Images of simulated cities help artificial intelligence to understand real streetscapes

Researchers from Osaka University develop a way to train data-hungry models to accurately assess images of urban landscapes, without the need for real models or images of actual cities

Peer-Reviewed Publication

OSAKA UNIVERSITY

 

IMAGE: OVERVIEW OF THE PROPOSED METHOD. A 3D CITY MODEL IS AUTOMATICALLY GENERATED ACCORDING TO THE SPECIFIED PARAMETERS USING PROCEDURAL MODELING, AND ANNOTATION DATA AND TRAINING IMAGES ARE GENERATED FROM THE 3D CITY MODEL USING A GAME ENGINE AND AN IMAGE TRANSLATION TECHNIQUE, RESPECTIVELY. view more 

CREDIT: 2023 TAKUYA KIKUCHI ET AL., ADVANCED ENGINEERING INFORMATICS

Osaka, Japan – Recent advances in artificial intelligence and deep learning have revolutionized many industries, and might soon help recreate your neighborhood as well. Given images of a landscape, the analysis of deep-learning models can help urban landscapers visualize plans for redevelopment, thereby improving scenery or preventing costly mistakes.

To accomplish this, however, models must be able to correctly identify and categorize each element in a given image. This step, called instance segmentation, remains challenging for machines owing to a lack of suitable training data. Although it is relatively easy to collect images of a city, generating the ‘ground truth’, that is, the labels that tell the model if its segmentation is correct, involves painstakingly segmenting each image, often by hand.

Now, to address this problem, researchers at Osaka University have developed a way to train these data-hungry models using computer simulation. First, a realistic 3D city model is used to generate the segmentation ground truth. Then, an image-to-image model generates photorealistic images from the ground truth images. The result is a dataset of realistic images similar to those of an actual city, complete with precisely generated ground-truth labels that do not require manual segmentation.

“Synthetic data have been used in deep learning before,” says lead author Takuya Kikuchi. “But most landscape systems rely on 3D models of existing cities, which remain hard to build. We also simulate the city structure, but we do it in a way that still generates effective training data for models in the real world.”

After the 3D model of a realistic city is generated procedurally, segmentation images of the city are created with a game engine. Finally, a generative adversarial network, which is a neural network that uses game theory to learn how to generate realistic-looking images, is trained to convert images of shapes into images with realistic city textures This image-to-image model creates the corresponding street-view images.

“This removes the need for datasets of real buildings, which are not publicly available. Moreover, several individual objects can be separated, even if they overlap in the image,” explains corresponding author Tomohiro Fukuda. “But most importantly, this approach saves human effort, and the costs associated with that, while still generating good training data.”

To prove this, a segmentation model called a ‘mask region-based convolutional neural network’ was trained on the simulated data and another was trained on real data. The models performed similarly on instances of large, distinct buildings, even though the time to produce the dataset was reduced by 98%.

The researchers plan to see if improvements to the image-to-image model increase performance under more conditions. For now, this approach generates large amounts of data with an impressively low amount of effort. The researchers’ achievement will address current and upcoming shortages of training data, reduce costs associated with dataset preparation and help to usher in a new era of deep learning-assisted urban landscaping.

Data generated at each step using the developed framework. Left figure: parameter settings and other elements used in this study. Right figure: examples of data generated using the parameter settings.

Comparison of the detection accuracy of models trained on datasets generated by the proposed method and models trained on real images. It is possible to use the proposed method to achieve similar or superior results to the model trained on real images. The areas where the model trained on generated images obtained better results than the model trained on real images are indicated by the red dashed lines.

CREDIT

2023 Takuya Kikuchi et al., Advanced Engineering Informatics

The article, “Development of a synthetic dataset generation method for deep learning of real urban landscapes using a 3D model of a non-existing realistic city,” was published in Advanced Engineering Informatics at DOI: https://doi.org/10.1016/j.aei.2023.102154

About Osaka University

Osaka University was founded in 1931 as one of the seven imperial universities of Japan and is now one of Japan's leading comprehensive universities with a broad disciplinary spectrum. This strength is coupled with a singular drive for innovation that extends throughout the scientific process, from fundamental research to the creation of applied technology with positive economic impacts. Its commitment to innovation has been recognized in Japan and around the world, being named Japan's most innovative university in 2015 (Reuters 2015 Top 100) and one of the most innovative institutions in the world in 2017 (Innovative Universities and the Nature Index Innovation 2017). Now, Osaka University is leveraging its role as a Designated National University Corporation selected by the Ministry of Education, Culture, Sports, Science and Technology to contribute to innovation for human welfare, sustainable development of society, and social transformation.

