Search

The Amazon forests act as a global reserve for carbon, have very high biodiversity, and provide a variety of additional ecosystem services. These forests are, however, under increasing pressure, coming mainly from deforestation, despite the fact that accurate satellite monitoring is in place that produces annual deforestation maps and timely alerts. Here, we present a proof of concept for rapid deforestation monitoring that engages the global community directly in the monitoring process via crowdsourcing while subsequently leveraging the power of deep learning. Offering no tangible incentives, we were able to sustain participation from more than 5500 active contributors from 96 different nations over a 6-month period, resulting in the crowd classification of 43,108 satellite images (representing around 390,000 km2). Training a suite of AI models with results from the crowd, we achieved an accuracy greater than 90% in detecting new and existing deforestation. These findings demonstrate the potential of a crowd–AI approach to rapidly detect and validate deforestation events. Our method directly engages a large, enthusiastic, and increasingly digital global community who wish to participate in the stewardship of the global environment. Coupled with existing monitoring systems, this approach could offer an additional means of verification, increasing confidence in global deforestation monitoring. [ABSTRACT FROM AUTHOR]

The creation of crop type maps from satellite data has proven challenging and is often impeded by a lack of accurate in situ data. Street-level imagery represents a new potential source of in situ data that may aid crop type mapping, but it requires automated algorithms to recognize the features of interest. This paper aims to demonstrate a method for crop type (i.e., maize, wheat and others) recognition from street-level imagery based on a convolutional neural network using a bottom-up approach. We trained the model with a highly accurate dataset of crowdsourced labelled street-level imagery using the Picture Pile application. The classification results achieved an AUC of 0.87 for wheat, 0.85 for maize and 0.73 for others. Given that wheat and maize are two of the most common food crops grown globally, combined with an ever-increasing amount of available street-level imagery, this approach could help address the need for improved global crop type monitoring. Challenges remain in addressing the noise aspect of street-level imagery (i.e., buildings, hedgerows, automobiles, etc.) and uncertainties due to differences in the time of day and location. Such an approach could also be applied to developing other in situ data sets from street-level imagery, e.g., for land use mapping or socioeconomic indicators. [ABSTRACT FROM AUTHOR]

A variety of metrics are commonly employed by map producers and users to assess and compare thematic maps'; quality, but their use and interpretation is inconsistent. This problem is exacerbated by a shortage of tools to allow easy calculation and comparison of metrics from different maps or as a map';s legend is changed. In this paper, we introduce a new website and a collection of R functions to facilitate map assessment. We apply these tools to illustrate some pitfalls of error metrics and point out existing and newly developed solutions to them. Some of these problems have been previously noted, but all of them are under-appreciated and persist in published literature. We show that binary and categorical metrics, including information about true-negative classifications, are inflated for rare categories, and more robust alternatives should be chosen. Most metrics are useful to compare maps only if their legends are identical. We also demonstrate that combining land-cover classes has the often-neglected consequence of apparent improvement, particularly if the combined classes are easily confused (e.g., different forest types). However, we show that the average mutual information (AMI) of a map is relatively robust to combining classes, and reflects the information that is lost in this process; we also introduce a modified AMI metric that credits only correct classifications. Finally, we introduce a method of evaluating statistical differences in the information content of competing maps, and show that this method is an improvement over other methods in more common use. We end with a series of recommendations for the meaningful use of accuracy metrics by map users and producers. [ABSTRACT FROM AUTHOR]

Critical knowledge gaps seriously hinder efforts for building disaster resilience at all levels, especially in disaster-prone least developed countries. Information deficiency is most serious at local levels, especially in terms of spatial information on risk, resources, and capacities of communities. To tackle this challenge, we develop a general methodological approach that integrates community-based participatory mapping processes, one that has been widely used by governments and non-government organizations in the fields of natural resources management, disaster risk reduction and rural development, with emerging collaborative digital mapping techniques. We demonstrate the value and potential of this integrated participatory and collaborative mapping approach by conducting a pilot study in the flood-prone lower Karnali river basin in Western Nepal. The process engaged a wide range of stakeholders and non-stakeholder citizens to co-produce locally relevant geographic information on resources, capacities, and flood risks of selected communities. The new digital community maps are richer in content, more accurate, and easier to update and share than those produced by conventional Vulnerability and Capacity Assessments (VCAs), a variant of Participatory Rural Appraisal (PRA), that is widely used by various government and non-government organizations. We discuss how this integrated mapping approach may provide an effective link between coordinating and implementing local disaster risk reduction and resilience building interventions to designing and informing regional development plans, as well as its limitations in terms of technological barrier, map ownership, and empowerment potential. [ABSTRACT FROM AUTHOR]

