2014: Listening to the Forest and its Curators: Lessons Learnt from a Bioacoustic Smartphone Application Deployment

Please download to get full document.

View again

of 10
9 views
PDF
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Document Description
Moran, S., Pantidi, N., Rodden, T., Chamberlain, A., Griffiths, C., Zilli, D., … Rogers, A. (2014). Listening to the Forest and its Curators: Lessons Learnt from a Bioacoustic Smartphone Application Deployment. In Proceedings of the SIGCHI Conference
Document Share
Document Tags
Document Transcript
  Listening to the Forest and its Curators: Lessons Learnt from a Bioacoustic Smartphone Application Deployment Stuart Moran 1 , Nadia Pantidi 1 , Tom Rodden 1 , Alan Chamberlain 1 ,   Chloe Griffiths 2 , Davide Zilli 3 , Geoff Merrett 3 and Alex Rogers 3 Mixed Reality Lab University of Nottingham, UK {firstname.lastname} @nottingham.ac.uk Located Knowledge Lab Wales, UK C.Griffiths@LocatedKnowledgeLab.org Electronics and Computer Science University of Southampton, UK {dz2v07;gvm;acr} @ecs.soton.ac.uk ABSTRACT Our natural environment is complex and sensitive, and is home to a number of species on the verge of extinction. Surveying is one approach to their preservation, and can be supported by technology. This paper presents the deployment of a smartphone-based citizen science  biodiversity application. Our findings from interviews with members of the biodiversity community revealed a tension  between the technology and their established working  practices. From our experience, we present a series of general guidelines for those designing citizen science apps. Author Keywords Bioacoustics; biodiversity; citizen science; community  practices; mobile; participatory sensing; tradition; tension. ACM Classification Keywords H.5.2 [Information Interfaces and Presentation]: User Interfaces - Interaction styles  INTRODUCTION Debates concerning our environment often center on the threat people pose to the biological diversity of our planet. A raft of national and global initiatives has emerged to highlight the importance of biodiversity and the threat  posed by the extinction of species. A critical part of many of these initiatives centers on gathering detailed information about the diversity of our natural world and the status of the variety of species involved. Initiatives such as the IUCN red list (www.iucnredlist.org) maintain a detailed record of species under threat. Core to this is the undertaking of surveys of our natural world to capture evidence of species and to understand the nature of the particular population. The scale involved in capturing the biological diversity of our planet is vast. The need to capture information about species is becoming ever more critical given the rate with which natural habitats are disappearing. This is particularly important as environmentalists need evidence to inform environmental policies and to argue for programmes of intervention. The scale of the task at hand and its societal importance makes biodiversity a natural candidate to exploit the growing trend of citizen science. Initiatives such as Galaxy Zoo (www.galaxyzoo.org) or Folding @ home (http://folding.stanford.edu/) have utilized digital technologies to recruit volunteers from the general public to tackle critical scientific challenges. Volunteers have been  provided access through a growing number of internet tools to catalogue galaxies, transcribe handwritten texts and help in the folding of protein structures. The increasingly ubiquitous nature of smartphones that offer sophisticated sensing capabilities has opened up the possibility of moving these volunteer citizen scientists out of the home and into the field allowing the general population to be engaged in undertaking the species surveys so critical to biodiversity and consequently conservation. This paper presents the development and deployment of a smartphone-based citizen science biodiversity application. The application seeks to recruit members of the general  public to find evidence of a rare cicada in the UK that might already be extinct. The challenge of finding evidence for the cicada is that its distinctive chirp is at a frequency that is only just audible to the human ear. However, it is within the range of the microphones in most smartphones. Hence, it is  possible to record, process and detect a cicada’s song directly on the phone; with location determined by the GPS. We discuss findings from a set of studies that involved interviews with employed biodiversity professionals and volunteer amateur naturalists as well as observations of a  biodiversity event where the app was launched to the general public. In addition to commenting upon the application itself, our studies highlight the distinctive challenges that emerge as citizen science goes mobile and engages with scientific endeavors in the field. Although offering considerable potential advantages in terms of scale and engagement with the public, the turn to citizen science in biodiversity also raises tension in terms of the nature of the scientific endeavor, and its current culture and practices. We elaborate some of these key tensions and suggest a   Copyright is held by the owner/author(s). Publication rights licensed to ACM. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed or profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others han ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific ermission and/or a fee. Request permissions from permissions@acm.org. CHI 2014 , April 26–May 1, 2014, Toronto, Ontario, Canada. Copyright ©ACM 978-1-4503-2473-1/14/04...$15.00. http://dx.doi.org/10.1145/2556288.2557022   number of general guidelines for those involved in the development of citizen science applications. BACKGROUND While citizen science is a relatively new term, the tradition of public participation and collaboration in scientific research goes back centuries. Indeed much of this has focused on capturing information about the natural world. The National Audubon Society’s Christmas Bird Count (http://birds.audubon.org/christmas-bird-count) was established in 1900, and continues even today, with amateur and professional bird watchers joining forces in cataloguing  bird species. There are instances of recording even earlier than this, as during the 17 th  century farmers, hunters, clergymen and other members of the public actively contributed in collecting and classifying various nature datasets for research purposes [3,18]. Today, citizen science has been able to take advantage of internet based technologies. Citizen scientists work remotely on large-scale data sets that are increasingly  becoming a key feature of scientific investigation in the natural sciences. For example, Galaxy Zoo engaged the  public in examining images of galaxies and classifying their  basic characteristics. While the bat detective project (www.batdetective.org) asks volunteers to identify potential  bat calls in audio surveys collected by researchers. Other  projects have gamified aspects of such classifying tasks to make them more appealing to a wider audience, motivating and educating the public involved in the project [1]. Web- based communities have also been formed to support socialising and sharing between citizen scientists such as eBird [24] and Key To Nature [17]. Advancements in mobile and ubiquitous technologies have allowed a move away from domestic desktop interaction, and enabled people to contribute valuable new datasets by engaging with their environment in-situ . Participatory sensing approaches enable people equipped with sensors or their own mobile devices to collect information from their environment as part of their everyday activities and interpret it within context [5,6]. Environmental impact has  been a particular focus for participatory sensing with the general public tracking and measuring climate change and  pollution [2,8,19]. Existing work in participatory sensing for sustainability has shown positive results in making people more aware of, and actively engaged with, their environment. It has also shown the importance of the in-situ  experience and how it can contribute to the interpretation of the collected data [6,12,20]. Education has also played an important role, with environment-visiting based work such as Ambient Wood [21], Ubigreen [11] and e-Science in schools [23] using mobile devices/sensors to educate participants about the environment. These systems emphasize managed groups [4] openly exploring, and reflecting on the physical space [13,21]. In distinct contrast, Citizen Science has emphasized using an anonymous crowd to survey/analyze data. Work related to this has revealed a number of challenges in terms of privacy, storage, dissemination and interpretation of the data. Paxton and Benford [20] stress the “ complex balance between the various needs, limitations and the preferences of both the users and the  sensing tools” and how this can affect both the experience of the user as well as the quality of the data interpretation. In considering this, along with the capabilities of mobile  phones today, we can see that participatory sensing lends itself to the field of biodiversity monitoring, in terms of supporting the recording and classification of species. Bioacoustic Participatory Sensing One of the main benefits in the use of mobile phones for  biodiversity is in the use of their microphones as a sensor [14]. In many respects this can serve as an ideal tool for species surveying. As would be expected, acoustic sensors typically focus on species with  recordable  and  predictable  vocalizations, such as amphibians, insects and birds [26]. With an ongoing trend in the increased adoption of smartphones by the public, there is a move toward utilizing the onboard microphone for recording purposes. There are already projects which make use of microphones, such as CrowdSense@Place [7] for the classification of places  based on sound and NoiseTube [16] for assessing noise  pollution. Much of this work relies on server-side post  processing of the recordings using software such as the Extensible Bioacoustic Tool (XBAT) [10]. Equally, with the computational power available on smartphones, there exist possibilities for localised classification of acoustics. This type of automatic detection for surveying has been described as the ‘ ultimate’ smartphone application [15], and as a paradigm shift for many ecologists [26]. In the following section we describe a smartphone application that seeks to bring about such a shift. CICADA HUNT MOBILE APPLICATION   The New Forest cicada ( Cicadetta montana s. str.) is the only cicada native to the UK. The prevalence of the species is currently in question as the last unconfirmed   sighting of the cicada was in 2000. Along with a general decline in the species there are three problems, which make surveying for the cicada particularly challenging for professionals. 