Website: https://resou.osaka-u.ac.jp/e

---

JOURNAL

Advanced Engineering Informatics

DOI

10.1016/j.aei.2023.102154 

METHOD OF RESEARCH

Computational simulation/modeling

SUBJECT OF RESEARCH

Not applicable

ARTICLE TITLE

Development of a synthetic dataset generation method for deep learning of real urban landscapes using a 3D model of a non-existing realistic city

SOCIAL DEVELOPMENT GOALS 

New foresight report identifies urgent policy actions needed to put SDGs back on track

Ahead of the UN’s SDG Summit (18-19 September), ground-breaking analyses shows how by enacting five ‘extraordinary turnarounds’ SDGs implementation can be accelerated

Reports and Proceedings

THE CLUB OF ROME

Ahead of the UN’s SDG Summit (18-19 September), Earth4All, an international team of economists and scientists, and the Foundation for European Progressive Studies (FEPS), unveil groundbreaking research showing that policymakers can ensure the implementation of SDGs by 2050. The report ‘SDGs for All: Strategic scenarios’ equips policymakers with practical solutions designed to accelerate SDG implementation and to respond to the planetary emergency. It concludes that policymakers can step up the implementation of the SDGs by 2030 and beyond and achieve wellbeing for all. But this is only possible by enacting five ‘extraordinary turnarounds’ that break with current trends.

 

UN Secretary-General António Guterres has urged world leaders to come to the Summit “not with beautiful speeches, but with concrete actions, plans and commitments to accelerate the implementation of the 2030 Agenda and the Paris Agreement on Climate Change”.

 

The report responds to the UN Secretary-General’s call for more rigorous strategic analysis and foresight to support policymaking. It brings together analysis from leading scientists, economists and modelers, offering proven ways to implement the SDGs and dramatically improve the course of policymaking at major upcoming global meetings, including the UN SDG Summit, the UNFCCC COP28 and the UN Summit of the Future.

 

The ‘SDGs for All’ report is built around the future scenarios and the five extraordinary turnarounds first explored in Earth for All: A Survival Guide for Humanity, published in 2022.  The two scenarios are:

 

• ‘Too Little Too Late’ (TLTL) – a decision-making as usual approach resulting in deepening wealth inequality, growing social tensions and limited efforts to address climate and ecological risks. As a result, global temperature increases to 2.5°C by 2100 putting the stability of the earth system at risk. Wellbeing continues to dramatically decline globally, and it takes until 2100 to eradicate extreme poverty.   
• ‘The Giant Leap'  (GL) – an alternative, achievable path costing 2-4% GDP per annum empowers society to make ambitious decisions by implementing five extraordinary turnarounds simultaneously across poverty, inequality, empowerment, food and energy. As a result, temperatures would stabilise below 2°C, material consumption is reduced, extreme poverty is eradicated by 2050, social tensions fall dramatically, inequality is reduced, and wellbeing rises exponentially. If policymakers around the world embrace this ‘Giant Leap’, huge improvements for people and planet are possible.

 

“The Giant Leap scenario offers a way out of the current planetary emergency and a pathway for attaining the majority of SDGs by 2050.  However, this will require a radical transformation away from today’s extractive economy dominated by GDP growth to wellbeing economies that place a value on people, planet and prosperity”, comments Sandrine Dixson-Declève, co-president of The Club of Rome, co-author of Earth for All and co-lead of the Earth4All initiative.

 

The report models how progress towards achieving the SDGs would be in both scenarios.

• Poverty – In Too Little Too Late 20% of the global population live in poverty by 2050 – compared to 7% in the Giant Leap.
• Inequality – income gender parity is achieved in the Giant Leap, but severely worsens under Too Little Too Late with owners accounting for 75% of incomes.
• Emissions - CO2 emissions reach net zero under the Giant Leap in the 2040s. Compared to 2 Tonnes of CO2 per person in the Too Little Too Late scenario by 2050 – which would in result in 17 billion tonnes of carbon emitted globally per year.
• Public spending – An additional $8.8 trillion is spent globally on public services per year by 2050 in the Giant Leap - equivalent to $6,000 per person per year. Whereas the spending is $4,800 per person per year in Too Little Too Late in 2050.  

