Abstracts Track 2025

Area 1 - Data Acquisition and Processing

Nr:	59
Title:	NTC18 Standard (EC8): Topographical Conditions in the Siena Case Study – Application and Verification
Authors:	Mario Ariano, PierLorenzo Fantozzi and Dario Albarello
Abstract:	The Italian seismic code NTC18 (EC8) provides guidelines to compute expected amplification of the seismic ground motion induced by some morphological configurations. Specifically, two main 2D morphologies are identified as reference: cliffs and crests. Based on numerical simulations, an amplification factor (St) is computed depending on the steepness of the cliffs and aspect ratio of the crests. A critical aspect of these estimates is that the considered configurations are defined in terms of steepness angles and aspect ratios, without any scale indication. Moreover, the considered morphologies are very schematic, and this prevents their simple application in the natural context: in most case an expert judgement is necessary, and this makes the final estimates potentially controversial and difficult to validate based on empirical observations. To face this problem, in the frame of the PRIN project “SERENA”, an automated procedure has been developed to identify areas subjected to morphological amplification effects by following NTC18 (EC8) prescriptions, using the Digital Terrain Model. The proposed approach allows the use of topographic data at the highest resolution available. The procedure was applied to the area of the municipality of Siena, a representative example characterized by residential settlements located on specific morphologies, a predominantly homogeneous subsoil, and detailed historical documentation of past earthquakes. The aim is twofold: first, to provide consistent and reproducible estimates of the expected St values, verifying the amplification factors through direct comparison with results obtained from 2D numerical simulations; second to validate the regulatory guidelines related to reference configurations and the tables for the various components of seismic motion. The presentation will illustrate and discuss the results obtained, highlighting the implications for the application of NTC18 in real contexts.

Nr:	79
Title:	Interpolation Methods in Microscale Meterology
Authors:	Daniel Morrish
Abstract:	Monitoring air quality with low-cost sensors presents several advantages over traditional stations, especially in Aotearoa New Zealand. Traditional air quality monitoring stations are expensive to set up and operate, which limits their deployment to the most polluted areas. In contrast, low-cost sensors are more affordable, allowing for the creation of larger networks with better spatial coverage. This expanded network can provide better insight and help identify more representative locations for traditional stations. This study focuses on using geospatial analysis techniques to select sites for low-cost sensors to have better quality data for analysis. Additionally, we evaluate various interpolation methods to make informed cost-benefit decisions. These methods help in accurately measuring domestic heating emission hot spots in both time and space. The data for this study is collected from low-cost sensors deployed across Alexandra, a town in the Central Otago region in Aotearoa New Zealand’s South Island with a population of 6000, during the winter of 2023. This approach not only improves our understanding of air quality patterns but also supports better decision-making for future monitoring efforts.