1)   The New Forest cicada lives underground for seven to eight years as a larva, before emerging and taking on its adult form for just four to six weeks between May and July. Unlike periodical cicada, its emergence is not synchronized, and thus, this significantly decreases the number of cicadas present during the already small and tight timeframe. 2)   The size of the New Forest is approximately 600km 2 , making it impossible for a small number of experts to effectively cover the area during the few sunny days when the males sing. 3)   The cicada sings at a frequency of 13-14kHz, which is at the limits of hearing for most adults. This makes  one of the primary means of in situ identification of insects by experts unusable (i.e. listening). These difficulties in coverage, detection and a narrow season motivated the surveyors to explore the possibilities of a technological solution. Through a series of meetings with core stakeholders (expert entomologists and members of the New Forest National Park Authority), the use of  participatory sensing was agreed upon as a viable and effective technological medium for promoting and assisting in the survey of the cicada; particularly given the untapped ~13 million day-visits per year to the New Forest National Park. Design and Development The pervasiveness of smartphone ownership underpins their appeal as a participatory sensing tool in the forest setting. Furthermore, through trials it was found that the microphone in a typical smartphone was sensitive enough to pick up the frequency of the cicada. To assist in the  process of acoustic classification, an algorithm was developed to accurately distinguish and specifically detect the New Forest cicada [27]. Given the onboard computational power available on smart phones, it was also  possible to run the (typically server side) classification algorithm locally on the device in real time. This classifier is based on a hidden Markov model, which is fed the ratio of two key frequencies extracted through the Goertzel algorithm as a single feature vector (for more details see [27]). It was trained using recordings from Slovenia where the same species of cicada is still abundant. The classifier also has the potential to detect common species of Orthoptera, including: wood cricket, Roesel’s bush cricket and a field grasshopper. An interesting constraint during the design of the app was the difficulty in trialling it in-situ. For this reason, a number of circuit board ‘electronic cicadas’ were developed to emit a constant acoustic square wave between 15kHz–18kHz to imitate the cicada’s song. These were used during  preliminary field trials of the application which were subsequently used to iteratively inform the design of the mobile application. The final design (called ‘Cicada Hunt’) included a sonogram, which was used to visualise the algorithm, taking inspiration from the popular music application Shazam (www.shazam.com) and drawing on a user’s likely familiarity with it. The sonogram was thus visualised in a circular way, as opposed to the more commonly found horizontal format. The sonogram displayed the real-time fluctuations relating to the audio’s frequency and amplitude. This detail was not only intended to act as a means of education for users about frequency and sound,  but also to speak to professionals who frequently make use of sonograms. Interaction with the application was designed to be as simple and lightweight as possible, in order to serve Figure 1 ‘Cicada Hunt’ mobile phone application many different types of users. To use the app, users press the cicada icon (see Figure 1 (A)) to start a survey, which records and analyses the next 30 seconds of audio in the environment. If a cicada is detected, the users are informed and a report consisting of the location, time, and audio, is uploaded to a server. In the case of a negative record, a report is also  being sent with the location, date and time but without any sound file. Users can review the record of their sent reports on the app (see Figure 1 (B)) and can also access information about the app itself (instructions, project), the cicada species and habitat, and the New Forest National Park in general (see Figure 1 (C)). Access to all users’ records is provided through a password protected dashboard on the project’s website. A typical feature of a participatory sensing application is the use of a map to indicate where surveys have been conducted. This can help with motivating surveyors to fill in the recording gaps on the map and also to create a sense of community. However, during pre-development discussion with the stakeholders, this was not perceived as a desirable feature as it was anticipated that it might encourage users to wander off the tracks and venture deep into the forest (to fill the gaps). This is a particular problem, as the cicada and ground nesting bird seasons coincide. For this reason, this feature was consciously excluded from the application. Deployment and Use The application was launched on both the iOS app store and the Android Google play store during May 2013 (search for ‘Cicada Hunt’). This was intended to coincide with the emergence of the cicadas. The app was promoted through significant news press coverage and with a launch event during a Bioblitz. Over a period of 3 months, more than 4,000 surveys were conducted, and 1,500 unique downloads to phones; unfortunately no cicada was found. The top ten users accounted for almost !  