Citizen Science

OpenStreetMap (OSM) is currently the world's largest database, to which millions of volunteers contribute. By now, it has nearly 4 billion nodes, 400 million ways and almost 5 million spatial and logical relationships among the elements. This database is a precious source of information and the data it contains can be reused by everyone in accordance with its Open License, the Open Data Commons Open Database License (ODBL). Anyone is free to copy, distribute and adapt the data, providing the paternity to OSM and its contributors and guarantying that any alteration of the data is made available with the same license. Often the accuracy of the OSM data is called into question. This work tries to partly answer this question by providing a method for measuring the metric accuracy of a target layer, like for instance that corresponding to OSM buildings (the layer of building footprints), by comparison with a second reference layer, for instance an authoritative one, representing the ground truth. The method is therefore more general, allowing the evaluation of metric differences between two layers, but it can be applied to OSM data in order to check their quality. It is complemented by a web mapping tool, where users can upload their target and reference layers and have back all the relevant statistics. The prepared procedure also provides the ability to warp the target layer to the authoritative one with a least squares approach based on the affine or on multi-resolution spline transformation. The transformation parameters can then be applied to all layers (road, hydrography, etc) spatially homogeneous with the target one (i.e, the building footprints layer), making them spatially homogeneous to the reference. Besides the procedure and the web application, statistics for some large Italian and foreign cities will be shown (Milan, Paris, Berlin, S.Francisco, Boston), comparing OSM data with official data with accuracy at urban scale (from 1: 1000 to 1: 5000).

Water colour is a very informative indicator of the ecological state of marine and fresh-waters. Until recently, however, colour information has only been measurable with a suite of unwieldy scientific instruments. With the ubiquity of smartphones, several phone applications have been developed which offer the exciting prospect that water colour measurements and associated metadata can now be collected over wide spatial and temporal scales by citizen scientists. A new project starting in July 2017 aims to harness the enthusiasm of Australian citizen scientists interested in the water quality of their local regions. These individuals will generate a large pool of valid data to calibrate satellite information, and provide a synoptic overview of river, lake, and coastal water quality for natural resource management in Australia. The usefulness of the data for calibration of satellite remote sensing data and its applicability in water quality management such as an early warning system of harmful algal blooms will be assessed. Each participant will obtain a quantitative understanding of how local water bodies change seasonally and in response to short duration events like floods and cyclones. As Australia has vast underpopulated regions, a targeted group of citizen scientist will include indigenous rangers working in isolated regions in tropical Australia where scientific investigations are limited or non-existent. Once organisations and communities are engaged, CSIRO will provide training for the mobile phone application, deployment of a Secchi disk, use of a basic chemical water quality test kit and data transfer. Data will be uploaded by the participants and results will be returned to the users through a website and regular newsletters. This presentation will outline the early stages of implementation of the project, the benefit to citizen scientists in enhanced knowledge of water quality issues and environmental awareness, and the benefits provided in calibration and validation of satellite data.

The goal of the E2mC project is to demonstrate feasibility and usefulness of the integrating social media analysis and crowdsourced information within both the Rapid Mapping and Early Warning Components of Copernicus Emergency Management Service (EMS). Since 2012, the Copernicus EMS provides timely information for emergency response in relation to different types of disasters, as well as prevention, preparedness, response and recovery activities. It operates through four modules (Rapid Mapping, Risk & Recovery, European Forest Fire Information System (EFFIS) and European Flood Awareness System (EFAS)) that constitute to two main Service Components: Mapping and Early Warning. Through these two components, Copernicus EMS supports crisis managers, Civil Protection authorities and humanitarian aid actors dealing with natural disasters, manmade emergency situations, and humanitarian crises during the different phases of crisis management. The experience gained in these recent years of operational services has highlighted the need to enhance the service performance of both the Components, in terms of timeliness, quality and production capacity. Relating to the Mapping Component, satellite images are the key source for the maps production. One of the main user requirements of the EMS is to receive first crisis information within the first 24 hours after the disaster, while today it is not unusual to experience delays up to 72 hours, mainly due to the availability of the first usable post-event satellite image, caused by various casues, including satellite tasking and orbital constraints, bad weather conditions, and late activation. Moreover, the crisis maps, purely based on satellite information, have known quality limitations due to the physical constraints of satellite acquisitions (e.g. resolution or near-nadiral view) that negatively affect accuracy. Social media have proven particularly valuable in collecting information in emergency situations, such as in earthquakes, in particular in very destructive ones such as Haiti and Nepal, and in large floods and hurricanes, such as Hurricane Matthew and floods in the Philippines, in the different phases of such events. Preliminary analyses and surveys conducted in the requirements analysis phase of the project have highlighted that the operators involved in the crisis map production during an event (Rapid Mapping service) use social media information gathered manually, mainly with internet search and download. A fundamental innovation with E2MC is to provide a technology platform that supports the automated retrieval and analysis of social media information, combined with crowdsourcing, with the general goal of improving the quality and dependability of the information provided to professional users within the Copernicus network. This paper discusses the E2mC approach to the integration of social media and crowdsourcing within Copernicus rapid mapping and early warning service components. It discusses the functional architecture of the platform that will be developed by the E2mC team to support this integration. Preliminary results are discussed providing insights on the type, quality, amount and timeliness of the information that can be found on social media during an emergency and on the tasks that are best candidate for crowdsourcing.

