A year in an urban forest: Dairy Bush GigaPan 2009-2010
A year in an urban forest: Dairy Bush GigaPan 2009-2010 M. Alex Smith Biodiversity Institute of Ontario & Department of Integrative Biology University of Guelph,
[email protected] ABSTRACT
The urban environment is expanding at a never before seen rate. Existing natural environments within urban centres, such as forests or woodlots, are exposed to increasing anthropogenic pressures of degradation, fragmentation, biological invasion and destruction. One key to our capacity to understanding these changes will be ongoing monitoring through time. If such monitoring is democratized and publicly available then one may assume that a marginalized environment may become more valued by the human population. On the University of Guelph campus in Ontario, Canada, the “Dairy Bush” is an 8.5 ha woodlot that has been part of the city and the university campus since 1830. The sign outside the Bush reads, "The Dairy Bush is a unique and delicate example of Great Lakes St. Lawrence forest in Southern Ontario, and serves as an outdoor laboratory for University of Guelph students." Between August 2009 and September 2010 I visited the Dairy Bush weekly to document a year in this urban woodlot using GigaPan panoramic images. Keywords
GigaPan, forest, phenology, phenology, Canada, Guelph, urban, myrmechory, INTRODUCTION
Natural environments within urban centres are exposed to anthropogenic pressures of development, fragmentation, biological invasion and degradation. As urban environments are expanding at a never before seen rate (United Nations 2008, World Bank 2009) we can expect this change to increase in both scale and frequency. Urban forests should be considered a key component of the urban infrastructure that helps maintain a healthy environment for urban dwellers with significant benefits to carbon dioxide sequestration, air quality, hydrology, noise reduction and other societal benefits (Dwyer et al 1992). Monitoring within these forests for negative changes to existing urban forests will be a key to our capacity to maintaining these benefits. Using democratized technology, like the GigaPan, such activities could include: monitoring changes of plants within season, and between season, activity measures (human and other animal), and monitoring the changes associated with human use. In the Ontario city of Guelph, on the University of Guelph campus, the “Dairy Bush” is an 8.5 ha woodlot that has been part of the city and the university campus since 1873. The woodlot is unique in containing naturally growing and planted species that include rare and listed species in Southern Ontario. While the origins of the name, “Dairy Bush” are not known it was common agricultural practice in the area to maintain a small woodlot in order that cattle could congregate there in the heat of the warm summer months – and so it is possible that cattle from the agricultural college may have been given access to the bush in the summer. Since 1975 the wood lot has become a “living laboratory” for numerous ecology students over the past 30 years. Also during this time – the both city of Guelph, and the University have seen a marked population expansion at the southern end of the city, surrounding the University. This led to quite substantial developmental pressures – and the former fields surrounding the bush to the north-east are now home to student housing and other retail developments (Figure 1). The encroachment of human use has led to increased traffic within the bush – and will likely lead to increased degradation through time. A housing development for students was established directly south of the Dairy Bush in 2003, and more development is occurring in the adjacent (former) fields to the south (Figure 2). HISTORY
From land clearing in the 1800s to 1993, the Dairy Bush was an isolated forest fragment surrounded by farmland. Since 1993, the southern and eastern edges have adjoined to high-use residential properties managed by the University. While it was once thought, that, like most of southern Ontario, the Dairy Bush was entirely comprised of secondary growth, a
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
1
A year in an urban forest: Dairy Bush GigaPan 2009-2010
dendrochronological analysis of the sugar maple in the bush found that there was a cohort of maple in the bush that had germinated in the 1780s (Larson 2003). When the Ontario School of Agriculture and Experimental Farm (currently the Ontario Agricultural College (OAC)) first opened in 1874 the first professor of agriculture was a Scottish arboculturist, William Brown. The first mention of the area now called the Dairy Bush was in OAC's 18th annual report, written in 1892 – “….with trees as already named [norway spruce, white spruce, european larch, austrian pine, walnut, butternut, english ash, american ash, white birch, american elm, linden, norway maple, sugar maple], to which was added some young mountain ash, hickory, catalpa, and sweet chestnut, all of which are now fairly well established." (OAC Report 1892). In 1906, Jacob Zavitz (Head of the University of Guelph’s forestry program, and the Province’s first forester) planted a stand of white pine (Pinus strobus) - still clearly visible in aerial photos from the 1930s (Figure 1). From the 1905 report, “However these plantations are of value from the experimental and educational standpoint." (OAC Report 1905). In 1976, Douglas Larson, University of Guelph, Botany professor, began taking students into the Dairy Bush to use as a natural laboratory (Vowles 1999). He noted that the increasing residential development in the areas surround the Dairy Bush were resulting in significant degradation of the woodlot through increased trail use, soil compaction and litter. BIOLOGY
