Tomato Production At Risk Due to Decrease in Buzz Pollinators

Originally posted at the University of Tennessee website.

Tomatoes are the heart of many backyard gardens. Tomato crops are also an important economic revenue in North America. The vegetable we all know as the “T” of a great summer BLT, however, may be in jeopardy due to a decline in its pollinator species because of climate change.

According to a study published in Ecological Applications, tomato production is at risk in the Eastern United States due to climate-induced decrease in the richness of buzz pollinators. Approximately 70 percent of the world’s crops depend on insect pollination for production. Climate change is already affecting the abundance of pollinators, but researchers with the National Institute for Mathematical and Biological Synthesis (NIMBioS) wanted to know how climate change impacts the pollination of specific crops.

Wild bees are an essential component of agriculture and especially important for crops whose flowers use buzz pollination to spread their pollen. Also known as sonication, buzz pollination is a process in which the pollinators attach to the anther cone of the flower and contract their indirect flight muscles to produce vibrations. Wild buzz pollinators can increase pollen load, which increases the fruit production and quality of the crop.

“Buzz pollinators provide pollination services for select groups of plants, including several vegetable species,” said Greg Wiggins, education and outreach coordinator for NIMBioS. “Our focus on a specific group of pollinators allows us to highlight potential impacts the changes in distributions of these pollinators may have on agricultural production.”

Climate change already affects many species by forcing them to move in order to keep within suitable climatic conditions. In this study, researchers looked at how potential future changes in climate will impact the distribution of buzz pollinators and what effect that will have on the species richness of tomato crops in North America.

“Studying distributional shifts is especially important for species that pollinate the food we eat,” said Luis Carrasco Tornero, lead author and postdoctoral researcher with NIMBioS. “Our study exemplifies the potential impact of climate change on the production of tomato crops due to the distribution changes of pollinator wild bees. Our findings could serve as an example of individual crop impacts due to lack of overlap between crop plants and their pollinator, which might help raise awareness about the direct impacts of climate change to our lives in the next few decades.”

Researchers applied ecological models and scenarios of future climatic conditions to estimate the potential geographic distribution of the buzz pollinators, which are rarely used to investigate possible effects of climate change on plant-animal interactions and food production.

“We show that the magnitude of change projected by different scenarios varies across species, thus this variation needs to be taken into account when working at the community level, with multiple species,” said Mona Papeş, assistant professor in the UT Department of Ecology and Evolutionary Biology. “Our models estimated a decrease in the number of pollinator species in several regions across the US where field tomato crops are grown on a large scale.”

Beyond tomatoes, ecological niche models can be used to assess potential impacts of climate change on other crops that depend on buzz pollinators, such as peppers, eggplants, or potatoes. It also has applications outside agricultural systems.

“Our work further supports the growing body of evidence showing the potential detrimental impacts climate change can have on natural systems,” Wiggins said.

Researchers will continue their work in pollinator distribution and plan focus their next study on what factors contribute to the ability pollinators have to shift their distribution in the face of climate changes.

“We hope our results from the next project can help conservationists find solutions to manage pollinator populations and avoid local extinctions of some pollinator species,” Carrasco said.

The research project was supported by a NIMBioS Summer Research Experiences (SRE) program that provides undergraduate students from across the US the opportunity to spend eight weeks during the summer working in small teams on research that incorporates mathematics to answer questions about biological systems. Undergraduate students from three other universities participated in the research and are co-authors on the paper.

Lasers, forests, and birds


Understanding why some regions are higher in number of species than others is fundamental to protect our planet's biodiversity. One factor that seems to determine the number of species present in an area is environmental heterogeneity. A heterogeneous environment will have a higher number of habitats, conditions (such as temperature or humidity) or a higher number of resources. Habitat structure is another component of environmental heterogeneity. An area with a high variety of structures, such as vegetation forms or inanimate objects, will provide animal species a higher number of food resources, shelter, or even a higher number of microclimates. However, measuring vegetation structure, the main component of habitat structure, is not an easy task.

Lidar (Light detection and ranging), a remote sensing technology that uses pulses of laser to measure the height and density of the vegetation, is helping researchers to measure vegetation structure from the sky (and soon from the International Spatial Station!). This technology allows us to make vegetaion measurements for wide extents and avoids spending huge amounts of time and money in field surveys. Recently, I leaded a study, together with researchers from NIMBioS and the University of Tennessee, where we created 3D models of forests using Lidar, and studied how North American forests' structure determine the diversity of birds. We created different Lidar-derived structural metrics to study how the structure of a forest patch varies horizontally and vertically. We found that high heterogeneity of the horizontal component of the forest structure was especially important for enhancing diversity of birds. Forest patches with very high vertical heterogeneity presented less species than patches with intermediate levels of vertical heterogeneity, contrary to previous studies.

This type of study is very important for conservation, as we can identify the characteristics of those habitats that support high levels of biodiversity, and therefore prioritize our protection efforts. We could, for example, promote the maintenance of horizontal heterogeneity in forests to enhance their bird diversity. However, these effects might depend on the forest type, so our future work will look at how these relationships are affected by the typology of the forest, and how these effects can help us explain biodiversity patterns at continental scales.