of all surveys, suggesting a key group of power users (see Figure 2). While these usage statistics provide some indication of the level of use of the app, they say little about who used it and what the issues involved in using the app might be.   (A) (B) (C)   Figure 2 Top ten anonymous power users and their devices Method In order to build a richer picture of app use, we carried out a number of in-depth interviews with a range of employed  biodiversity professionals and volunteer amateur naturalists. Participants were selected based on the types of biodiversity activities they engage with. We focused on those that carry out surveying, store and manage data, and make strategic decisions regarding forest management. Participants were approached through the New Forest National Park Authority (NFNPA) and local surveying groups in Hampshire and Sherwood. Interviews were carried out at locations where participants conduct their activities (i.e. during the bioblitz, at the New Forest National Park Authority and at the Sherwood Forest Trust). A total of 15  participants were interviewed (~1 hour), with 8 male and 7 female, aged between 23-72. Interviews were conducted over a 3 month period. Participants are grouped by role and location: 1)   Forest Managers, NFPA [ FM1 ,  FM2, FM3 ] 2)   Ecologists, NFPA [ E1 ,  E2 , E3 ,  E4 ] 3)   Biological Data Managers, NFPA [ D1 ,  D2 ,  D3 ] 4)   Community Managers, Sherwood [ CM1 , CM2 ] 5)   Rangers, Sherwood [ R1 ,  R2 ] 6)   Ecologist,   Ceredigion [ E5 ] Ecologists and rangers are active fieldworkers who directly conduct surveys of the environment. They carry out similar curation-based tasks, but each specialises in a particular species including: birds, bats, lizards, dragonflies and pond life. The remaining roles are typically adopted by ecologists later on in their careers. For example, biological data managers who receive the data from surveys, storing the information in databases and managing how it is shared internally or across local authorities. Forest and community managers focus on strategic decisions related to cultural and natural heritage, landscape and habitat preservation, and  public engagement activities. During the interviews, our focus was the utility of a citizen science approach to the scientists involved, how the application was received by domain experts (professionals and amateurs) and how it might be used for surveying and educating the public. Interviews were transcribed in full, and an inductive thematic analysis was carried out. The data was iteratively coded at a low level, and grouped together to reveal themes. The interview results were augmented with a number of short participant-observation studies during a public  biodiversity event in the New Forest (Bioblitz). Bioblitzes are outdoor events where members of the public meet-up with professional naturalists to intensively survey as many species as possible over a 24 hour period. The aim is to  both engage the public with nature and collect genuine survey data. Approximately 60 people, mainly young families and older amateur naturalists, took part during the event. The event was held in an open field in the New Forest, with pop-up stalls providing different nature related activities for children and for signing up to organised surveys. We observed a number of these intensive surveys over the 24 hours including: moth, bat, plant, bird, seashore and pond surveys (see Figure 3). Each event saw small groups (up to 10) guided by a professional naturalist, conducting real surveys of the species over a few hours. This gave us the opportunity to better understand some of the existing practices of surveying, but to also examine how they are used to engage members of the public. The cicada app was also officially launched to the public during the event from a stall which provided free Wi-Fi. This gave us the opportunity to observe and discuss the use of the application in-situ. Through the interviews and the actual deployment of the cicada mobile application the tension between engaging and inviting citizens to the forest, and protecting the habitat of the species under consideration came to the fore. This tension underpinned the previous decision to exclude the map style interface traditionally associated with  participatory sensing approaches. As we began to unpack this, we found that the app was surfacing critical issues about the use of citizen science for biodiversity. However,  before we consider these issues in detail, it is worth understanding the nature of the biodiversity community. THE BIODIVERSITY COMMUNITY The biodiversity community includes both biodiversity  professionals and amateur naturalists who play an active role in the work of the community. Biodiversity  professionals are involved as part of their job with the natural environment. Their daily work entails a variety of activities depending on their specific role but, in general, they engage with monitoring and preserving natural habitats and the species that are part of those. Some of them might  be forest rangers; others might be ecologists, scientists (entomologists, botanists etc) or heritage specialists. Most of them share their time between being out in the field observing and collecting information about species, and deskwork analysing that information, creating reports and disseminating them for purposes of planning, policy making and educating. Amateur naturalist is a term used by the biodiversity  professionals to describe members of the general public that engage in species surveys and other biodiversity activities.    