Feed on
Posts
Comments

Deadline extended to 21.12.2020
18th International Conference on Information Systems for Crisis Response and Management (ISCRAM 2021)

May 23-26, 2021, VIRTUAL, Interactive ONLINE EVENT
https://www.drrm.fralinlifesci.vt.edu/iscram2021/

Track: Geospatial Technologies and Geographic Information Science for Crisis Management (GIS)
https://www.drrm.fralinlifesci.vt.edu/iscram2021/files/CFP/ISCRAM2021-Track10-Geospatial_Technologies

New Deadline for paper submissions: December 21, 2020

Track Description

With crisis and hazardous events being an “inherently spatial” problem, geospatial information and technologies have been widely employed for supporting disaster and crisis management. This was further highlighted during the response to the 2020 Coronavirus pandemic, which is relying extensively on spatial analysis for managing  the virus dissemination pathways and fighting against the virus propagation. Therefore, geospatial methods and tools – such as Spatial Decision Support Systems (SDSS), Geographic Information Systems (GIS) architectures, Volunteered Geographic Information (VGI), spatial databases, spatial-temporal methods, as well as geovisual analytics technologies –  have a great potential to contribute to, understand the geospatial characteristics of a crisis, estimate damaged areas, define evacuation routes, and plan resource distribution. Collaborative platforms like OpenStreetMap (OSM) have also been employed to support disaster management (e.g., in near real-time mapping). Nevertheless, all these geospatial big data pose new challenges for not only geospatial data visualization, but also data modeling and analysis; existing technologies, methodologies, and approaches now have to deal with data shared in various formats, different velocities, and uncertainties. Furthermore, new issues have been also emerging in urban computing and smart cities for making communities more resilient against disasters. In line with this year’s conference theme, the GIS Track particularly welcomes submissions addressing aspects of geospatial information in disaster risk and crisis research, and how this geospatial information should embrace the interdisciplinary nature of crisis situations. This includes exploring bridges between geospatial data science methods and tools and other related fields, including (but not limited to): computing disciplines (e.g. AI and machine learning), social sciences (e.g.  socio-spatial aspects of risk and resilience, community resilience, participation and governance) and humanities (e.g. spatial humanities and spatial digital arts). We seek conceptual, theoretical, technological, methodological, empirical contributions, as well as research papers employing different methodologies, e.g., design-oriented research, case studies, and action research. Solid student contributions are welcome.

Track topics are therefore focused on, but not limited to the following list:

– Geospatial data analytics for crisis management
– Location-based services and technologies  for crisis management
– Geospatial ontology for crisis management
– Geospatial big data in the context of disaster and crisis management
– Geospatial linked data for crisis management
– Spatially explicit machine learning and Artificial Intelligence for crisis management
– Urban computing and geospatial aspects of smart cities for crisis management
– Spatial Decision Support Systems for crisis management
– Individual-centric geospatial information
– Remote sensing for crisis management
– Geospatial intelligence for crisis management
– Spatial data management for crisis management
– Spatial data infrastructure for crisis management
– Geovisual analytics for crisis management
– Spatial-temporal modeling in disaster and crisis context
– Crisis mapping and geovisualization
– Collaborative disaster mapping, citizen participation
– Public policies and governance for geospatial information
– Case studies of geospatial analysis/tools during a pandemic situation
– Empirical case studies

Important Dates

In acknowledgement of the ongoing challenges posed by the COVID-19 pandemic, the original submission deadlines for ISCRAM 2021 have been extended by several weeks. No other changes are expected to be made to the paper submission and review process.


Revised submission deadline for CoRe papers: December 21, 2020

Revised notification of decision for CoRe papers: January 31, 2021

Revised submission deadline for WiP and Practitioner papers: February 14, 2021

Revised notification of decision for WiP and Practitioner papers: March 15, 2021

The 2021 ISCRAM conference invites three types of paper submissions:

  • CoRe - Completed Research (from 4000 to 8000 words).
  • WiP - Work In Progress (from 3000 to 6000 words).
  • Practitioner (from 500 to 3000 words).

https://www.drrm.fralinlifesci.vt.edu/iscram2021/call-papers.php

IMPORTANT DATES:

- Submission deadline for CoRe papers: December 6, 2020

- Notification of decision for CoRe papers: January 17, 2021

- Submission deadline for WiP and Practitioner papers: January 31, 2021

- Notification of decision for WiP and Practitioner papers: February 28, 2021

* TRACK CHAIRS:

- Professor João Porto de Albuquerque*, j.porto@warwick.ac.uk University of Warwick
- Alexander Zipf zipf@uni-heidelberg.de University of Heidelberg
- Flávio Horita flavio.horita@ufabc.edu.br Federal University of ABC
- Michael A. Erskine michael.erskine@mtsu.edu Middle Tennessee State University

Klimawandel und Biodiversitätsverlust – das unglückliche Zusammenspiel von Zeit und Raum

lautet der Titel des nächsten Vortrags in der Vortragsreihe „Klimawandel – Herausforderungen für die Menschheit“ der Heidelberger Gesellschaft für Geographie e.V. (HGG).