Accuracy assessment, also referred to as validation, is a key process in the workflow of developing a land cover map. To make this process open and transparent, we have developed a new online tool called LACO-Wiki, which encapsulates this process into a set of four simple steps including uploading a land cover map, creating a sample from the map, interpreting the sample with very high resolution satellite imagery and generating a report with accuracy measures. The aim of this paper is to present the main features of this new tool followed by an example of how it can be used for accuracy assessment of a land cover map. For the purpose of illustration, we have chosen GlobeLand30 for Kenya. Two different samples were interpreted by three individuals: one sample was provided by the GlobeLand30 team as part of their international efforts in validating GlobeLand30 with GEO (Group on Earth Observation) member states while a second sample was generated using LACO-Wiki. Using satellite imagery from Google Maps, Bing and Google Earth, the results show overall accuracies between 53% to 61%, which is lower than the global accuracy assessment of GlobeLand30 but may be reasonable given the complex landscapes found in Kenya. Statistical models were then fit to the data to determine what factors affect the agreement between the three interpreters such as the land cover class, the presence of very high resolution satellite imagery and the age of the image in relation to the baseline year for GlobeLand30 (2010). The results showed that all factors had a significant effect on the agreement. [ABSTRACT FROM AUTHOR]

Citizens are increasingly becoming involved in data collection, whether for scientific purposes, to carry out micro-tasks, or as part of a gamified, competitive application. In some cases, volunteered data collection overlaps with that of mapping agencies, e.g., the citizen-based mapping of features in OpenStreetMap. LUCAS (Land Use Cover Area frame Sample) is one source of authoritative in-situ data that are collected every three years across EU member countries by trained personnel at a considerable cost to taxpayers. This paper presents a mobile application called FotoQuest Austria, which involves citizens in the crowdsourcing of in-situ land cover and land use data, including at locations of LUCAS sample points in Austria. The results from a campaign run during the summer of 2015 suggest that land cover and land use can be crowdsourced using a simple protocol based on LUCAS. This has implications for remote sensing as this data stream represents a new source of potentially valuable information for the training and validation of land cover maps as well as for area estimation purposes. Although the most detailed and challenging classes were more difficult for untrained citizens to recognize, the agreement between the crowdsourced data and the LUCAS data for basic high level land cover and land use classes in homogeneous areas (ca. 80%) shows clear potential. Recommendations for how to further improve the quality of the crowdsourced data in the context of LUCAS are provided so that this source of data might one day be accurate enough for land cover mapping purposes. [ABSTRACT FROM AUTHOR]

A study was undertaken to assess Envisat Advanced Synthetic Aperture Radar (ASAR) ScanSAR data for quantifying forest growing stock volume (GSV) across three boreal regions with varying forest types, composition, and structure (Sweden, Central Siberia, and Québec). Estimates of GSV were obtained using hyper-temporal observations of the radar backscatter acquired by Envisat ASAR with the BIOMASAR algorithm. In total, 5.3·106 km² were mapped with a 0.01° pixel size to obtain estimates representative for the year of 2005. Comparing the SAR-based estimates to spatially explicit datasets of GSV, generated from forest field inventory and/or Earth Observation data, revealed similar spatial distributions of GSV. Nonetheless, the weak sensitivity of C-band backscatter to forest structural parameters introduced significant uncertainty to the estimated GSV at full resolution. Further discrepancies were observed in the case of different scales of the ASAR and the reference GSV and in areas of fragmented landscapes. Aggregation to 0.1° and 0.5° was then undertaken to generate coarse scale estimates of GSV. The agreement between ASAR and the reference GSV datasets improved; the relative difference at 0.5° was consistently within a magnitude of 20-30%. The results indicate an improvement of the characterization of forest GSV in the boreal zone with respect to currently available information. [ABSTRACT FROM AUTHOR]