A key red flag in the report is that by 2050, the level of global warming is too high in both scenarios, with devastating consequences in every corner of the globe. However, global warming eventually plateaus below 2°C under the Giant Leap and gives a chance for humanity to thrive again. The reality of overshooting above 1.5°C in both scenarios gives serious cause for concern regarding the lack of planetary emergency plans currently in place to address climate change and the resulting increase in shocks and stresses.

The modelling has also shown that gender equality is woefully off track; at current rates it would take 257 years to reduce the overall gender gap.

“This must serve as a wake-up call for society”, said Maria João Rodrigues, President of the Foundation for European Progressive Studies. She continued: “The upcoming UN SDG Summit will mark the half-way point to the 2030 deadline for achieving the SDGs, and that is why we asked Earth4All to analyse the progress made and what we need to do to get back on track with their unique system dynamics model. Action can no longer be avoided, it must be done in a systemic way and adopted on both international and national levels.”

 

Sandrine Dixson-Declève, co-president of The Club of Rome, co-author of Earth for All and co-lead of the Earth4All initiative, concluded: “Our economic and financial systems are broken and we are reaching dangerous levels of inequality. The Too Little Too Late scenario when applied to the SDG’s condemns future generations to a dangerously destabilised planet. The climate system is likely to cross multiple tipping points and social tensions are likely to increase. This must be avoided at all costs. By contrast our Giant Leap scenario significantly reduces this risk, thereby contributing to greater resilience and the possible emergence from emergency.

 

“Although the achievement of the SDGs especially SDG13 and SDG5 is in grave danger, failing to meet them is not an option. The recipe for success: SDG implementation must be coupled with emergency planning to prepare for future shocks and stresses and leaders must take up the call for radical transformation.”

 

To achieve the pace and scale of change required for meaningful SDG progress by 2050, the report identifies several urgent policy levers which need to be implemented simultaneously:

• Significant new investments are essential – they must be accompanied by massive increases in public spending, along with higher taxation of extremely wealthy individuals and private corporations.
• Fundamental reform of the International Monetary Fund’s process for allocating Special Drawing Rights (SDRs) – to support countries that need them most. And dealing with the sovereign debt overhang is also essential to give low-income countries more fiscal space and relief.
• Governments must quickly reverse the steady erosion of workers’ rights and implement new safety nets such as a universal basic dividend.
• Governments must massively scale up investment in women and girls to reverse the huge declines in terms of income, safety, education and health.
• Global food systems must be radically transformed, starting with the repurposing of agricultural subsidies towards supporting low-carbon and regenerative agriculture practices to improve food production efficiency and sustainability. Food supply chains must shift towards localised food production, and farmworker rights must be prioritised and protected.
• Global energy systems must shift from inefficient fossil energy systems to a clean and optimised energy system that reduces consumption in high-income countries, ensures electricity access to all, and enhances greater efficiencies across the global energy system.

 

Earth4All and FEPS will present the findings of the SDGs for All report to the UN and stakeholders prior to and during the SDG Summit.

---

METHOD OF RESEARCH

Computational simulation/modeling

ARTICLE TITLE

SDGs for All: Strategic scenarios

ARTICLE PUBLICATION DATE

14-Sep-2023

DECRIMINALIZE DRUGS

Overdose deaths from fentanyl laced stimulants have risen 50-fold since 2010

The trend marks the fourth wave in the US overdose crisis, which began with prescription opioid deaths in the early 2000s and has since continued with other drugs

Peer-Reviewed Publication

UNIVERSITY OF CALIFORNIA - LOS ANGELES HEALTH SCIENCES

 

IMAGE: A SIMPLIFIED SCHEMA OF THE FOUR WAVES OF THE US OVERDOSE MORTALITY CRISIS. WAVES 1 AND 2 INCLUDE DEATHS INVOLVING COMMONLY PRESCRIBED OPIOIDS, AND HEROIN, RESPECTIVELY, BUT EXCLUDING FENTANYL CO-INVOLVED DEATHS. WAVE 3 AND WAVE 4 SHOW FENTANYL DEATHS NOT INVOLVING, AND INVOLVING, STIMULANTS RESPECTIVELY AS DISTINCT TRENDS. DATA FROM CDC WONDER. view more 

CREDIT: FRIEDMAN AND SHOVER, 2023, DOI: 10.1111/ADD.16318

New UCLA-led research has found that the proportion of US overdose deaths involving both fentanyl and stimulants has increased more than 50-fold since 2010, from 0.6% (235 deaths) in 2010 to 32.3% (34,429 deaths) in 2021. 