Prof. Dr. Marcus Koch (Universität Heidelberg) trägt hierzu am Dienstag 08.12.2020 um 19 Uhr vor:

Der Prozess des Verlustes der biologischen Vielfalt läuft ungebremst auf globalem Maßstab; und die derzeitigen Aussterberaten gehören zu den höchsten, die in den letzten 500 Millionen Jahren geschätzt wurden. Parallel verändert sich das Klima mit rasanter Dynamik wie es auch während der vielen Wechsel von Eis- und Warmzeiten der letzten 2,5 Millionen Jahre gewesen ist.
Was ist anders an den Dynamiken, die nun ausgerechnet jetzt die Arten aussterben lassen?

Die Veranstaltungsreihe findet im Wintersemester 2020/21 online statt.
Zugang hierzu haben Mitglieder der HGG und angemeldete Schulklassen. Der Zugangscode wird den Mitgliedern und Neumitgliedern per Mail oder per Post zugeschickt. Für das Wintersemester 2020/21 bieten wir für Neumitglieder einen reduzierten Mitgliedsbeitrag in Höhe von 6 € für Studierende und 12 € für vollzahlende Mitglieder an. Das Anmeldeformular finden Sie zum Download auf der HGG-Homepage oder Sie können es per Mail unter <hgg@UNI-HEIDELBERG.DE anfordern.

Die Vorträge dieses Semester:
Der Klimawandel und seine Konsequenzen für die Nutzungssysteme stehen seit einigen Jahrzehnten im Zentrum der wissenschaftlichen Debatte und medialer Berichterstattung. Nicht erst seit den Klimastreiks der Fridays-for-Future-Bewegung und den wiederkehrenden Berichten des IPCC (Intergovernmental Panel on Climate Change) bilden die Probleme an der Schnittstelle zwischen Klimawandel, Wasserverfügbarkeit und Nachhaltigkeit wichtige Aufgaben einer verantwortungsbewussten und zukunftsorientierten Geographie.
HGG

This new video by the 3DGeo group presents the challenges of 3D Earth observation and our advances in 4D change analysis in the frame of the Auto3Dscapes project:

Direct link to the video: https://youtu.be/Fdwq-Cp0mFY

Many thanks to Claudia Denis and David Jäger for helping to realize the video! We are also very happy about the fruitful ongoing collaboration with the CoastScan team of TU Delft and the environmental research station Schneefernerhaus!

Interested in the research? The method of spatiotemporal segmentation is presented in this paper:

Anders, K., Winiwarter, L., Lindenbergh, R., Williams, J. G., Vos, S. E., & Höfle, B. (2020). 4D objects-by-change: Spatiotemporal segmentation of geomorphic surface change from LiDAR time series. ISPRS Journal of Photogrammetry and Remote Sensing, 159, 352-363. doi:10.1016/j.isprsjprs.2019.11.025

Stay updated on Auto3Dscapes research via this News Blog, ResearchGate, and Twitter under #Auto3Dscapes! And if you enjoy videos about research projects, visit the 3DGeo YouTube channel, with recent videos about SYSSIFOSS and AHK-4D.

The PhD project Auto3Dscapes is supported by the Heidelberg Graduate School of Mathematical and Computational Methods for the Sciences (HGS MathComp), founded by DFG grant GSC 220 in the German Universities Excellence Initiative.

On December 8th at 6.30 pm the next international mapathon organized by the disastermappers heidelberg takes place! This time we map together with MAMAPA, an integration project for migrants and refugees from Mannheim, CartONG, LECLARA Larabanga and OSM Ghana.

We will map HOT-Tasks 8839 and 9928, the former including the city of Wa in northwestern Ghana, the latter in Accra.

You can join the event via https://audimax.heiconf.uni-heidelberg.de/mzt4-vfgk-hy4y-djfp

The motivation:

Up to now, not all map data is available everywhere in the world, which often limits the possibilities for disaster prevention and preparedness, needs assessments for relief organisations and many other purposes. So be part of the worldwide OSM mapping community! Everyone is welcome – no matter if you are an experienced mapper or a newcomer. A prepared tutorial provides a suitable introduction for beginners, so that no previous knowledge is required.

What you need:

Prepare a headset to be able to swap and a mouse which will make mapping easier. Unfortunately we can’t provide you with snacks and drinks at the moment – so remember to prepare something; then the event will be even more fun!