0   5   0   1   0   0   1   5   0   2   0   0 24723615376585240383836    2   5   0   N  u  m   b  e  r  o   f  r  e  p  o  r   t  s iPhone 4iPhone 5iPhone 4SGalaxy SIII12345678910Rank    Figure 3 Bio Blitz in the New Forest National Park Amateur naturalists self identify with the term and distinguish themselves from the professionals despite the fact that sometimes they are just as knowledgeable as a result of their long-term engagement with species recording. Very often, they are locals, residents near a park area that work closely with the county recorders in regular, organized group surveys or people who just record things on their own while walking the dog. The biodiversity community, both professionals and amateur naturalists, share an affinity and sense of wonder for the natural environment that motivates them to engage with its preservation. In many cases this seems to have been instilled in them from a young age. Many of the  professionals that we interviewed vividly described memories of their childhood with family walks in the forest and the excitement of seeing and getting to know more about the species: “From an early age with my parents and they instilled that kind of… the wonder of nature and the world is such a diverse place and people are so diverse […] and they take it for granted. To understand it is just beautiful and it changes, day to day, season to season […] you always learn something new no matter where you went.” CM1 Protecting the natural environment is of utmost importance for this community. Biodiversity professionals identified two activities as vital in achieving that: conducting surveys and educating the public; in particular children. Instilling a love of nature, biodiversity and its preservation in the younger generations is of great importance for all  professional and amateur naturalists: “I’m in to what I do because of my dad. You know, a natural interest from a young age. I went out with him and I fell in love with nature, but a lot of kids these days don’t have that […] you need to get them while they’re young to kind of nurture that interest and that passion.” R1 “If I have enthused, you know, 5 kids really into it, really excited, for me that makes my job worthwhile.” R2 Survey as a Tool of Data Collection The emotional attachment to their work is also reflected in the professionals’ view of surveys and the tools used for surveying. Despite the advances and availability of technology in this digital era, professionals tend to conduct Figure 4 Amateur naturalists surveying species on paper surveys using pen and paper, and ID books (see Figure 4). Several reported using a camera or audio recorder and a GPS to define location, but equally admitted to avoiding using digital devices when possible as they lack the romanticism of the old ways. Pointing at a smartphone, one of our participants said: “There is nothing romantic about this, you can see the records but you cannot feel them as you would with the ones in your nature diary.” E1 In the same context, another referred to the feel of their field guide: “You have your field guide that you have had for years and you know, you kinda know the pages because of the way you have twisted them? It’s things like that isn’t it?” R2 Amateur naturalists conduct surveys in the same way, often with even less digital support unless this is done in the context of a joint organized survey with the professionals. After a survey is complete, the collected records usually have to be transferred into a digital form so that they can be disseminated, checked and archived in county and national databases, and planning, management and policy making offices or other relevant organizations. Not only is this  process tedious, but it is also clear that people are not  particularly inclined to share data: “You suspect there’s an awful lot of useful data out there, but some people umm don’t like sharing it […] it’s easy for them to put in a note book and then it’s more hard work and perhaps not so much what they want to do is to then get those records into the biological record centres.” E2 “[Another heritage group] been surveying for 40 odd years, and they’ve got a fantastic database but they won’t let us see it.” D3 Checking the work of volunteers Accessing, disseminating and checking survey records is also challenging with respect to the volunteers. Biodiversity  professionals and amateur naturalists both stressed a distinct lack of consistency among the recordings of volunteers: “When you get some data in, from some people that will be very meticulous, to the letter and ya know […] and then you get really sketchy data, and to try to put the two lots of data together… was
Similar documents
View more...
Search Related
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks