space to avoid planting in the same ground as the previous season.
  • In the lab, our initial findings are that garden soils do not fit well with traditional soil testing methods. The very high content of organic matter and low incidence of rocks brings immediate problems to the lab testing process. The first step taken when a lab receives a soil sample is to pass the sample media through a sieve. The larger pieces are lightly ground and sieved again. The aim is to isolate the soil from non-soil matter in order to restrict laboratory tests to just the soil content itself. The organic matter is often shredded by this process, which can alter the results of the laboratory tests. The primary problem here is that the organic material in our sampled garden soils is mostly forest by-products: timber waste. This material is generally inert in the garden setting and not accessible to plants. When this organic matter is included in a soils analysis, the organic matter compounds are incorporated into the test results and  skew the report away from the actual state of the garden’s soil.

The next steps in understanding garden soils are in research and application. In research, soil testing should be reconsidered with gardens in mind. There may be alternative processing techniques to reduce variability between test results and garden soil content. Theoretical models may be able to produce a metric which could be used to adjust the results of a standard soil test to reflect garden conditions more accurately.

In application, greater precision of terminology would allow for a more refined view and management of garden systems. In particular, bed-types should be grouped by their method of establishment (i.e. was soil transported to the garden, or not), rather than the presence or lack of a garden border. Additionally, organic mulches and compost should be considered in finer detail. The source of the product is important to determine what chemical content is being applied to the soil top. The physical structure of the product is important to relay the extent to which the mulch content will likely be incorporated into the soil, itself.

We’re still actively working to process and test samples. We look forward to sharing more results, in the near future.

#OverlyHonestMethods is a hashtag that is trending on Twitter.  With this hashtag (which is simply an easy way to sort and find posts), scientists share the honest, ugly truth behind research.  Some examples:

  • “Data was not recorded on Sundays because I didn’t feel like coming in, and not recorded on this day because spiders”
  • “Got a random number by asking my mom for a 3 digit number b/c I was too lazy to use an actual random number generator”
  • “Only read the abstract of the paper cited because I don’t have any money to pay for the full paper.”

This past week, I felt like I was swimming in my own #OverlyHonestMethods research sorrow.

For starters, my particular project in the Garden Ecology Lab is to document pollinator biodiversity within 24 Portland-area gardens.  I LOVE this project.  It’s the project I’ve wanted to do since I arrived at OSU, in 2007.  But, there are two minor issues with this project.

Distribution of 24 gardens, being sampled for insect pollinators, 2017-2019.
Blue, yellow, and white pan traps ~ placed into home gardens to passively sample insect pollinators.
One of our beautiful, Portland-area garden study sites.

First, it takes me 6.5 hours to drive to all sites, in one day.  This is without doing any of the actual research.  I had originally planned to sample all gardens June 21-23 ~ but this plan was quickly scrapped when I realized that there would be no way that we could physically drive to all gardens, set traps, sample for 10 minutes, and then return to pick up all traps the next day.

Working dawn to dusk, we were only able to sample 13 of our 24 gardens, June 21-22.  So, we pushed our second set of garden samples (the remaining 11 gardens) to June 29-30.  Not ideal ~ but this is why we are replicating our study across three years, and will be sampling gardens once a month, for 3-5 months, within each year.

The other major issue with this study, this month, is that I am chairing the International Master Gardener Conference.

In less than one week, I’ll be welcoming nearly 1,300 Master Gardeners to Portland ~ for a conference that begins on July 9th (pre-tours), and ends on July 17th (post-tours).  That means that my crew and I have been stuffing 1,300 envelopes and bags.  We’ve printed and are putting 3,900 meal tickets into 1,300 badges.  I can’t over-emphasize how much work this conference has been (and continues to be!).  On the one hand, sampling pollinators just before this conference is the LAST thing I needed to do.  On the other hand . . . after spending too many late nights in a hot room, filled with boxes and boxes of conference envelopes, sampling garden pollinators is exactly what I needed.

Of course, when it rains, it pours.  Last week, we also had issues with our Native Plant study.  On Tuesday, I get a call from Aaron, who tells me that:  (1) someone trespassed onto our plots, and sprayed herbicide, and (2) someone pulled our plants up, by the roots, in one of our study blocks (replicates).

Native plant plot. This poor plant was ripped out of the ground, and/or had herbicide applied in the vicinity.
Native plant plot. This one was spared the wrath of wreckless weeders and herbicide applicators.

It’s a long (and enraging) story.  But, long story short ~ we lost all of the plants in our fifth study block.  We only have five blocks / replicates in this study (with 24 plant species ~ it is both expensive and expansive to include more blocks).  So, in one sad, sad day ~ we lost 20% of our replication, which will have negative impacts on our statistical power.

How will we cope?  We’ll regroup and replant.  We were already planning on repeating this study in 2018. Now, it seems like we’ll have to repeat in 2018 AND 2019 ~ which is a bummer . . . because this will extend Aaron’s time in grad school, will cost me 50% more to get him through grad school, and generally makes a sad, sad day for all.

But, the silver linings are: I love working with Aaron, and don’t mind supporting him for an extra year, and Aaron had already mentioned that he might want to stay on for a Ph.D., which would necessarily lengthen and/or expand the scope of his study.

C’est la research.  Perhaps in 2-3 years, we’ll all be able to have a good chuckle about this challenging month.


Gardens are unique and understudied systems that can have multi-faceted and positive impacts on environmental and public health.  But, key to realizing the potential, positive impact of gardens are the decisions that are made when planning, installing and maintaining garden beds and features.  These decisions are especially important, because gardeners manage and maintain a significant amount of land in the United States.  Take lawns, for example.  Studies suggest that lawns represented the single largest irrigated crop in the United States, and that there are more acres of lawn than the combined acreage of corn, alfalfa, soy, orchards and rice1.

Of course, lawns are just one component of a garden ~ perhaps the least interesting component, from an ecological point of view.  Gardens are special, because of their unique levels of plant abundance and diversity2, which in some cases can be considered ‘biodiversity hotspots’3.   In New York, my lab group documented the important role that plant abundance and diversity in urban and suburban gardens can play in conserving pollinator biodiversity4, 5, 6.  Recently, some of the top researchers in the country argued that conservation plans could better harness the positive environmental benefits of gardens and landscapes7.  But, before we can get there, we need to answer some basic questions.

This is where the Garden Ecology Lab comes in.  Our group works at the interface of ecology and sociology, to try and understand the benefits of gardens to the environment and to human health and well-being.  We want to document the biodiversity of plants, pollinators and other organisms in Oregon gardens, and analyze what factors constrain or promote garden biodiversity.  I’ve done this work in New York, but want to repeat these first steps in Oregon.  Ultimately, the goal is to understand how gardens ~ and the decisions we make in our gardens ~ either promotes or constrains ecosystem services, such as pollination, pest control, and more.

Our group is diverse, and includes students interested in ecology, horticultural therapy and urban soils.  Extension and outreach is embedded in all that we do, such that we plan to work closely with gardeners (as citizen scientists) to describe and understand garden biodiversity, and to communicate findings to broader audiences. We’ll be looking for garden study sites and cooperating gardeners in the coming months, and invite you to get to know us, just a bit more.


1Milesi, C., S. W. Running, C. D. Elvidge, J. B. Dietz, B. T. Tuttle, R. R. Nemani. 2005. Mapping and Modeling the Biogeochemical Cycling of Turf Grasses in the United States. Environmental Management 36:426–438.

2Thompson, K. K. C. Austin, R. M. Smith, P. H. Warren, P. G. Angold, K. J. Gaston. 2003. Urban domestic gardens (I): putting small-scale plant diversity in context. Journal of Vegetation Science 14:71-78.

3Gea Galluzzi, G., P. Eyzaguirre, V. Negri. 2010. Home gardens: neglected hotspots of agro-biodiversity and cultural diversity. Biodiversity and Conservation 19: 3635–3654.

4Fetridge, E., J. S. Ascher, G. A. Langellotto.  2008. The bee fauna of residential gardens in a suburb of New York City (Hymenoptera: Apoidea).  Annals of the Entomological Society of America 101:1067-1077.

5Matteson, K. C., G. A. Langellotto. 2010. Determinates of inner city butterfly and bee species richness. Urban Ecosystems 13:333-347.

6Matteson, K. C., J. S. Ascher and G. A. Langellotto. 2008. Richness and composition of the bee fauna of urban gardens in New York City (Hymenoptera: Apoidea). Annals of the Entomological Society of America 101:140-150.

7Hall, D. M., G. R. Camilo, R. K. Tonietto, J. Ollerton, K. Ahrne, M. Arduser, J. S. Ascher, K. C. R. Baldock, R. E. Fowler, G. W. Frankie, D. Goulson, B. Gunnarsson, M. E. Hanley, J. I. Jackson, G. Langellotto, D. Lowenstein, E. S. Minor, S. M. Philpott, S. G. Potts, M. H. Sirohi, E. M. Spevak, G. Stone, C. G. Threlfall.  2016. The city as a refuge for insect pollinators: conservation for the city. Conservation Biology. Online First.

We study gardens: the plants, insects, animals, people, decisions and management practices that either improve or degrade a garden’s ability to promote environmental and human health.

An underlying premise of our work is that gardens are important and understudied systems, that are key to building more sustainable, healthy and just communities.  

Garden Ecology Lab, Fall 2017: (left to right) Signe Danler, Gail Langellotto, Lucas Costner, Isabella Messer, Michael Nelson, Aaron Anderson

Our garden pollinator work is supported by a generous donation from Spike Wadsworth and Y. Sherry Sheng.