Data sources for sustainable long term environmental monitoring system for any official or operational use should be reliable, consistent and unbiased both in spatial distribution of observations and observed values distriribution domain. Citizen observations (CO) have challenges to meet any of these requirements. However, there is a potential for both cost-effective data gathering and citizen involvement with co-operation and information dissemination benefits beyond acquiring environmental data. In fact, at current state of technology the problem is not acquisition of georeferenced data from mobile communication devices, rather the problem is how to motivate observers to commit to the information provision activity among other possible competing activities. With outlined process of introducing participatory citizen observations, the motivation of observers and the expectations of data gathering organizations can meet. By committing to extended Open311 information exchange practices the questionnaires for formalizing the provided CO data, similar questions can be provided across different user platforms and interfaces, supporting multiple languages. Practices of ensuring service discovery, data availability and quality are outlined for multi-organization framework. For water qality related CO measurements with Secchi3000 technology are reviewed with potential for satellite Earth Observation (EO) synergy.

In 2010 has been launched a participative system (web-based - meduse.acri.fr ) to report on presence/absence of jellyfish on French Riviera. All citizens, locals and tourists, were invited to report on jellyfishes. The result is double : i) a near-real-time mapping of risk, widely open to public, to manage their recreative bathing and ii) a mean to get statistical data, useful to carry out scientific studies of correlation of jellyfish presence and specific environmental conditions (that can be observed in situ and from space) ; scientific papers have been published making use of this “crowd” data and, by this, have demonstrated the quality of this approach for scientific purposes. This precursor system has been fully adopted by a very large public and can reasonably considered as a deep success (around 30.000 observations reported – that makes this data set the largest in Europe regarding jellyfish - more than 1,3 millions of connections, 7.500 registered observers…). This reporting/mapping system is still largely alive and is regularly promoted by public media (mainly for public health purpose). Based on this (unexpectedly large) success, and lessons learnt regarding acceptance of public to this type of monitoring, we have developed a “generic” system (Simplex (TM)) based on the contribution of public, or semi-public for scientific applications. The development of this system is supported by the French Ministry of Environment and allows any type of environmental monitoring in complement to other types of observation (EO, modeling, in situ…). SimplexTM consists in three complementary branches i) one generic and simple smartphone application for an observing cohort, ii) one web-GIS allowing management of this cohort by a coordinator and crossings with external input (environmental and socio-economic) and iii) NRT direct access to Copernicus (EO and models) data at the time/location of the observer and for the coordinator. Simplex(TM) is presently in beta-version and under test for four applications with various stakeholders (sea mammals monitoring, macro-wastes at sea, …). Results and possible benefit for EO validation will be presented at the EO Open Science workshop.

The value of science and data has, as of late, been a hot button topic. Demonstrators are marching and headlines are being written about future government investment in data collection in support of science. And even while some governments are pulling back, we are in the midst of a global sensing revolution. Innovations in satellite and rocket manufacturing have ushered in a step-change: 2017 was the first year in history that more than 100 Earth observation satellites were launched in a single year. Today, most locations on the land surface now have their picture taken multiple times per day. Ecological, geological and human-caused changes on the Earth’s surface are visible like never before: the demographics of remote tropical trees, flowering, dropping their leaves, and flushing. Alpine lakes, dammed by glacial ice, draining and re-filling. Sand encircling Indian Ocean Islands, ebbing with seasonal currents. And yet, each of these dynamic, diverse Earth systems are under intense and increasing threat from climate change, human expansion and loss of biological diversity. If it is coupled with accessibility for scientific use, this global sensing revolution gives the scientific community an unprecedented opportunity to conduct Earth science in real time. So far, the results are encouraging, with more and more commercial data becoming accessible to scientists and researchers. But, this new vantage comes with challenges: multiple sensors and platforms, operated by scores of agencies, mean analyses must be increasingly interoperable to keep pace. The study of change on our Planet must be rigorously tuned to technological changes in the very sensors that allow us to track that change. How we meet the challenge of synthesizing and drawing insights from these evolving data streams—as a global community—will determine our planet’s fate.