Area 2 - Spatial Data Mining

Nr:	58
Title:	Analytical “Decisiveness” as a Robust Measure of Factor’s Absolute Importance in Determining Spatial Distribution
Authors:	Langping Li and Hengxing Lan
Abstract:	Measuring the importance of predisposing factors is fundamental for their weighting, selection, and interpretation in spatial analysis. Various methods have been developed to measure the importance of these factors in determining the spatial distribution of phenomena, broadly categorized into “relative importance” and “absolute importance” indices. Relative importance indices rely on comparisons among factors, meaning the importance of one factor is assessed in relation to others. In contrast, absolute importance indices evaluate each factor independently, based solely on its correlation with the phenomenon. Absolute importance has distinct advantages. First, it can be calculated using data from just one factor, whereas relative importance requires data from at least two factors. Second, absolute importance values change only if the specific factor’s data changes, while relative importance values may fluctuate with changes in other factors, complicating their interpretation. Third, absolute measurements allow consistent comparisons of the same factor across different cases or regions, a feature often lacking in relative evaluations. Despite these advantages, most existing studies focus on measuring relative importance, with absolute importance receiving limited attention. A systematic review of landslide studies revealed that only ~4% assessed absolute importance. Furthermore, among the limited methods for measuring absolute importance—such as the accountability & reliability method, AUC (area under the success rate curve), geographical detector method, and sensitivity index method—none are without inherent difficulties. This study proposes a novel analytical index called “decisiveness” for measuring the absolute importance of factors in determining spatial distributions. Decisiveness evaluates a factor’s ability to distinguish between susceptible and insusceptible areas. A factor with high decisiveness can effectively divide a target area into zones that are highly favorable or unfavorable for the occurrence of the phenomenon. Using the certainty factor method, favorability values for all grid cells in the target area are calculated and integrated to derive the decisiveness index, which ranges from 0 to 1. A higher decisiveness value indicates greater importance. A case study on landslide spatial distribution demonstrated that decisiveness effectively reflects the ability of predisposing factors to differentiate between landslide-susceptible and insusceptible areas. In comparison to existing methods, decisiveness introduces a more reliable and straightforward approach to measuring absolute importance. While existing methods provide valuable insights, they often involve complex calculations and inherent limitations. Decisiveness eliminates these complications, offering a robust framework for evaluating factor importance. With this novel index, it is possible to not only evaluate the importance of different factors in a single application but also compare the importance of the same factor across diverse cases. This capability is especially valuable in understanding how factors contribute to spatial phenomena in varying regions or under different conditions. By addressing the limitations of previous approaches, decisiveness provides a significant advancement in spatial analysis, offering practical and accurate tools for decision-making and research.

Area 3 - Remote Sensing

Nr:	75
Title:	SAR-Derived Superficial Soil Moisture Estimation at the Field Scale Over an Agricultural Area: A Study Case Using L-Band Data
Authors:	Giulia Graldi, Filippo Bocchino, Alireza Hamoudzadeh, Lorenza Ranaldi, Deodato Tapete, Alessandro Ursi, Maria Virelli, Patrizia Sacco, Valeria Belloni, Roberta Ravanelli and Mattia Crespi
Abstract:	Among the four types of drought –meteorological, hydrological, agricultural and socioeconomic– the agricultural drought is the most challenging to monitor, as its effects can only be detected months after its manifestation (Boken, et al., 2005). Agricultural drought is defined as a shortage of water availability in the soil, and a subsequent loss of yield. Soil moisture is thus the main parameter for monitoring this phenomenon, but retrieving information on its spatial and temporal distribution is challenging itself, given its great variability. The spatial and temporal variability of soil moisture can be studied by using active and passive microwave Earth Observation data. Historically, soil moisture products have been available at resolutions ranging from tens of kilometers to sub-kilometric scales. In between data acquired with active microwave technologies, Synthetic Aperture Radar (SAR) data are commonly used for soil moisture analysis at high spatial scales, such as the field scale, given their high spatial resolution. However, when estimating superficial soil moisture from SAR data at such scales, the assumptions made at coarser resolutions may no longer hold. For example, at the field scale, variations in the soil roughness conditions due to the agricultural practices and the seasonal effect of the vegetation should be considered. In the frame of the GRAW project, funded by the Italian Space Agency (ASI), this contribution presents a study case on the use of L-band data of the SAOCOM mission (Azcueta,et al.,2021) for retrieving surface soil moisture at the field scale. The case study is located in an agricultural area in Spain where surface soil moisture measurements from the REMEDHUS ( González-Zamora et al. 2019) network are available as reference data. Given the presence of a not negligible geolocation error in SAOCOM data in respect to the spatial scale of analysis (Recchia et al., 2022), a proper preprocessing workflow was applied to the data for correcting their geolocation. After that, the variations in the soil roughness conditions were identified through an anomaly detection approach (Zhu et al., 2019), while the crop types were derived from the EUCROPMAP 2022 map (European Commission, Joint Research Centre, 2022), for modelling the seasonal vegetation contribution.