Radar satellite images and the "double-bounce" effect


During the last few weeks I've been playing around with radar images from satellite sensors, in order to figure out what are the best datasets that we can use for land cover mapping in the UK. The use of radar images is promising and we might be able to increase significantly the certainty of the next generation of UK Land Cover Maps. 

That been said, analyzing and interpreting radar images is challenging and completely different from optical images. Currently I'm mainly working with Sentinel-1 satellite, which carries a Synthetic Aperture Radar (SAR) sensor. A SAR sensor is what is called an active sensor. In opposition to a passive sensor, which detects and processes the light reflected from the Sun (optical sensors such as Landsat), a SAR sensor emits a microwave beam and detects its reflection after interacting with the Earth surface. 

The intensity of the beam that returns to the sensor will be a proportion of the emitted signal, backscattered to the sensor from a certain point in the Earth surface. This intensity is called backscatter coefficient. The coefficient will depend on the type of surface and also in things like its shape or roughness. Without entering into the "polarization" of the signal, we could simplify things saying that the roughest the surface is, the highest the signal intensity will be. In other words, we will see brighter pixels for rougher surfaces. 

As an example, a calmed body of water will not backscatter much of the microwave beam (Fig. 1a). However, a more complex surface such as an heterogeneously vegetated land will backscatter some of the signal (Fig. 1b).  Another curious case of reflection occurs when the beam encounters vertical structures. In this case, it can happen that the beam reflects in a flat horizontal surface and then reflects again in the flat vertical surface, returning to the sensor with a high intensity (Fig. 1c). This is called the double-bounce or the corner-reflection effect. The double-bounce effect commonly occurs with ships in the sea with calmed waters, or in tall buildings surrounded by flat concrete areas.  

A few days ago I was exploring some Sentinel-1 images from the north-west of England (where I live), and I realized that this effect was very noticeable with some highly organized structures in the middle of the sea. These were the offshore wind farms of Morecambe bay. The Waley Wind Farm is one of the largest of the world and it is very spectacular to see it from space. It's also a great example of how radar satellite images can show us a different view of the Earth surface, difficult to sea with optical images. See how different it looks the Waley Wind farm from radar and optical images in the next image.

I'm enjoying exploring the capabilities of radar images. I will possibly write more entries with some curiosities about radar sensors in the future. I would like also to study how radar images can be used for automatic detection of bird colonies, but that will have to wait for now!

Dissecting avian behavior from space


Remote sensing has been helping scientists to collect and monitor ecological information since the launch of the first earth-observation satellites. The release of free Landsat satellite images, with continuous coverage from the late 70's, the improvement of Lidar and Radar sensors, and the recent widespread use of unmanned aerial vehicles (UAV's, such as drones), has radically changed how animal ecologists study animal movement, species distributions or environmental management. 

In the July's number of the Sensed Newsletter (published by the Remote Sensing and Photogrammetry Society of UK, RSPSoc), I discuss about how these advances are helping ornithologist to track and count birds, as well as to study their behaviour from space.  I was very pleased they ask me to collaborate with this magazine. Have a look at the article!

UK Land Cover Map 2015


One of the main reasons why I moved to the UK and started working at the CEH was the opportunity of being involved in the creation of the UK Land Cover Map 2015 (LCM2015). This is a nation-wide habitat mapping project that will have a big impact in future UK research and it will be used by government departments, environmental management agencies, charities and many more. After spending all 2016 and part of 2017 working on it, the LCM2015 was finally released last April.  


LCM2015 is derived from satellite images (mainly Landsat-8) and provides land-cover information of the whole UK. The main product is a 22 land-cover class vector map, based on a parcel-based spatial framework, from which a 25m-pixel raster product is derived. However, many other secondary products at different scales can be obtained. 

One of my favourite features about the LCM2015 is that the algorithm used to predict the land cover type (based on satellite spectral data and other ancillary information) is a Random Forest. As you probably know, this is a classificatory algorithm that I have been using for a long time in my research on animal distribution modelling. It is also becoming very popular to perform supervised classifications of satellite images, and we used it for the LCM2015, obtaining great results. Using this algorithm allowed us to provide a "probability map", giving the user an estimate of how accurate is the assignment of each parcel or pixel to a land-cover type. I think this is going to be a key feature of future LCM's, as it will allow us to add the uncertainty of the habitat identification, when using remote sensing derived data, into our species distribution models. 

The production of the LCM2015 meant many months of hard work, but it has been a great experience for me. Now, I can't wait to start exploring its potential for ecological research, especially for bird distribution research in the UK (stay tuned!). Here I link a short video of the LCM2015, created by the CEH (I did that 3D model!): 




Business Impact School


Last year I attended the Business Impact School (BIS), held at the impressive Willis Tower in London. The BIS is organized by the NERC Valuing Nature Programme and every year they fund 25 early career researchers to attend a three day workshop with speakers from the business and scientific world. These workshops aim to create a scientific community with deep understanding of how biodiversity conservation, ecosystem services and natural capital can translate into industry and private sector decision-making. 

