Search Results (6)

Geodata, geographical information science (GISc), and GeoAI (geo-intelligence workflows) play an increasingly important role in predictive disaster risk reduction and management (DRRM), aiding decision-makers in determining where and when to allocate resources. There have been discussions on the ethical pitfalls of these predictive systems in the context of DRRM because of the documented cases of biases in AI systems in other socio-technical systems. However, none of the discussions expound on how to audit geo-intelligence workflows for biases from data collection, processing, and model development. This paper considers a case study that uses AI to characterize housing stock vulnerability to flooding in Karonga district, Malawi. We use Friedman and Nissenbaum’s definition and categorization of biases that emphasize biases as a negative and undesirable outcome. We limit the scope of the audit to biases that affect the visibility of different housing typologies in the workflow. The results show how AI introduces and amplifies these biases against houses of certain materials. Hence, a group within the population in the area living in these houses would potentially miss out on DRRM interventions. Based on this example, we urge the community of researchers and practitioners to normalize the auditing of geo-intelligence workflows to prevent information disasters from biases. Full article

Aerosol optical depth (AOD) observations have been widely used to generate wide-coverage PM_2.5 retrievals due to the adverse effects of long-term exposure to PM_2.5 and the sparsity and unevenness of monitoring sites. However, due to non-random missing and nighttime gaps in AOD products, obtaining spatiotemporally continuous hourly data with high accuracy has been a great challenge. Therefore, this study developed an automatic geo-intelligent stacking (autogeoi-stacking) model, which contained seven sub-models of machine learning and was stacked through a Catboost model. The autogeoi-stacking model used the automated feature engineering (autofeat) method to identify spatiotemporal characteristics of multi-source datasets and generate extra features through automatic non-linear changes of multiple original features. The 10-fold cross-validation (CV) evaluation was employed to evaluate the 24-hour and continuous ground-level PM_2.5 estimations in the Beijing-Tianjin-Hebei (BTH) region during 2018. The results showed that the autogeoi-stacking model performed well in the study area with the coefficient of determination (R²) of 0.88, the root mean squared error (RMSE) of 17.38 µg/m³, and the mean absolute error (MAE) of 10.71 µg/m³. The estimated PM_2.5 concentrations had an excellent performance during the day (8:00–18:00, local time) and night (19:00–07:00) (the cross-validation coefficient of determination (CV-R²): 0.90, 0.88), and captured hourly PM_2.5 variations well, even in the severe ambient air pollution event. On the seasonal scale, the R² values from high to low were winter, autumn, spring, and summer, respectively. Compared with the original stacking model, the improvement of R² with the autofeat and hyperparameter optimization approaches was up to 5.33%. In addition, the annual mean values indicated that the southern areas, such as Shijiazhuang, Xingtai, and Handan, suffered higher PM_2.5 concentrations. The northern regions (e.g., Zhangjiakou and Chengde) experienced low PM_2.5. In summary, the proposed method in this paper performed well and could provide ideas for constructing geoi-features and spatiotemporally continuous inversion products of PM_2.5. Full article

Due to the characteristics of the Southern Amazonas Mesoregion (Mesorregião Sul do Amazonas, MSA), conducting on-site surveys in all licensed forestry areas (Plano de Manejo Florestal, PMFS) is an impossible task. Therefore, the present investigation aimed to: (i) analyze the use of geointelligence (GEOINT) techniques to support the evaluation of PMFS; and (ii) verify if the PMFS located in the MSA are being executed in accordance with Brazilian legislation. A set of twenty-two evaluation criteria were established. These were initially applied to a “standard” PMFS and subsequently replicated to a larger area of 83 PMFS, located in the MSA. GEOINT allowed for a better understanding of each PMFS, identifying illegal forestry activities and evidence of timber laundering. Among these results, we highlight the following evidences: (i) inconsistencies related to total transport time and prices declared to the authorities (70% of PMFS); (ii) volumetric information incompatible with official forest inventories and/or not conforming with Benford’s law (54% of PMFS); (iii) signs of exploitation outside the authorized polygon limits (51% of PMFS) and signs of clear-cutting (43% of PMFS); (iv) no signs of infrastructure compatible with licensed forestry (24% of PMFS); and (v) signs of exploitation prior to the licensing (19% of PMFS) and after the expiration of licensing (5%). Full article

Brazil and China have a long-term joint space based sensor program called China-Brazil Earth Resources Satellite (CBERS). The most recent satellite of this program (CBERS-4) was successfully launched on 7 December 2014. This work describes a complete procedure, along with the associated uncertainties, used to calculate the in-flight absolute calibration coefficients for the sensors Multispectral Camera (MUX) and Wide-Field Imager (WFI) on-board CBERS-4. Two absolute radiometric calibration techniques were applied: (i) reflectance-based approach and (ii) cross-calibration method. A specific site at Algodones Dunes region located in the southwestern portion of the United States of America was used as a reference surface. Radiometric ground and atmospheric measurements were carried out on 9 March 2015, when CBERS-4 passed over the region. In addition, a cross-calibration between both MUX and WFI on-board CBERS-4 and the Operational Land Imager (OLI) on-board Landsat-8 was performed using the Libya-4 Pseudo Invariant Calibration Site. The gain coefficients are now available: 1.68, 1.62, 1.59 and 1.42 for MUX and 0.379, 0.498, 0.360 and 0.351 for WFI spectral bands blue, green, red and NIR, respectively, in units of (W/(m²·sr·μm))/DN. These coefficients were determined with uncertainties lower than 3.5%. As a validation of these radiometric coefficients, cross-calibration was also undertaken. An evaluation of radiometric consistency was performed between the two instruments (MUX and WFI) on-board CBERS-4 and with the well calibrated Landsat-7 ETM+. Results show that the reflectance values match in all the analogous spectral bands within the specified calibration uncertainties. Full article

The Heat Energy Assessment Technologies (HEAT) project uses high-resolution airborne thermal imagery, Geographic Object-Based Image Analysis (GEOBIA), and a Geoweb environment to allow the residents of Calgary, Alberta, Canada to visualize the amount and location of waste heat leaving their houses, communities, and the city. To ensure the accuracy of these measures, the correct emissivity of roof materials needs to be known. However, roof material information is not readily available in the Canadian public domain. To overcome this challenge, a unique Volunteered Geographic Information (VGI) application was developed using Google Street View that engages citizens to classify the roof materials of single dwelling residences in a simple and intuitive manner. Since data credibility, quality, and accuracy are major concerns when using VGI, a private Multiple Listing Services (MLS) dataset was used for cross-verification. From May–November 2013, 1244 volunteers from 85 cities and 14 countries classified 1815 roofs in the study area. Results show (I) a 72% match between the VGI and MLS data; and (II) in the majority of cases, roofs with greater than, or equal to five contributions have the same material defined in both datasets. Additionally, this research meets new challenges to the GEOBIA community to incorporate existing GIS vector data within an object-based workflow and engages the public to provide volunteered information for urban objects from which new geo-intelligence is created in support of urban energy efficiency. Full article