Nr:	80
Title:	Refined Satellite Altimetry from GEDI and SWOT for Enhanced Monitoring of Inland Water Levels in Italy
Authors:	Alireza Hamoudzadeh, Lorenza Ranaldi, Filippo Bocchino, Roberta Ravanelli, Giulia Graldi, Valeria Belloni, Deodato Tapete, Alessandro Ursi, Maria Virelli, Patrizia Sacco and Mattia Crespi
Abstract:	Freshwater sources and inland water bodies are crucial for agriculture and human water consumption, necessitating effective monitoring to evaluate the impacts of climate change and human activities. Advances in remote sensing technology now facilitate cost-effective, long-term tracking of surface water levels [1,2]. This study enhances continuous water level time series by integrating satellite altimetry data from the Global Ecosystem Dynamics Investigation (GEDI) and Surface Water and Ocean Topography (SWOT) missions, improving accuracy, precision, and revisit frequency. GEDI, a LiDAR altimeter onboard the International Space Station, delivers high-resolution measurements (25m footprint) within latitudes 51.6°N to 51.6°S. Data from March 2019 to March 2023, available on Google Earth Engine (GEE), was analyzed for Italian lakes (2019–2022) [2]. An outlier detection workflow leveraging GEDI data and a 3NMAD-based repetitive test improved the precision of the retrieved water levels. For Northern Italian lakes, GEDI achieved an intrinsic precision of 0.11m and sub-10cm precision for smaller lakes in Lazio. SWOT, operational since April 2023, employs a Ka-band Radar Interferometer to observe 86% of Earth’s surface with 100m pixel resolution and a 21-day revisit interval [1]. For Northern Italian and Swiss lakes, SWOT-retrieved water levels showed a 92% correlation with gauge measurements and a precision of approximately 0.06m. For Central Italian ungauged lakes, the water level spatial NMAD was under 10cm. In summary, this study shows how GEDI and SWOT can provide complementary and reliable solutions for enhancing global inland water level monitoring.

Nr:	81
Title:	Preliminary Assessment of Agricultural Drought Using PRISMA Data: Insights Across Diverse Crop Types in Italy
Authors:	Filippo Bocchino, Giulia Graldi, Alireza Hamoudzadeh, Lorenza Ranaldi, Deodato Tapete, Alessandro Ursi, Maria Virelli, Patrizia Sacco, Antonio Denaro, Laura Rosatelli, Camillo Zaccarini, Valeria Belloni, Roberta Ravanelli and Mattia Crespi
Abstract:	In recent years, severe drought conditions affected many areas in Italy, impacting the agricultural sector and leading to substantial yield losses. Agricultural drought is influenced not only by meteorological conditions but also by their effects on vegetation, which depend on the phenological stage of the crops and soil moisture levels. To account for all these variables, accurate monitoring of agricultural drought using remote sensing techniques is crucial for evaluating both the severity and the impacts of droughts in both quantitative and qualitative terms. Among remote sensing data, hyperspectral satellite imagery offers a promising opportunity for monitoring drought impacts on crops, thanks to its high spectral resolution, which enables sensitivity to specific parameters associated with vegetation stress. In this context, hyperspectral data from the Italian Space Agency (ASI)’s Hyperspectral Precursor of the Application Mission (PRISMA) sensor represents a valuable resource to analyze spectral signatures of crops affected by drought. The PRISMA sensor captures hyperspectral imagery across 240 contiguous bands spanning wavelengths from 400 to 2500 nm, with a spectral resolution of less than 12 nm and a spatial resolution of 30 meters. Additionally, a 5-meter panchromatic image is provided, facilitating the association of spectral signatures with on-the-ground features. In this study, results derived from PRISMA imagery were compared with in-situ data provided by the Institute of Services for Agricultural and Food Market (ISMEA). The in-situ dataset consists of drought damage assessments for various agricultural fields collected by ISMEA across Italy during an experimental campaign conducted in 2022. For each field, drought damage percentages were estimated using standardized, field-based procedures. Given that the damage data are field-specific, and the investigated fields are relatively small, spanning only a few thousand square meters, it was essential to address the well-known geolocation errors of PRISMA imagery. To ensure accurate analysis at this fine spatial scale, an innovative terrain-independent correction method was applied. Spectral bands significantly affected by atmospheric water vapor and CO₂ absorption were removed during pre-processing. Subsequently, spectral signatures at plot level were computed using the spatial median, enabling the comparison of spectral characteristics across fields with the same crop type but varying levels of drought damage. One case study focused on durum wheat fields in Foggia (Southern Italy) during the 2022 season. Two cloud-free PRISMA images (29/04/2022 and 14/06/2022) were analyzed, revealing differences in the 700–1100 nm spectral range between fields with 0% and 36% damage. These bands are associated with leaf structure and water content, parameters influenced by drought stress. Although this study is constrained by the limited number of analysed fields (which in turn was constrained by the spatial distribution of the survey sample and related collection date of the ISMEA damage records), the preliminary results underscore the potential of PRISMA imagery for detecting and monitoring the impacts of drought on crops.