 

By 2021, stimulants such as cocaine and methamphetamine had become the most common drug class found in fentanyl-involved overdoses in every US state.  This rise in fentanyl/stimulant fatalities constitutes the ‘fourth wave’ in the US’s long-running opioid overdose crisis –the death toll of which continues to rise precipitously. 

 

“We’re now seeing that the use of fentanyl together with stimulants is rapidly becoming the dominant force in the US overdose crisis,” said lead author Joseph Friedman, an addition researcher at the David Geffen School of Medicine at UCLA. “Fentanyl has ushered in a polysubstance overdose crisis, meaning that people are mixing fentanyl with other drugs, like stimulants, but also countless other synthetic substances. This poses many health risks and new challenges for healthcare providers. We have data and medical expertise about treating opioid use disorders, but comparatively little experience with the combination of opioids and stimulants together, or opioids mixed with other drugs. This makes it hard to stabilize people medically who are withdrawing from polysubstance use.”

The findings are published in the peer-reviewed journal Addiction.

The analysis illustrates how the US opioid crisis began with an increase in deaths from prescription opioids (wave 1) in the early 2000s and heroin (wave 2) in 2010.  Around 2013, an increase in fentanyl overdoses signalled the third wave.  The fourth wave – fentanyl overdoses with stimulants – began in 2015 and continues to grow.

Further complicating matters is that people consuming multiple substances may also be at increased risk of overdose, and many substances being mixed with fentanyl are not responsive to naloxone, the antidote to an opioid overdose.  

The authors also found that fentanyl/stimulant overdose deaths disproportionately affect racial/ethnic minority communities in the US, including Black and African American people and Native American people. For instance, in 2021, the prevalence of stimulant involvement in fentanyl overdose deaths was 73% among 65 to 74-year-old Non-Hispanic Black or African American women living in the western US and 69% among 55 to 65-year-old Black or African American men living in the same area.  The rate among the general US population in 2021 was 49%.

There are also geographical patterns to fentanyl/stimulant use.  In the northeast US, fentanyl tends to be combined with cocaine; in the southern and western US, it appears most commonly with methamphetamine. 

 

“We suspect this pattern reflects the rising availability of, and preference for, low-cost, high-purity methamphetamine throughout the US, and the fact that the Northeast has a well-entrenched pattern of illicit cocaine use that has so far resisted the complete takeover by methamphetamine seen elsewhere in the country,” Friedman said.

 

The study was funded by the UCLA Medical Scientist Training Program (National Institute of General Medical Sciences training grant GM008042) and the National Institute on Drug Abuse at the National Institutes of Health (K01DA050771). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

---

JOURNAL

Addiction

DOI

10.1111/add.16318 

METHOD OF RESEARCH

Data/statistical analysis

SUBJECT OF RESEARCH

People

ARTICLE TITLE

Charting the fourth wave: Geographic, temporal, race/ethnicity and demographic trends in polysubstance fentanyl overdose deaths in the United States, 2010–2021

ARTICLE PUBLICATION DATE

14-Sep-2023

The fourth wave of the US overdose crisis: fentanyl and stimulants

Peer-Reviewed Publication

SOCIETY FOR THE STUDY OF ADDICTION

 

IMAGE: A SIMPLIFIED SCHEMA OF THE FOUR WAVES OF THE US OVERDOSE MORTALITY CRISIS. WAVES 1 AND 2 INCLUDE DEATHS INVOLVING COMMONLY PRESCRIBED OPIOIDS, AND HEROIN, RESPECTIVELY, BUT EXCLUDING FENTANYL CO-INVOLVED DEATHS. WAVE 3 AND WAVE 4 SHOW FENTANYL DEATHS NOT INVOLVING, AND INVOLVING, STIMULANTS RESPECTIVELY AS DISTINCT TRENDS. DATA FROM CDC WONDER. view more 

CREDIT: FRIEDMAN AND SHOVER, 2023, DOI: 10.1111/ADD.16318.