The partners for this mapathon:

The MAMAPA project is conceived both to aid humanitarian cartography and, at the same time, to support the integration of “new arrivers” (migrants, refugees) in the Rhein-Neckar metropolitan region. During the project mapathons, locals and recent arrivers build tandems at each workstation and collaborate on the mapping task. Ample opportunity is provided for contact and conversation in the few hours spent together. (For further information, please visit https://mamapa.org or mannheimermapathons)

CartONG is a NGO located in Chambéry, France, and provides cartographic and related expertise and services ( e.g. data management, mobile data collection) for humanitarian organizations worldwide. Major clients include Doctors without Borders and the UNHCR. When these organisations are in the field, CartONG works with them to provide maps as well as to generate and manage data essential for mission success. (For further information, please visit https://cartong.orghttps://cartong.org/volunteering)

OSM Ghana, a non-profit organisation, is dedicated to completing the mapping of the country as well as providing training in the use of free and open source geographic tools for humanitarian and developmental goals. The group, active in the FOSS Community, has been involved in the recent past in numerous projects including Open Cities Accra, mapping with LECLARA (Larabanga, Ghana) for sustainable tourism development and participation in the organisation of an African Technical Exchange in Accra for the Secondary Cities project. (For further information, please visit https://osmghana.org )

LECLARA Larabanga is an organization of local activists in the town of Larabanga, in the Savannah Region of Ghana. They pursue a number of goals important for development in Larabanga, both cultural and educational. Since 2006, (together with the current Project Director of MAMAPA), much was accomplished for primary and secondary schooling (Project “LIEI”), with a particular emphasis on offering educational opportunities for young girls in the village.Located a few kilometers from Mole National Park and home to a Mosque and Koran dating from the 15th/16th centuries, Larabanga has exceptional touristic potential. Since 2016, to spur local development, LECLARA has been involved in building an infrastructure for sustainable tourism. As well, local youth have found opportunities as trained guides.Mapping is essential to this effort. With support from CartONG and OSM Ghana and as the recipient of a HOT Microgrant (2018-19) Larabanga and the surrounding area are now completely mapped. At the same time, workshops on OSM mapping techniques and mobile data collection have been held locally for members of LECLARA. While tourism there (as everywhere else) is currently “on hold”, up-to-date maps are as well an important tool for defending the health and safety of the population both during the current pandemic and into the future. (For further information, please visit https://visitlarabanga.org or LIEI (pre-2016) https://liei-ghana.org .)

Further upcoming events by disastermappers heidelberg.

  • December 8th, 2020, MaMaPa: International Mapathon with Migrants and CartONG and OSM Ghana
  • December 15th, 2020, Christmas Workshop
  • January 19th, 2021, GIS Workshop with Viva Con Agua – Access to drinking water
  • February 23rd, 2021, Talk and Mapathon – Access to Healthcare Infrastructure with PD Carsten Butsch

3rd CALL FOR PAPERS –  18th International Conference on Information Systems for Crisis Response and Management (ISCRAM 2021)

May 23-26, 2021, Virginia, USA – https://www.drrm.fralinlifesci.vt.edu/iscram2021/ Virginia Tech

Track: Geospatial Technologies and Geographic Information Science for Crisis Management (GIS)
https://www.drrm.fralinlifesci.vt.edu/iscram2021/files/CFP/ISCRAM2021-Track10-Geospatial_Technologies

Deadline for paper submissions: December 6, 2020

* Track Description

With crisis and hazardous events being an “inherently spatial” problem, geospatial information and technologies have been widely employed for supporting disaster and crisis management. This was further highlighted during the response to the 2020 Coronavirus pandemic, which is relying extensively on spatial analysis for managing  the virus dissemination pathways and fighting against the virus propagation. Therefore, geospatial methods and tools – such as Spatial Decision Support Systems (SDSS), Geographic Information Systems (GIS) architectures, Volunteered Geographic Information (VGI), spatial databases, spatial-temporal methods, as well as geovisual analytics technologies –  have a great potential to contribute to, understand the geospatial characteristics of a crisis, estimate damaged areas, define evacuation routes, and plan resource distribution. Collaborative platforms like OpenStreetMap (OSM) have also been employed to support disaster management (e.g., in near real-time mapping). Nevertheless, all these geospatial big data pose new challenges for not only geospatial data visualization, but also data modeling and analysis; existing technologies, methodologies, and approaches now have to deal with data shared in various formats, different velocities, and uncertainties. Furthermore, new issues have been also emerging in urban computing and smart cities for making communities more resilient against disasters. In line with this year’s conference theme, the GIS Track particularly welcomes submissions addressing aspects of geospatial information in disaster risk and crisis research, and how this geospatial information should embrace the interdisciplinary nature of crisis situations. This includes exploring bridges between geospatial data science methods and tools and other related fields, including (but not limited to): computing disciplines (e.g. AI and machine learning), social sciences (e.g.  socio-spatial aspects of risk and resilience, community resilience, participation and governance) and humanities (e.g. spatial humanities and spatial digital arts). We seek conceptual, theoretical, technological, methodological, empirical contributions, as well as research papers employing different methodologies, e.g., design-oriented research, case studies, and action research. Solid student contributions are welcome.

Track topics are therefore focused on, but not limited to the following list:

– Geospatial data analytics for crisis management
– Location-based services and technologies  for crisis management
– Geospatial ontology for crisis management
– Geospatial big data in the context of disaster and crisis management
– Geospatial linked data for crisis management
– Spatially explicit machine learning and Artificial Intelligence for crisis management
– Urban computing and geospatial aspects of smart cities for crisis management
– Spatial Decision Support Systems for crisis management
– Individual-centric geospatial information
– Remote sensing for crisis management
– Geospatial intelligence for crisis management
– Spatial data management for crisis management
– Spatial data infrastructure for crisis management
– Geovisual analytics for crisis management
– Spatial-temporal modeling in disaster and crisis context
– Crisis mapping and geovisualization
– Collaborative disaster mapping, citizen participation
– Public policies and governance for geospatial information
– Case studies of geospatial analysis/tools during a pandemic situation
– Empirical case studies

* Important Dates

Full research and insight papers:
– Submission deadline: December 6, 2020
– Decision notification: January 17, 2021

Short (WiPe) papers and Practitioner papers:
– Submission deadline: January 31, 2021
– Decision notification: February 28, 2021

* Paper submission guidelines

https://www.drrm.fralinlifesci.vt.edu/iscram2021/call-papers.php

IMPORTANT DATES:

- Submission deadline for CoRe papers: December 6, 2020

- Notification of decision for CoRe papers: January 17, 2021

- Submission deadline for WiP and Practitioner papers: January 31, 2021

- Notification of decision for WiP and Practitioner papers: February 28, 2021

* TRACK CHAIRS:

- Professor João Porto de Albuquerque*, j.porto@warwick.ac.uk University of Warwick
- Alexander Zipf zipf@uni-heidelberg.de University of Heidelberg
- Flávio Horita flavio.horita@ufabc.edu.br Federal University of ABC
- Michael A. Erskine michael.erskine@mtsu.edu Middle Tennessee State University

Next week the HOT Summit will take place. The conference takes already place for the sixth time and it is the fifth consecutive time that we from HeiGIT/GIScience Heidelberg do contribute a session. This years topic is 10 Years of Humanitarian OpenStreetMap: The Past, Present, and Future of Humanitarian Mapping. It will be a one day conference in conjunction with the Understand Risk conference and happens on Friday 4th December. The fully virtual conference is organized in three blocks across several time zones, to reflect a truly global event.

Benjamin Herfort from HeiGIT will participate together with Hannah Ker (MapAction) and Geoffrey Kateregga (HOT) in an extended dialogue session towards the sustainability of community engagement and data production in humanitarian mapping activities. In our session we want to explore what humanitarian mapping in OSM looks like and if it is serving the community in a sustainable way. You can take a look at our slides already here and get prepared with questions and comments in advance.

Marcel Reinmuth (HeiGIT) will contribute a lightning talk: OpenStreetMap for healthcare access.
In which he will present latest results on using OSM data for healthcare access estimations and a brief overview on what is missing in healthcare data in OSM. Here you can see the slides and links to further resources.

Map of Overall and Humanitarian Building Mapping in OSM since 2008

Map of Overall and Humanitarian Building Mapping in OSM since 2008

Related work and Links:

What is the idea behind the Notebook?

In the case of an emergency (e.g. floods, earthquakes, political crisis) it is important to know where the health facilities are located. Furthermore, it is important to identify which cities/districts have a reduced or no access at all to health facilities before an emergency. Many countries still posses a centralized health system, making the tasks of the emergency workers even more difficult. In order to get accurate information from the health facilities, we retrieve the latest data that is available in the OpenStreetMap database. Two years ago, we wrote a Blogpost where we introduced a Jupyter Notebook that used our openrouteservice (ORS) Isochrones API in order to determine the access to health facilities in Madagascar. The notebook has been improved and updated and is now ready to be used with the latest version of the ORS API.

Check out the new interactive version of the notebook in nbviewer.

The biggest improvement of this new Jupyter Notebook is the automation and globalization of the analysis. In other words, the user just has to insert the ISO-3 code and the name of the desired country at the begin of the script. For example, if we want to make an analysis  for Bolivia, we just need to insert the ISO-3 code (”BOL”) and the oficial name (”Bolivia”). This is a big improvement because the user doesn’t have to get his own data (e.g. shapefiles).  By entering the ISO-3 code, the user automatically downloads a geojson file with the administrative boundaries (admin_level 2), a geojson with the points of the health facilities from the ohsome API and finally, a population raster from worldpop.org.

Another important upgrade is the implementation of the Python module rasterstats. The module replaces an old script that was used for the statistics and it includes a function called zonal statistics.  The function returns the statistics of the raster. This allows us to count and sum up the population for each district or isochrone in an easy and sophisticated manner. Lastly, the results are displayed in a choropleth map with multiple layers. We implemented GeoPandas and Folium in this last part.

Analysis of two countries - Comparing Health Care Access for Azerbaijan and Czech Republic

Workflow

Let’s have a look at the script. In the following examples, we will apply the notebook in the Republic of Azerbaijan and the Czech Republic. The first step is to enter the ISO-3 code and the name of the country  that we want to analyse. The script will automatically download the boundaries, the health facilities (nodes) and the “World Population” raster.

After this step, the analysis begins. The first task of the analysis is to create a districts dictionary that will save, for example, the population data from the raster. The overview map will show the user how the health facilities are distributed in the country.

Another important step is to calculate the access to health facilities per district. For this step, the script grabs the isochrones that we got from the ORS API. Combined with the population data stored in the raster, we are then able to calculate how many persons have access to a health facility on a district level.

Finally, the script saves the output as a geojson file. In order to check if the data has been written properly, the script displays the dictionary that was created at the beginning as a Pandas DataFrame. The final choropleth map has three layers. It allows the user to switch between the population count and the percent of the population in each district that is able to reach the health facilities via car or foot in a certain amount of time. The cursor displays the name of the district and the data.

Fig 1. ISO-3 code and name from Azerbaijan

Fig 2. Administrative Boundaries from Azerbaijan and health facilities clusters

Fig 3. The final results from Azerbaijan displayed as a Pandas Dataframe

Results

The last step of the script is to display the results of the analysis in an interactive choropleth map with three layers. For example, we can observe that in Azerbaijan (see Fig. 4), the persons living in the west and in the capital, Kabu, have a better access to the health facilities.

If we make the analysis for the Czech Republic, we get the choropleth map depicted in Fig 5. Comparing the result in Azerbaijan to the result in the Czech Republic, we could assume that the health facilities in the Czech Republic are more evenly distributed. As a result of this, the percent of people that have access to a health facility in a district increases. This a very basic comparison that is easy to achieve with this new notebook.

It’s important to underline that this script has still some limitations. The topography and relief of a country (e.g. a mountain range) are not taken into account in the ORS API. We are looking forward to improve this aspect and build a notebook that is even closer to the reality.

If you have thoughts or ideas how we can better implement this notebook in order to provide an even more realistic result, don’t hesitate to contact us here: info@heigit.org

Fig 4. Health Care Access for Azerbaijan

Fig 5. Health Care Access for the Czech Republic

Related Work and Literature

Gullies are landforms with specific patterns of shape, topography, hydrology, vegetation, and soil characteristics. Remote sensing products (TanDEM-X, Sentinel-1 and Sentinel-2) serve as inputs into an iterative algorithm, initialized using a micro-mapping simulation as training data, to map gullies in the northwestern of Namibia. A Random Forest Classifier examines pixels with similar characteristics in a pool of unlabeled data and gully objects are detected where high densities of gully pixels are enclosed by an alpha shape. Gully objects are used in subsequent iterations following a mechanism where the algorithm uses the most reliable pixels as gully training samples. The gully class continuously grows until an optimal scenario in terms of accuracy is achieved. Results are benchmarked with manually tagged gullies (initial gully labelled area <0.3% of the total study area) in two different watersheds (408 km2 and 302 km2, respectively) yielding total accuracies of >98%, with 60% in the gully class, Cohen Kappa >0.5, Matthews Correlation Coefficient >0.5, and ROC Area Under the Curve >0.89. Hence, our method outlines gullies keeping low false-positive rates while the classification quality has a good balance for the two classes (Gully/NoGully). Results show the most significant gully descriptors as the high temporal radar signal coherence (22.4%) and the low temporal variability in Normalized Difference Vegetation Index (NDVI) (21.8%).

Classification results for three gully probability classes (high, medium, low) for WS2. a) shows the classification results and b) the realgully locations used  for  validation  (black  dots),  initial  gully  label  (green  polygons)  used  for training.  c-e)  Detailed  view  of  gully  mapping  underlain  withMicrosoft® Bing™ MapsAerialimagery.

Classification results for three gully probability classes (high, medium, low) for WS2. a) shows the classification results and b) the real gully locations used for validation (black dots), initial gully label (green polygons) used for training. c-e) Detailed view of gully mapping underlain with Microsoft® Bing™ Maps Aerial imagery.

This research builds on previous studies to face the challenge of identifying and outlining gully-affected areas with a shortage of training data using global datasets, which are then transferable to other large (semi-)arid regions.

Find all details in the full papers:

Orti, M.V., Winiwarter, L., Corral-Pazos-de-Provens, E., Williams, J.G., Bubenzer, O. & Höfle, B. (2020): Use of TanDEM-X and Sentinel products to derive gully activity maps in Kunene Region (Namibia) based on automatic iterative Random Forest approach. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

Orti, M.V., Negussie, K., Corral-Pazos-de-Provens, E., Höfle, B. & Bubenzer, O. (2019): Comparison of Three Algorithms for the Evaluation of TanDEM-X Data for Gully Detection in Krumhuk Farm (Namibia). Remote Sensing. Vol. 11 (11), pp. 1-22.

Methods of gully detection and monitoring are the core research subject of the PhD project of Miguel Orti in the 3DGeo Research Group on the development of gully identification and measurement methods combining remote sensing and crowdsourcing techniques.

This research is part of the DEM_HYDR2024 project supported by TanDEM-X Science Team, therefore we would like to express thanks to the Deutsches Zentrum für Luft-und Raumfahrt (DLR) as the donor for the used TanDEM-X datasets. We acknowledge the financial support provided by the Namibia University of Science and Technology (NUST) within the IRPC research funding programme and to ILMI for the sponsorship of field trips to identify suitable study areas. Finally, we would like to express gratitude towards Heidelberg University and the Kurt-Hiehle-Foundation for facilitating the suitable work conditions during this research.

India accounts for nearly 18% of the worlds population. The country is also one of the main carrier of the worlds disease burden. Despite the general increase in life expectancy and decreasing mortality due to communicable diseases and malnutrition in recent years, the numbers of non-communicable diseases are rising substantially. Cardiovascular diseases such as diabetes and ischemic attacks are on the rise (Dandona 2017). Throughout the country the socioeconomic and health diversity is high and the Indian health system from under substantial shortcomings relating to infrastructure, the quality, and availability of services (Angell et al. 2019; Bhargava and Paul 2018).

OSM health care facilities in India over time

OSM health care facilities in India over time

Access to health services is determined by various factors, such as affordability, availability, information, accessibility, and compatibility. Affordability and information are cited as the main barriers that determine access (Butsch, 2011). Therefore, it is important that data on the spatial location of healthcare facilities is made freely available to all citizens. OpenStreetMap as a central global place for free geodata can be a solution to this. Public health officials can employ the data on care infrastructure to support their decision making. Businesses can use it to built new services on top. Citizens can easily interact with the data, getting answers on questions where and how to access a care service provider in their local community.

RMSI - a GIS consulting company, contributing since 2018 to OSM - leads efforts on an import of data on healthcare facilities in India since April 2019. The data source of the import are three different archives published by the Indian government on their open data portal: https://data.gov.in/. The archives include facilities like: hospitals, clinics, health centers and blood banks.
In the following we explore in which state when health-related objects were imported. We use HeiGIT technology to assess the imports to OSM: the ohsome API and the ohsome dashboard.

Tagging scheme for healthcare facilities in OSM:

  • amenity=doctors :: min. 1 doctor present, non-inpatient care
  • amenity=clinic :: min. 10 doctors present, non-inpatient care
  • amenity=hospital :: inpatient care
  • healthcare=* :: tag to cover more details on healthcare facilities (e.g. midwives, nurses, hospices, centres, blood banks, birthing centres)

Note that the tag “healthcare=*” is often used complementary to the other tags and therefore we do not evaluate that tag separately.

ohsome Results

Before the import started, only 5399 hospitals and 1557 other facilities were represented by OSM in India. Beginning in spring 2019 we have found a surge in new facilities added, across all states.

OSM hospitals in India over time.

OSM clinics in India over time.

OSM doctors in India over time.

First significant imports started at the beginning of 2019 with the clinics tag, followed by the imports of hospitals in May 2019 and in (April) September 2019 with the doctors tag. For most Indian states the number of healthcare related objects has (min.) doubled and was carried out step by step. For example, at the beginning of our time series (March 2018), most doctors were tagged in Karnataka (SW-India) (53), this number changed by the end of May 2020 and rose from 53 to 73, and to 115 by the end of June 2020, where our time series ends. For other states - like Assam (NE-India) - only one doctor was tagged at the beginning of March 2018. Then the import started, and within one week there were about 100 new facilities.
The imports for hospitals draw a similar picture, but this tag shows the most imports, e.g. in Maharashtra (SW-India) the number of tagged hospitals rose from 682 to 6343 and they are still ongoing - what we can assume from the curve. For other states, like Andhra Pradesh (SE-India), the import stagnates since January. Another significant import of clinics happened in Telangana (southern India), where within one week the number of tags rose from 49 to 2252 in June 2019. Further imports for this state happened later in October/November 2019, but some tags were deleted in April 2020. In general, most significant imports happened in November 2019 and in some states they are ongoing.

Looking at the density of tagged healthcare facilities per state, the highest density per 100k inhabitants is given in Lakshadweep (tropical archipelago), followed by Goa (SW-India) and Puducherry (SE-India). On the other hand the lowest density per 100k inhabitants is given in Bihar (NE-India), Jammu and Kashmir (northern India) and Jharkhand (NE-India):

OSM healthcare facility density by Indian state.

Conclusion/Future

We have found that over the course of less than a year healthcare objects in India went from 6956 to 48101, including ~33000 hospitals. This is a huge increase in critically important geo data. Further, this data is now also easy accessible through the OSM ecosystem of diverse services. Based on this data, we were able to determine in more detail which Indian states have a high density of health care infrastructure and which have a low density of health care facilities. However, this is still a rather macro perspective. With the new data now in OSM, everyone can assess the distribution on a spatial scale that suits their use case best.

In an upcoming analysis we will focus on the attributes. What thematic information we can derive from health facilities in India. How dense are these facilities tagged. What is actually tagged, and furthermore for what facilities and where in India can we derive capacity information.

Related Work

“Local Knowledge” is constituting the exceptional value of Volunteered Geographical Information and thus also considered as an important indicator of data quality. We are interested in how much local information is captured in OpenStreetMap data. In this blog post we explore the temporal evolution of mapping in OSM and the information stored in its database, by taking an explorative  look at four different cities in Germany, Nepal and the Philippines: Heidelberg, Kathmandu, Pokhara and Manila.

Heidelberg is generally considered to be relatively well mapped and has experienced mapping activity over a decade for now. Mapping in Kathmandu has been impacted heavily by data created for disaster response in the aftermath of the 2015 earthquake disaster in Nepal. This resulted in a significant increase in activity from mappers around the world. As comparison, we also will take a look at Pokhara. Pokhara is Nepals second largest city and lays approximately 200 km west of Kathmandu and belongs to the more rural part of Nepal. Manila is the capital and the economical and cultural center of the Philippines.

The image below shows a potential classification of OSM data in regard to the types of information it might contain. This is mostly targeted towards a humanitarian mapping context and may need adaptation for a more general evaluation of mapping phases, but we use it here as an example only. While buildings and road network completeness are of interest for level 0-1 (mostly based on remote mapping such as in international humanitarian mapathons, the further levels 2-4 are considered to source from local knowledge.

Fig. 1: Image from Twitter post by Rebecca Firth from HOT

In the following we will compare different aspects of development of OSM Data, including

  • completeness of road network and buildings (level 0-1)
  • exploratory analysis of local information for facilities and POIs (level 3)
  • overall information richness (level 4)

Examining this evolution should give us some insights on how long it takes volunteers to provide local information (especially in a context where mapping started with remote mapping) and how far the process is at the different locations. In order to perform this analysis the ohsome API by HeiGIT was utilized to access the OSM full history data. The API provides different endpoints to extract and aggregate data about the objects, users and single contributions.

NOTE: In addition to this Blogpost a Jupyter Notebook (source code) was released, which allows u to generate an interactive map and plots for regions of your choice.

Level 0-1: Roads and Buildings

Road network

Especially humanitarian mapping in OSM often started with roads and buildings, which can be traced remotely from satellite imagery. The situation is different in areas where there is already an existing strong OSM community for a longer time. For example in total, over 1.75 million kilometers of highway were mapped in Kathmandu. The related graph shows clearly the impact of the 2015 earthquake: the road network increased by approximately 15% directly after the disaster and by 30% until today. Pokharas increase since the earthquake is even bigger, with doubling its mapped length of highway objects since when. Especially two spikes are noticeable: one in the direct aftermath of the earthquake and one in the year 2016. The development of Heidelbergs road network length, showed in contrast a more constant development, with a small growth rate over the last decade. Manilas road network showed seemed to be still in the phase of active mapping: its length increased by approximately 7% over the last year.

Buildings

The mapping of buildings in comparison to the length of the road network showed a slightly delayed development. After initial mapping of buildings in 2011, only very few buildings were mapped in Pokhara until about one year after the earthquake. Afterwards the number of mapped buildings showed a rapid growth over a few months in the second quarter of 2016. Kathmandu showed three main growth events: one in the end of 2012, marking the first noticeable amount of contributions, one directly following the earthquake in 2015 and another smaller one in 2019. Manila showed a steady data evolution, with the exception of one spike in 2019 which indicates that similar to the road network buildings are still not mapped completely.

Level 3: Temporal Evolution of Facilities and POIs

Facilities

The third level is characterized by information about facilities. Here we take a look at the temporal development of educational facilities, access to drinking water, healthcare facilities and information about the road network (in this case bridges and tunnels). The plot below shows the total count of objects belonging to these groups.