In August 2009, I selected a position along an existing path that divides the natural side from the Zavitz planation. This area was selected in part due the height of the understory – the reduced level here permitted longer views within the panorama. It was also the intersection of several of the trails. This makes it likely that some of the 2009 Ecological Methods students would be performing some of their studies within the area visible in the GigaPan. Summer 2009 (Figure 3), and further that this locality is the most likely to experience degradation through habitual trail use. In 2009 we had a comparatively long fall (Figure 4) – with freeze up and permanent snow fall that stayed on the ground was delayed approximately two weeks. Winter 2009/2010 (Figure 5) was characterised by mild temperatures and little snow fall. Spring 2010 (Figure 6) was uncharacteristically warm – with temperatures approaching 300C in early April. Summer 2010 (Figure 7) was the first full 360 degree panorama of at the Dairy Bush site. One very interesting characteristic of this technology is the frequency with which it was possible to observe animals or plants that were missed while on the ground (eg Figure 8). Phenological changes
The year-long Dairy Bush GigaPan has captured many examples of seasonal changes in both native and invasive understory plants and animals and examples of environmental damage. An example of each is expanded upon below. Native Plants: Bloodroot (Sanguinaria canadensis)
Bloodroot (Sanguinaria canadensis) is a native perennial flowering plant common to the damp forests of eastern North America. Bloodroot emerges in the spring, flowers before the canopy has covered and is dependent on ants for seed dispersal, an example of obligate myrmecochory (Pudlo et al 1980). Bloodroot accomplishes this through the production of an external food body, the elaiosome – that contains protein, starch, vitamins and lipids as ant attractants. In 2010, the Dairy Bush GigaPans captured the emergence (Figure 9), and flowering (Figure 10) of one bloodroot plant on April 6 and 11th respectively. The plants produced seed pods (Figure 11) on May 05 which were cut off and opened on June 10 (Figure 12). While ant activity on the plant has not been captured in the panorama, carpenter ants (Camponotus pennsylvanicus) have appeared in other nearby GigaPan snapshots (Figure 13), and other species (Aphaenogaster rudistexana and Lasius) have been uncovered nearby. In other studies ants of the genus Aphaenogaster have been shown to disperse between 70 and 80 % of the bloodroot’s seeds (Ness et al 2009). I have frequently observed Aphaenogaster in the area just off the right hand side of the GigaPan under and within fallen logs. Habitat disturbance and degradation can disrupt the ant fauna as well as the relationship between ant and the plant whose seeds they disperse. Pudlo et al (1980) found that in more disturbed forest, the ant dispersal of bloodroot seeds was reduced due to a reduction in ant diversity, while Ness (2004) also reported that seed dispersal distance is negatively influenced by proximity to the forest edge (ie. degradation). As it appears likely that the Dairy Bush will continue to undergo more
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
2
A year in an urban forest: Dairy Bush GigaPan 2009-2010
development and increased use (Figure 2) – it will be useful to continue to monitor the ant diversity and the density of bloodroot plants as a measure of the degradation of the forest. Such a monitoring protocol could be facilitated with selectively positioned GigaPan panoramas taken through time. Invasive Plants: Garlic Mustard (Alliaria petiolata)
Garlic mustard is a biennial flowering plant which has expanded from its native range in Europe to become a very common understory plant in eastern North America. The Dairy Bush trail network has several first and second year growths of Garlic Mustard. These could be measured and enumerated to contribute to ongoing public surveys of invasive species in North America (e.g. The Garlic Mustard Field Survey (Global Invasions Network) http://www.goearthtrek.com/GarlicMustard/GarlicMustard2.html) Other inquiries