New research published in the scientific journal Addiction has found that the proportion of US overdose deaths involving both fentanyl and stimulants has increased more than 50-fold since 2010, from 0.6% (235 deaths) in 2010 to 32.3% (34,429 deaths) in 2021.  By 2021, stimulants (such as cocaine and methamphetamine) had become the most common drug class found in fentanyl-involved overdoses in every US state.  This rise in fentanyl/stimulant fatalities constitutes the ‘fourth wave’ in the US’s long-running opioid overdose crisis –the death toll of which, continues to rise precipitously. 

“We’re now seeing that the use of fentanyl together with stimulants is rapidly becoming the dominant force in the US overdose crisis,” says lead author Dr Joseph Friedman, of the University of California, Los Angeles.  “Fentanyl has ushered in a polysubstance overdose crisis, meaning that people are mixing fentanyl with other drugs, like stimulants, but also countless other synthetic substances. This poses many health risks and new challenges for healthcare providers. We have data and medical expertise about treating opioid use disorders, but comparatively little experience with the combination of opioids and stimulants together, or opioids mixed with other drugs. This makes it hard to stabilize people medically who are withdrawing from polysubstance use.”

People consuming multiple substances may also be at increased risk of overdose, and many substances being mixed with fentanyl are not responsive to naloxone, the antidote to an opioid overdose.  

The authors also found that fentanyl/stimulant overdose deaths disproportionately affect racial/ethnic minority communities in the US, including Black and African American people and Native American people. For instance, in 2021, the prevalence of stimulant involvement in fentanyl overdose deaths was 73% among 65 to 74-year-old Non-Hispanic Black or African American women living in the western US and 69% among 55 to 65-year-old Black or African American men living in the same area.  The rate among the general US population in 2021 was 49%.

There are also geographical patterns to fentanyl/stimulant use.  In the northeast US, fentanyl tends to be combined with cocaine; in the southern and western US, it appears most commonly with methamphetamine.  Friedman says, “We suspect this pattern reflects the rising availability of, and preference for, low-cost, high-purity methamphetamine throughout the US, and the fact that the Northeast has a well-entrenched pattern of illicit cocaine use that has so far resisted the complete takeover by methamphetamine seen elsewhere in the country.”

The analysis illustrates how the US opioid crisis began with an increase in deaths from prescription opioids (wave 1) in the early 2000s and heroin (wave 2) in 2010.  Around 2013, an increase in fentanyl overdoses signalled the third wave.  The fourth wave – fentanyl overdoses with stimulants – began in 2015 and continues to grow.

-- Ends –

Research reported in this press release was supported in part by the National Institute on Drug Abuse of the National Institutes of Health under award number K01DA050771. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

This paper is Open Access on the Wiley Online Library: https://onlinelibrary.wiley.com/doi/10.1111/add.16318  or you may request a copy from Jean O’Reilly, Editorial Manager, Addiction, jean@addictionjournal.org.

To speak with authors Dr. Joseph Friedman and Dr. Chelsea Shover, please contact them at the University of California, Los Angeles by email (joseph.robert.friedman@gmail.com, CLShover@mednet.ucla.edu).

An Addiction Audio podcast with Dr. Friedman is available here (link will be live from 14 September):
https://shows.acast.com/addiction-audio/episodes/poly-drug-use-and-us-opioids-with-joseph-friedman

Full citation for article: Friedman J and Shover C. (2023) Charting the Fourth Wave: Geographic, Temporal, Race/Ethnicity, and Demographic Trends in Polysubstance Fentanyl Overdose Deaths in the United States, 2010-2021. Addiction. DOI: 10.1111/add.16318

Funding: Dr. Friedman received support from the UCLA Medical Scientist Training Program (National Institute of General Medical Sciences training grant GM008042). Dr. Shover was funded by the National Institute on Drug Abuse (K01DA050771).

Declaration of interests:  None.

Addiction is a monthly international scientific journal publishing peer-reviewed research reports on alcohol, substances, tobacco, and gambling as well as editorials and other debate pieces. Owned by the Society for the Study of Addiction, it has been in continuous publication since 1884.