Area 4 - Domain Applications

Nr:	63
Title:	Engaging and Effective Hands-on Skill Training of Undergraduate Students in Mobile and Web GIS, High-Precision GPS, Drone RS, and GeoAI Through Asset Inventory Applied, Civic Engagement Projects
Authors:	Wenjie Sun
Abstract:	Service learning, also known as “civic engagement”, is a form of experiential education, in which students use academic knowledge and skills to address genuine community needs and solve real-world problems. Students typically work in small groups and go through the entire life cycle of a real-world GIS project, from meeting with “clients” and prioritizing their needs, to project management, from data collection, acquisition, mapping, and analysis, to the final product delivery. The captivating nature of applied projects makes them an ideal means of teaching to engage students in active and interactive learning. During this process, it is important to help them visualize the “big picture” and appreciate the challenging and exciting moments of solving a real-life problem. These projects keep students well-motivated and let them see the relevance of GIS skills acquired from the classroom in practical scenarios. Moreover, the skills, training, and experiences from these applied research opportunities have effectively connected students to fields of work and assisted them in advancing to further career opportunities in GIS. The author will share her experiences and reflections from 4 civic engagement projects in her Applied GIS Projects class from Fall 2024. In these projects, students worked for one “internal client” (Carthage College Facility Management) and one “external client” (Hawthorn Hollow Nature Sanctuary and Arboretum, a local nonprofit conservation agency in Kenosha, WI), addressing their respective asset inventory needs. In the Hawthorn Hollow project, the team conducted field data collection using high-precision GPS (Trimble DA2 with 1cm subscription service) and mobile GIS app (ESRI ArcGIS Survey123), and then created a visitor map for hard copy handout and an interactive web app (which can be embedded on the Hawthorn Hollow website and accessed on visitor smartphones via a QR code) using these data to show informational and educational signage, a trail, and different tree species regions at the arboretum. In two Facility Management projects, the teams utilized similar GIS and GPS skills and setup for building rooftop utility equipment mapping, and tree mapping along campus drive, both of which will be used by office managers and field crew for maintenance purposes. In the third Facility Management project, the team explored and calibrated vehicle and parking spot detection GEO AI models using high-resolution RGB and multi-spectral drone imagery and ArcGIS Pro’s deep learning tools to assess Carthage parking lot utilization by day of week and time of day. The accuracy rate of detection is now high enough to proceed to the implementation phase, in collaboration with a separate Drones class to acquire and run the model on the entire intended time series of drone imagery. Students gained technical skills with high-precision GPS, field data collection, mobile GIS, cartography, and web app creation; cultivated critical “soft skills” of communicating with non-GIS professionals, working in a collaborative group environment, time management, problem-solving, giving oral presentations, and composing written reports. All these skills are highly sought after when students compete for internships, jobs, and graduate schools. A tangible and otherwise expensive need is met for the college/external nonprofit organization and a long-term partnership is formed for yet more project opportunities for future students. It’s a win-win for all!.