The Dairy Bush GigaPan also suggests other methods of capturing and analyzing long-term environmental data. For instance, foliar damage from trophospheric ozone could be tracked using the cherry bushes near the front of the panorama. These bushes could also be tracked for estimations of herbivory from larval Lepidoptera. The yearly rates of leaf-litter decomposition could be inferred from the GigaPan series. Finally – so many snails have been captured in snapshots – and one wonders if it would be possible to individually mark snails and attempt to “recapture” them in subsequent GigaPans. PHOTOGRAPHY
Panoramas were taken with a Canon PowerShot G10 (14.7 MegaPixels) camera on a Velbon PH-157Q tripod set approximately 1.5m high. The position varied slightly twice throughout the year, but the variance was less than 1 metre. Suggested focusing methodology for GigaPan panoramas is to manually lock the focus at a fixed focal distance. However, the range of depth of field presented in the forest resulted in the decision to utilize autofocus to capture greater detail in a structurally heterogeneous volume shot as the Dairy Bush GigaPan. Technical changes and observations invariably occur through any long-term data acquisition process – and the Dairy Bush GigaPan was no different. For example, the GigaPan stitcher version of August 2009 (0.4.3864 (Windows)) took, on average, more than 2 minutes per photograph to stitch. This was part of the pragmatic decision to shoot the panorama at 231 images (Figure 3; 231 (33 columns by 7 rows) stitch times were 9hrs, 12 min (2:23 per picture)). Due to changes in the stitching and uploading software (version 1.0.0805 (Windows)) dramatically reduced stitching and uploading times the Dairy Bush GigaPan has approximately tripled in size during the course of the year (Figure 6; 640 images, 5hr 16min stitch time, (30 seconds per picture). There were physical elements that I have learned in this year that I would adjust in re-starting a long-term panorama capturing program. Some are likely to my own photographic naiveté, but I list all in case they assist others. The Dairy Bush panorama is less than 360 – and the centre of focus happens to be facing nearly directly west. This orientation, combined with the available time of day (lunch time or late afternoon), resulted in difficult to capture exposures (See the over-exposed sections of sunshine through the trees in Figures 4, 5 and 6). Restarting this process, I would orient the centre of the panorama away from the direct sun – and would extend its degrees of coverage to a full 360o (as has been initiated in year 2 of the process, evident in Figure 7). Additionally, I would move the GigaPan off the beaten path. The original reasoning in placing it along a pathway was that it was a space that people could find – and in doing so – better relate to. It would also be an area most likely to exhibit patterns of damage through increased trail use. However, one unforeseen consequence of the location was that the markers to label tripod placement were regularly disturbed, removed or destroyed. As a consequence of this there are slight changes in panorama orientation that are evident between winter and spring 2010. I reiterate the recommendation that anyone undertaking this type of long-term GigaPan is to give in to the desire to expand on the panorama you are taking. Storage will expand – opportunity will not. The portions of the panorama that you end up not desiring can be stripped afterwards – they cannot be re-captured. In the second year of the Dairy Bush GigaPan series – I am shooting the full 360 degrees. The source files for the Dairy Bush GigaPan series currently (Sept 2010) includes 280 GB of storage space containing more than 2.75 million image files (source and tiles).
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
3
A year in an urban forest: Dairy Bush GigaPan 2009-2010
Technical thoughts for the future – or multiple modest proposals…
In the future, it would be interesting to see more infrastructures on the GigaPan viewer that would permit better links between GigaPans taken at different times (e.g. the snap shot time series of the bloodroot phenology is hard to find and follow through for a beginner viewer. Would a roll-over link that takes you from a snapshot at one space to the next panorama at the same spac.e work?). The projection of the Dairy Bush GigaPan on the sphere is impressive – the canopy rises above you – but most viewers likely don not experience this due to the extra steps involved in downloading the .kmz, opening another program and viewing there. Just as there is a link to full-screen, it would be very valuable to have an automatic link on the panoramas page to a full screen projection using the Google Earth html plug-in. The parts that are really missing in the G.Earth environment – are the snapshots. Can these be exported into the kmz or a full-screen projection? If so that would be an unmatched tool for outreach, education and exploration. On the ground – I would love to explore more innovative and clever ways to integrate the GigaPan robot into the bush. Could the camera and robot be mounted in place and take photographs automatically? What camera/lens combination would result in the best depth of field? Focus stacking – single images created from multiple images at different depths of field – resolved and unified in a GigaPan framework would be truly astounding.