If you are an early career researcher wanting to improve your knowledge on linking natural capital with the business world, this workshop is perfect (try to apply for next one!). One of the most interesting topics addressed last year was how to effectively communicate science to the business world. The organization was great and I met a lot of brilliant young researchers there, all from very different scientific backgrounds.

They made some videos from the event. Here I post one that I'm speaking about my research and the workshop itself.


RSPSoc Annual Conference and... prize!


Last month I spent a few days in Nottingham to attend to the annual conference of the Remote Sensing and Photogrammetry Society (RSPSoc). It was the first time for me to participate in a "pure" remote sensing (RS) congress, as my research has always been focused on the ecological and conservation applications of RS and GIS. It turned out to be a great place to present my new research on habitat mapping using Sentinel-2 satellite images, and to interact with researchers of the RS community. I also brought a poster on herons and egrets colony distribution in order to show how we can use satellite images and unmanned aerial vehicles (UAV) for ecological research. My line manager, Dan Morton, was also there presenting our progress with the Land Cover Map 2015. 

Bringing to the table all these topics was a little bit overwhelming as it was difficult to decide which talk to attend at every session (there where several talks at the same time) in order to make the congress the most useful for me. However, I had the opportunity to listen to very interesting talks about land cover mapping, and had great feedback about my work with Sentinel-2 from top researchers of the RS community. 

I was particularly impressed by the new studies using high resolution images to improve the spatial resolution of historical Landsat images. These methodologies gave me a lot of ideas to work on the detection of bird colonies from space. Also, the amount of works using Sentinel-1 images (radar) for detecting change, natural disasters and all kind of ecological applications made me think a lot about the potential of these satellites. It seems they are opening a very interesting path... Should I give this area a chance? 

It was a great conference and I received excellent advice from very experience researchers, in relation to the Sentinel-2 work and to my heron research. What a pity I didn't attend the final ceremony, as I received the Best Poster Award for the herons and egret research! Thanks to the RSPSoc for a great organization and for the prize!



Symposium - Space: the final frontier for biodiversity monitoring?


A couple of weeks ago, I attended the symposium: "Space: the final frontier for biodiversity monitoring", held by the Zoological Society of London, where top scientist from diverse backgrounds discussed the applications of remote sensing techniques for global biodiversity monitoring.

I presented two posters. One of them showing practical examples of the Land Cover Map 2007 (created by CEH) for biodiversity studies, and the second one describing part of my PhD research on using historical satellite images to study changes over time in habitat selection of avian species. 

You can read the description of the event on the Northwest East Observation Network (NEON) website, as two of the members of the network (Christopher Marston and myself) participated in the symposium.

New Job at the Centre for Ecology and Hydrology


Last January, I started working as a research associate at NERC Centre for Ecology and Hydrology (CEH) located in Lancaster, UK. I will be applying remote sensing and GIS techniques for ecology research and biodiversity conservation. Until now, I have been immersed in the production of the next generation of Land Cover Map for the UK (LCM2015), working with the preprocessing and classification of satellite images to produce accurate habitat mapping. I hope a can show you the results in the near future! 

Climate Change and Animal Populations Workshop


This month I participated in a Workshop on the effects of climate change on animal populations, held in the beautiful city of Erice, Sicily. I had the opportunity to spend some days with my PhD supervisor and my collaborators from Japan and Italy. It was a wonderful place to present our first results on the effects of global change on waterbird populations and habitat selection strategies. 

I had also the opportunity to hear talks and speak with world-top animal ecology researchers. I was specially impressed by the lectures of Professors Stan Boutin from Canada, Brian Hutley from the UK, or Raimundo Real from Spain among many others.

Big questions arose, and great discussions took place after each talk. One of the bring-back messages of the workshop was the difficulty to find direct evidence of the effects of climate change on animal populations. Some examples of adaptation of mammals and birds to climate change were shown, and a lot of discussions about how to identify micro-evolution in response to climate change emerged. Researchers are quite skeptical about climate change affecting drastically animal populations around the globe. Habitat alteration is still a major effect, and adaptation is helping to survive to temperature and other variables changes. However, adaptations in many cases should show some limits in the future. Maybe soon. What will happen then? When will this happen? 

I learnt a lot from those discussions about adaptation/micro-evolution, and how difficult is to really show evidence of climate change driving the decrease of animal populations. I missed, however, a little bit of discussion about other animal groups. Birds and mammals are, in some cases, being quite successful to adapt to climate changes until now. But, is this true for other vertebrates? What about amphibians, the most threatened group? Also, how are invertebrates being affected? Acidification and the rising temperature of the sea might be also affecting enormously marine animal populations.  

A lot of research is still needed. The main lesson I learned is to keep fighting against climate change while always be skeptical!

This little guy (black redstart) would wake me up everyday really early during my stay in Erice.