This page contains a Flash digital edition of a book.
ecology Taking a view on shedloads of data


Human activity is a major factor in the development of ecologies, particularly where urbanisation or infrastructures are involved. Building a motorway disrupts the ecologies through which it passes on a very large scale during construction, and leaves them permanently changed in the longer term, but also creates new corridors between them as embankments mature into habitats of their own. A housing development (quite apart from the construction phase) drops a whole new ecological island into the middle of an existing one. Less obvious is the role which aesthetics play in such impacts. Successful resistance to development in its area by a human community will prevent ecological change in one place and may either prevent it from occurring at all or, in many cases, redirect it elsewhere. Some protests (motorways through breeding grounds for endangered toads, for example, or tidal barrages across estuaries) are on


explicitly environmental grounds, but many more are powered by a desire to preserve aesthetic aspects. Perhaps the most prominent example of this is the


resistance to siting of wind turbines which, practical arguments aside, produce extreme reactions of love and hate among those who live in the areas where they might be erected. Since the arrival of a turbine farm in an area will inevitably have positive, negative and zero sum impacts on local ecologies, aesthetic perceptions become a deciding factor in ecological outcome. The exact anticipated visual, rather than actual, impact of turbines thus becomes an ecological force. Objectively assessing that visual impact can be problematic. Enter ‘viewshed’ analysis. A ‘viewshed’ is a landscape component visible


from a given viewpoint. In this context, it will concern the number of turbine structures visible, at any one time, to a human observer positioned within the landscape under consideration. Viewshed analysis


(a common facility in geographical information software) establishes exactly what can and cannot be seen from each specified cell within a digital elevation model of the terrain.


Atmos Consulting, which specialises in renewable


energy and environmental issues, is one user refining this approach. The computer model which Atmos is developing seeks to reduce the complexity of viewshed analysis for proposed wind turbine developments to a prediction of cumulative visual impact. The aim is to provide information in a form which takes relative distance, size and perspective into account while delivering easily comprehended output that will be of practical use in public consultations and planning decisions. As well as simplifying the consumer view of the data, Atmos’ model also seeks to increase resolution so that individual turbines (rather than aggregated farms) contribute to the outcome.


Nepal, are found to spatially coexist with humans by temporal adjustment: they alter their habits to use paths at times when their human competitors do not. Te authors set up automated cameras inside and outside a designated park to monitor the movements of tigers, their primary prey species, and human movements by category (resident, tourist, military and vehicular) with recording of time, GPS location, and so on, using statistical point- process modelling to estimate densities and traditional nonparametric methods to test for differences between loci. Shared and unshared pathways also feature


prominently in another of my recent analytic involvements: a long-term mapping, spatially and temporally, of fauna routes in a moorland habitat. Te routes range in scale from those of shrews and other small rodents up to the largest local predator, the fox, with agricultural species, such as sheep or horses, and humans themselves as additional referent layers. A few routes having


been mapped in detail at a given scale, and their structures studied, volunteers go out and locate points on new ones. Data on the known routes are then used to try and predict how the new dots might be joined up; effort


is then concentrated on locating new points using these predictions. Raw and processed data pass back and forth between handheld computers in the field and a base analytic centre, refining models on the fly. As gaps are filled in, analyses are repeated and improved. When a sufficient route density has been built up, the process is applied to predicting and locating nodes, thus accessing new route systems. It’s not a precise science (when was statistics ever?), but offers considerable improvement over blanket fingertip searches. With the paths themselves located,


Further


information Analytica www.lumina.com Atmos Consulting www.atmosconsulting.com FragStats


GenStat www.vsni.co.uk/soſtware/ genstat


Statistica www.statsoſt.co.uk


XlispStat lib.stat.cmu.edu/xlispstat


16 SCIENTIFIC COMPUTING WORLD


www.umass.edu/landeco/ research/fragstats/fragstats.html


relationships between them are examined in ways analogous to those in the Chitwan study above. Once a pathway is opened up by a small animal, a larger one will oſten use it as the line of least resistance and thus enlarge it, aſter which it will again be utilised by a larger animal still, and so on. Te smaller species may then abandon the route and forge a new one, share it, or adjust their usage of it. Tere is a surprising degree of overlap between predators and prey, but also a considerable parallelism between routes of different scales. Statistical analysis of spatially related routes, combined with temporal traffic mapping, offers a lot of insight into the ecological relations at work within the habitat.


Playgrounds can be serious places My opening characterisation of ecology as a statisticians’ playground should not be taken as a trivialisation of the issues and implications involved. Far from it: the seriousness of meddling with mechanisms too big and complexly interrelated to comprehend is, of course, part of the draw. Before the advent of modern statistical analysis, there was no way even to begin the conversation about this seriousness. Before the arrival of widespread


computerised methods, there was no way to meaningfully develop it. Jabot and Bascompte recently[5]


pointed out that, even now, we


examine reductionist parts of the complex whole, and multitrophic methods are needed to build a more holistic analysis. Grimm and Railsback[6]


comment that ‘modelling systems


across scales and at multiple levels – especially to link population and ecosystem dynamics to individual adaptive behaviour – is essential for making the science predictive’. But statisticians are, first and foremost, explorers for whom both the scale and the seriousness of the challenge make the playground metaphor unavoidable. We are, like Newton, children mesmerised by pebbles and shells on the shore of an immeasurable ocean; but pebble by pebble, shell by shell, immensity is comprehended.


References and Sources For a full list of references and sources, visit visit www.scientific-computing.com/features/ referencesoct12.php


www.scientific-computing.com


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52