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
4
A year in an urban forest: Dairy Bush GigaPan 2009-2010
FIGURES
Figure 1: Google Earth KMZ file with image overlays of the original field map for the Ontario Agricultural College from 1873, aerial photographs of the Dairy Bush from 1930, 1955, 1978 and 1994, location of cored trees and four GigaPan images.
Figure 2. A GigaPan of construction on the southern end of the Dairy Bush (02 June, 2010). http://gigapan.org/gigapans/50997/.
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
5
A year in an urban forest: Dairy Bush GigaPan 2009-2010
Figure 3. First GigaPan of the Dairy Bush in late summer 2009 (31 August, 2009). http://www.gigapan.org/gigapans/31738/. A full screen GoogleEarth projection of this panorama is available here. (Link shortened with http://bit.ly/, and Google Earth PlugIn necessary.
Figure 4. A GigaPan of the Dairy Bush in fall (18 November, 2009). http://www.gigapan.org/gigapans/37446/. A full screen GoogleEarth projection of this panorama is available here. (Link shortened with http://bit.ly/, and Google Earth PlugIn necessary.)
Figure 5. A GigaPan of the Dairy Bush in winter (20 January, 2010). http://www.gigapan.org/gigapans/41677/. A GoogleEarth projection of this panorama is available here. (Link shortened with http://bit.ly/, and Google Earth PlugIn necessary.)
Figure 6. A GigaPan of the Dairy Bush in spring (05 May, 2010). http://www.gigapan.org/gigapans/49029/. A full screen
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
6
A year in an urban forest: Dairy Bush GigaPan 2009-2010
GoogleEarth projection of this panorama is available here. (Link shortened with http://bit.ly/, and Google Earth PlugIn necessary.)
Figure 7. A 360o GigaPan of the Dairy Bush in late summer (01 September, 2010). http://www.gigapan.org/gigapans/58071/. A full screen GoogleEarth projection of this panorama is available here. (Link shortened with http://bit.ly/, and Google Earth PlugIn necessary.)
Figure 8. A snapshot of the Dairy Bush in Spring (2010) showing a bright green assassin bug (Reduviidae) that was captured by the panorama in Figure 6. I did not see the bug while in the bush. For more detail, see http://www.gigapan.org/gigapans/49029/snapshots/138373/.
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
7
A year in an urban forest: Dairy Bush GigaPan 2009-2010
Figure 9. A snapshot of the Dairy Bush in spring 2010 (6 April, 2010) showing the emergence of bloodroot plant. For more detail, see http://www.gigapan.org/gigapans/46507/snapshots/135264/
Figure 10. A snapshot of the Dairy Bush in spring 2010 (11 April, 2010) showing the bloodroot flowering. For more detail, see http://www.gigapan.org/gigapans/47735/snapshots/135263/
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
8
A year in an urban forest: Dairy Bush GigaPan 2009-2010
Figure 11. A snapshot of the Dairy Bush in spring 2010 (2 June, 2010) showing bloodroot seed pods. For more detail, see http://www.gigapan.org/gigapans/51015/snapshots/144514/
Figure 12. A snapshot of the Dairy Bush in spring 2010 (10 June, 2010) showing bloodroot seed pods opened and dispersed. For more detail, see http://www.gigapan.org/gigapans/51634/snapshots/146115/
Figure 13. A snapshot of the Dairy Bush in spring 2010 (18 May, 2010) showing a one potential bloodroot seed dispersing ant species, Camponotus pennsylvanicus, on white pine tree trunk close to the bloodroot plant. For more detail, see http://gigapan.org/gigapans/49841/snapshots/141314/.)
CONCLUSION
One key to our capacity to understanding the changes caused by the increasing pressures of urbanization will be ongoing monitoring through time. If such monitoring is democratized and publically available then one may assume that a marginalized environment may become more valued by the human population. The weekly GigaPan panorama’s that I shot in the Dairy Bush between August 2009 and September 2010 were an attempt to make this urban woodlot more accessible to the student and City population. Urban forests are exposed to increasing anthropogenic pressures of degradation, fragmentation, biological invasion and destruction – and there is evidence of such changes in the GigaPan series. Additionally, the series captured the phenology of the native flora and fauna of the forest.
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
9
A year in an urban forest: Dairy Bush GigaPan 2009-2010
In conclusion, I have learned a lot from undertaking the Dairy Bush GigaPan series this year. On a personal level, it has given me an hour each week where I stood silently in a forest – all only a ten minute walk from my office. I greatly appreciated this. It has also made meager contributions to popularizing the Dairy Bush itself (Lovell 2010, Vowles 2010, Smith 2010). This type of popularization is important for it keeps the Bush in people’s minds within the City and University communities. Hopefully, this will insert the Dairy Bush into any community discussion about further residential or business development plans for the area. The sign outside the Bush reads, "The Dairy Bush is a unique and delicate example of Great Lakes St. Lawrence forest in Southern Ontario, and serves as an outdoor laboratory for University of Guelph students.” – hopefully the GigaPan series will play a role in facilitating the future use of this outdoor laboratory. ACKNOWLEDGEMENTS
I would like to thank Mary Jo Knelly, Illah Nourbakhsh, Randy Sargent, Paul Heckbert and Dror Yaron of Carnegie Mellon University for their assistance and patience in providing the necessary equipment to undertake this project. Their enthusiasm is notable and infectious! Thank you to the Fine Family for their foresight in creating the Fine Foundation for Science – the program that first allowed me to experience the GigaPan. I would like to thank the two anonymous reviewers of this manuscript – their thoughtful comments improved the final product. I would like to thank Rick Turner for taking the Dairy Bush GigaPan when I was not in the country, and finally I would also like to thank Doug Larson and Kevin McCann (University of Guelph) for their interest for this project and passion for the Dairy Bush. The forest needs them. REFERENCES
1.
Beattie, A.J., D. C. Culver, R. J. Pudlo. (1979). Interactions between ants and the diaspores of some common spring flowering herbs in West Virginia. Castanea, 44:177-186.
2.
Dwyer, J. F., E. G. McPherson, H. W. Schroeder, and R. A. Rowntree. (1992) Assessing the benefits and costs of the urban forest. Journal of Arboriculture 18(5):227-234.
3.
Larson D.W. (2003) Dairy Bush Management Proposal (unpublished).
4.
Ness, J. H. (2004) Forest edges and fire ants alter the seed shadow of an ant-dispersed plant. Oecologia 138: 228-454.
5.
Ness, J. H., D. F. Morin and I. Giladi (2009) Uncommon specialization in a mutualism between a temperate herbaceous plant guild and an ant: are Aphaenogaster ants keystone mutualists? Oikos 118: 1793-1804.
6.
Lovell, J. 2010. Camera mount gets big picture, smallest details. Guelph Tribune. February 26, 2010. http://www.guelphtribune.ca/news/article/203869.
7.
Pudlo, R. J., A. J. Beattie, and D. C. Culver. (1980) Population consequences of changes in an ant-seed mutualism in Sanguinaria canadensis. Oecologia 146:32-37.
8.
Smith, M. A. (2010). GigaPan – a democratization tool, like DNA barcoding. iBOL Bulletin: The Newsletter for the International Barcode of Life project. 1(3):4-6. September 2010. www.ibol.org/gigapan/.
9.
United Nations (2008) World Urbanization Prospects: The 2007 Revision (United Nations, New York). http://www.un.org/esa/population/publications/wup2007/2007WUP_Highlights_web.pdf
10. Vowles, A. 1999 Exploring Roots of the Dairy Bush. At Guelph July 11, 1999. http://www.uoguelph.ca/atguelph/99-0811/articles/dairy.html 11. Vowles, A. 2010 Biologist Zooms In on Ants and Other Species. At Guelph. May 2010. http://atguelph.uoguelph.ca/2010/05/biologist-zooms-in-on-ants-and-other-species/ 12. World Bank (2009) The Little Green Data Book 2009 (http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/ENVIRONMENT/EXTEEI/0,,contentMDK:22180399~page PK:148956~piPK:216618~theSitePK:408050,00.html )
Proceedings of the Fine International Conference on Gigapixel Imaging for Science, November 11–13 2010
10