---

JOURNAL

Addiction

DOI

10.1111/add.16318 

METHOD OF RESEARCH

Observational study

SUBJECT OF RESEARCH

People

ARTICLE TITLE

Charting the Fourth Wave: Geographic, Temporal, Race/Ethnicity, and Demographic Trends in Polysubstance Fentanyl Overdose Deaths in the United States, 2010-2021

ARTICLE PUBLICATION DATE

14-Sep-2023

 

Most Ohio students who earn manufacturing-related credentials work in other industries

Findings hold implications as policymakers push to grow U.S. manufacturing

Reports and Proceedings

RAND CORPORATION

Most students who complete manufacturing-related credentials in Ohio do not end up employed in manufacturing in the state, highlighting a challenge that faces policymakers as they push to create more U.S. manufacturing jobs, according to a new RAND Corporation report.

 

Among those who earned a manufacturing-related credential from a public postsecondary institute in Ohio from 2006 to 2019, fewer than 40% worked in manufacturing in the state within one year after completing their education.

 

Wages are not a likely contributor to the trend. Students who enter other fields after completing a manufacturing-related credential earn less than their peers who pursued a manufacturing career -- a trend that continues for at least five additional years.

 

“These findings suggest that there is a much larger supply of highly skilled workers with manufacturing-related expertise than is currently being used by the manufacturing industry,” said Christine Mulhern, an author of the study and an economist at RAND, a nonprofit research organization.

 

The U.S. manufacturing industry is experiencing a resurgence and faces a growing need for skilled workers. Recent reports project that demand for skilled manufacturing workers will outpace supply in coming years.

 

RAND researchers focused on Ohio because the state has one of the nation’s largest manufacturing industries, and as such may be instructive for understanding the challenges and opportunities of expanding the manufacturing sector.

 

To identify promising strategies to expand the supply of skilled manufacturing workers to meet employers’ growing demands, RAND researchers examined Ohio’s postsecondary education system and manufacturing employment in Ohio.

 

The study analyzed administrative information from the Ohio Longitudinal Data Archive to describe education and employment patterns in Ohio between 2006 and 2019. The information covers public postsecondary institutions in Ohio, including community colleges, four-year colleges and Ohio Technical Centers.

 

The number of students pursuing manufacturing-related education in Ohio’s public postsecondary institutions has increased in recent years. Among individuals who complete a manufacturing-related credential in Ohio, more than 80% are white, and more than 85% are male.

 

However, some of the growth in enrollments in recent years has been due to an increase in female and Asian students in four-year manufacturing-related programs. Black students disproportionately enroll in shorter-term programs.

 

“This suggests that expanding the diversity of students in manufacturing-related programs may be important for expanding the diversity of the manufacturing workforce,” said Lisa Abraham, a RAND economist and co-author of the study.

 

Researchers found that the drop-off between earning a manufacturing-related credential and being employed in the sector was larger for women and individuals from underrepresented minority backgrounds.

 

Researchers also examined retention within the industry among the 2013 population of Ohio’s full-time manufacturing workers. 77% of these workers were still employed in manufacturing in 2016 and 63% were still employed in manufacturing in 2019. The most common path for those who exited manufacturing was leaving Ohio’s full-time workforce, although up to 15% left for a full-time job in another industry in the state.

 

In addition, researchers examined pathways into the manufacturing workforce among recent entrants to Ohio’s workforce. About 64% of these workers entered manufacturing from a job in another Ohio industry, and 11% entered from an Ohio postsecondary institution. The report finds that drawing workers from other industries could be a promising avenue for expanding the pipeline of manufacturing workers.

 

Researchers say that more research is needed to understand the education-to-employment pipeline for the manufacturing industry and how it compares with the pipelines

for other industries, both within and outside Ohio. Interviews with students, colleges and employers, as well as surveys of these groups may be helpful to understand the mechanisms underlying these patterns.

 

The report also suggests that employers may be able to curtail attrition by increasing investments in upskilling workers to provide more opportunity and attract new workers to manufacturing by better promoting available jobs and the benefits of working in the industry.

 

Support for the study was provided by the Lumina Foundation.

 

The report, “Strengthening the Manufacturing Workforce in Ohio,” is available at www.rand.org. Lucas Greer also co-authored the report.