While Heidelberg showed a more or less constant behavior, the 3 other cities showed a more irregular growth pattern. Manila experienced a steady increase over the last 10 years, which further accelerated in the last one to two years. This indicates an ongoing mapping of facilities and infrastructure. Kathmandu’s graph showed a strong increase between 2012 and 2013 and has since experienced irregular phases of growth which slowly leveled out recently. Notable is, that the 2015 earthquake response mapping didn’t had a significant impact on this development. Pokhara’s development, started slow and has grown between 2016 and 2018, before leveling out afterwards.

Point of Interests

As point of interest (POI) we consider objects containing the tags name and amenity. The amount of those are of interest to understand the amount of information besides geometries, facilities and critical infrastructure. A comparison to the development of roads and buildings indicates that the mapping of POIs followed the mapping of essential map features like buildings and roads. In particular the development for Kathmandu and Pokhara, which experienced short concentrated periods of highway and road mapping, showed a delayed evolution in regard to the mapping of POIs. This might indicate that mapping buildings and roads and mapping POIs were two separate processes. Manila and Heidelberg showed at least some form of co-occurence between the mapping of buildings and roads and the mapping of POIs which might indicate a simultaneous mapping process of the different features.

Level 4: Overall Information Richness

Temporal Evolution for Buildings and Roads

Following the scheme above, the main characteristic of the fourth level is a high amount of stored information in the objects in the map.  We will take a look at the number of additional tags per object. For definition of Richness of VGI data see Ballatore & Zipf (2015).

The graph below shows clearly that the amount of additional tag information in Heidelberg is very high for roads and buildings. For instance, more than 50% of the buildings contain five or more tags. Manila, Kathmandu and Pokhara had a significant lower portion of buildings and streets containing additional information. An exception was the road network of Manila which was comparable to Heidelberg.

Spatial Distribution for Buildings and Roads

(the map can only be viewed using the Jupyter Notebook)

Exploring the spatio-temporal domain using the leaflet map, shows that Manila and Heidelberg both showed alternating pattern of activity over a longer stretch of time. Pokhara and Kathmandu instead showed region wide morex extensive activity over short periods.

This suggests, that Heidelberg and Manila, both had a variety of spatially separated processes, while Kathmandu and Pokhara, were affected events covering the whole cities.

Conclusion

Mapping patterns in Kathmandu and Pokhara were clearly distinct from those in Heidelberg, both with respect to the temporal development of buildings and roads and the amount of tags. This indicates that a lower amount of local knowledge was present in the OSM data of the two cities. Mapping in Manila showed at least some resemblance in comparison to the development of Heidelberg, but also contains overall less information yet and the buildings and road network are still undergoing constant mapping.

In case you are interested to learn more about ohsome take a look at the How to become OHSOME series or take a look at the literature below. In case you want to take a look at a region of your choice, just add your bounding box in the cell of the jupyter notebook and rerun the cells.

Links:

OSHDB and ohsome API git repositorys

Humanitarian OSM Stats Global statistics for Humanitarian Open Street Map Team projects

ohsome HeX- Open Street Map History Explorer

Literature

Raifer, M., Troilo, R., Kowatsch, F., Auer, M., Loos, L., Marx, S., Przybill, K., Fendrich, S., Mocnik, F.-B.& Zipf, A. (2019): OSHDB: a framework for spatio-temporal analysis of OpenStreetMap history data.Open Geospatial Data, Software and Standards 2019 4:3. https://doi.org/10.1186/s40965-019-0061-3

Auer, M.; Eckle, M.; Fendrich, S.; Griesbaum, L.; Kowatsch, F.; Marx, S.; Raifer, M.; Schott, M.; Troilo, R.; Zipf, A. (2018): Towards Using the Potential of OpenStreetMap History for Disaster Activation Monitoring. ISCRAM 2018. Rochester. NY. US.

Grinberger, A. Y; Schott, M; Raifer, M.; Troilo, R.; Zipf, A. (2019): Analyzing the spatio-temporal patterns and impacts of large-scale data production events in OpenStreetMap In: Minghini, M., Grinberger, A.Y., Juhász, L., Yeboah, G., Mooney, P. (Eds.). Proc. of the Academic Track at the State of the Map 2019, 9-10. Heidelberg.

Ballatore, A. and Zipf, A. (2015): A Conceptual Quality Framework for Volunteered Geographic Information. COSIT - CONFERENCE ON SPATIAL INFORMATION THEORY XII. October 12-16, 2015. Santa Fe, New Mexico, USA. Lecture Notes in Computer Science, pp. 1-20.

Ludwig, C, Fendrich, S, Zipf, A. Regional variations of context‐based association rules in OpenStreetMap. Transactions in GIS. 2020; 00: 121. https://doi.org/10.1111/tgis.12694

« Newer Posts - Older Posts »