Nr:	74
Title:	Simulating Winter Maintenance Efforts Using a Geographically Weighted Regression Model
Authors:	Nafiseh Mohammadi and Alex Klein-Paste
Abstract:	Many countries need to perform “Winter Road Maintenance” (WRM) to counteract snow/ice conditions on roads and provide mobility and safety in winter traffic. However, due to the major cost and environmental impacts of WRM, many road administrators seek ways to evaluate WRM’s efficiency and optimize it. WRM is complex per se as it results from a multilevel decision-making procedure including strategic, tactical, and operational decisions. The simulation of WRM services might facilitate this optimization as it provides a holistic approach to investigating and supporting all the decision levels. In response, Norwegian Public Road Administration (NPRA) initiated the R&D project “WinterSim”, aiming to create the knowledge for a GIS-based simulation tool quantifying efforts (maintenance operations) required to maintain roads at specific levels of services. In an earlier work, the authors proposed a regression-based model entitled “Effort Model” as a potential computational core for the simulation tool (Mohammadi et al., 2007). Considering a spatial-statistical approach in data acquisition and data analysis, the model incorporates effective weather variables (like snow days, cold days, etc.) and non-weather variables (like maintenance class, annual traffic, road width, etc.) and returns the number of conducted operations at a given location over Norway’s state road network. It was calibrated with WRM production data, roads’ geometric and topographic characteristics, and historical weather data for the whole country. It consists of three sub-models, in the form of Generalized Linear Regression, estimating number of plowing, salting, and combined plowing-salting operations over a given road segment. The model could estimate number of operations with 70% accuracy but showed limitations in capturing the spatial variability of operations due to Norway’s diverse climate and topography. To upgrade the Effort Model, we have adopted the Multiscale Geographically Weighted Regression (MGWR) method, recreating our three sub-models based on that. Preliminary results show a significant improvement in estimation accuracy, ranging from 10% in the plowing sub-model to 20% in the salting sub-model. In addition to more accurate estimations, MGWR provides valuable by-products including coefficient raster layers and model diagnostics tables. A key advantage of MGWR lies in its ability to assign different spatial scales to non-stationary variables like ours. It allows the sub-models to capture the varying influence of the variables across different locations. This information provided by the coefficient surfaces is very helpful for WRM simulation because the relationships between needed operations and effective factors are not uniform across the entire country. Simultaneous analysis of a variable’s coefficient raster and its significance level provided by the diagnostic tables gives valuable insights into how that factor varies across the study area and to what extent it influences the operation demands. This understanding can allow the simulation tool to guide and shape policy decisions related to each variable and each operation type. Prediction of required efforts in an unseen period, the next 3-5 years in the case of Norway, is the future work for enhancing the simulation tool’s applicability.

Nr:	76
Title:	Walkable Neighbourhoods: GIS-Based Assessment and Cartographic Visualization from Multiple Pedestrian Perspectives
Authors:	Ana Kuveždić Divjak, Karlo Kević and Mateo Bošnjak
Abstract:	Urbanization is rapidly increasing, placing significant demands on city infrastructure and quality of life. This study investigates the concept of walkability and applies a GIS-based methodology for assessing walkability in the City of Zagreb, Croatia, utilizing open data sources. This research incorporates three pedestrian perspectives – residents, employees, and visitors (tourists) – to provide a comprehensive evaluation of pedestrian accessibility across the city’s districts. Walkability was quantified using indicators such as pedestrian network, population density, green and water areas, and points of interest. The assessment results were visualized through an interactive web-based thematic map, enabling users to browse, compare, and interpret walkability data effectively. Findings indicate that central districts like Donji Grad exhibit high walkability indices, whereas peripheral areas such as Podsused-Vrapče and Sesvete show lower levels. The study highlights the importance of walkability in sustainable urban planning and offers actionable insights for improving pedestrian infrastructure. The interactive map serves as a valuable tool for urban planners and policymakers to monitor and enhance walkability, contributing to the development of more sustainable and resilient urban environments.