 

The RAND Education and Labor division is dedicated to improving education and expanding economic opportunities for all through research and analysis.

---

METHOD OF RESEARCH

Data/statistical analysis

SUBJECT OF RESEARCH

People

ARTICLE TITLE

Strengthening the Manufacturing Workforce in Ohio

ARTICLE PUBLICATION DATE

14-Sep-2023

 

Pollination by more than one bee species improves cherry harvest

Peer-Reviewed Publication

UNIVERSITY OF GOTHENBURG

 

IMAGE: JULIA OSTERMAN, RESEARCHER AT THE UNIVERSITY OF GOTHENBURG. view more 

CREDIT: VERONIKA CONRADY

To obtain the biggest cherry harvest, trees should be pollinated by both honey bees and mason bees. A new study led by a researcher at the University of Gothenburg shows yet another benefit of biodiversity.

Like many other fruit trees, most sweet cherry cultivars depend on cross-pollination to produce their fruit. This means that there need to be several different cultivars of sweet cherry trees in an orchard for the bees to transport pollen from one to another.

“Sweet cherry trees are usually planted in alternate rows of different cultivars. In some cases, you can put different cultivars in the same row, but this can make the harvesting logistics tricky. In other words, the bees have to fly from one row of trees to the next to ensure that the trees set fruit,” says Julia Osterman, a biologist at the University of Gothenburg and lead author of the study published in the scientific journal Ecology and Evolution.   

Two bee species produce a synergy effect

Working with German researcher colleagues at Martin Luther University Halle-Wittenberg, Julia Osterman found that if trees were pollinated by more than one bee species, they produced more cherries. The researchers observed bees in a total of 17 cherry orchards in eastern Germany. Some growers used honey bees in hives as pollinators, while others used wild mason bees. Some orchards used both species to different extents. The researchers noticed a synergy effect in those orchards in which both species of bee were present.

“It had an impact on the sweet cherry fruit set. The orchards with honey bees and lots of mason bees could have cherries on up to 70 per cent of the blossom. In orchards with only honey bees or only mason bees as pollinators the rate could be as low as 20 per cent,” says Julia Osterman.

Many growers were already using two species of bees, often as a back-up if the weather was too cold for the honey bees when the cherry trees were in bloom, as cherries flower early. Honey bees only become active once the temperature is above 12°C, but mason bees can cope with lower temperatures. The sharp increase in fruiting occurred when both species were active. The researchers are now speculating on the reasons for this.

Bamboo sticks as nest material

“One theory is that the presence of mason bees affects the foraging behavior of honey bees,” says Julia Osterman. “This disturbs them and so they change rows more often, resulting in more cross-pollination. But all we know at the moment is that interaction between the bees produces a synergy effect.”

Of course, this is valuable data for cherry growers who can attract wild mason bees to their orchards by providing good nest material.

“Mason bees are solitary and don’t make honey in honeycombs like honey bees,” Julia Osterman explains. “They are more focused on collecting pollen to feed their offspring. They like to crawl into tube-shaped spaces where they can lay their eggs. Fruit growers can encourage mason bees to nest in their orchards by placing bamboo or wood with holes drilled in it at the site. However, it only seems to work up to a certain limit, after which you won’t attract any more mason bees, no matter how much nest material you bring in.”

Similar results have been observed in almond orchards and Julia Osterman’s next step will be to investigate whether this synergy effect applies to other fruit trees, as well as trying to determine exactly how the two bee species are affecting each other.

Facts: The study was conducted in Saxony-Anhalt and Thuringia in Germany in spring 2020 with researchers from Martin Luther University Halle-Wittenberg and a research institute in Erfurt and is published in Ecology and Evolution. To the study: Mason bees and honey bees synergistically enhance fruit set in sweet cherry orchards

The combination of honey bees and mason bees as pollinators yields a much bigger harvest than with only one species.

Installing bamboo canes or wood with drilled holes attracts mason bees to orchards.

CREDIT

Julia Osterman

A mason bee collecting pollen in a cherry orchard.

CREDIT

Maxime Eeraerts

---

JOURNAL

Ecology and Evolution

DOI

10.1002/ece3.10289 

METHOD OF RESEARCH

Observational study

SUBJECT OF RESEARCH

Animals

ARTICLE TITLE

Mason bees and honey bees synergistically enhance fruit set in sweet cherry orchards