Feed on

Dear mappers,

on Monday 17.5.2021, 18.00pm a third international mapathon together with disastermappers heidelberg, MAMAPA, CartONG, HuMap and OSM Ghana will take place, which will give you some insights in mapping for humanitarian purposes from the perspective of different organizations and people around the world!

For beginners we will offer a mapping introduction as usual, so you don’t need any knowledge or experience. All you need is a computer, internet, a mouse and some tasty snacks and drinks!
We hope to see many of you!

If you want to participate, please register in advance here:

EN: https://mamapa.org/h-onlinemapathonregistration

DE: https://mamapa.org/h-onlinemapathonanmeldung


Recently, a new research paper “Detecting OpenStreetMap missing buildings by transferring pre-trained deep neural networks” (Pisl, J., Li, H., Herfort, B., Lautenbach, S., Zipf, A. 2021) has been accepted at the the 24th AGILE conference 2021. The conference will take place virually on June 8 to 11, 2021.

Accurate and complete geographic data of human settlements are crucial for effective emergency response, humanitarian aid and sustainable development. OpenStreetMap (OSM) can serve as a valuable source of this data. As there are still areas being incompletely mapped in OSM, deep neural networks have been trained to detect such areas from satellite imagery. However, in regions where little or no training data is available, the network training remains problematic.

In this study, we proposed a method of transferring a building detection model, which was previously trained in an area  well-mapped in OSM, to remote data-scarce areas. The transferring was achieved via fine-tuning the model on limited training samples from the original training area and the target area. We validated the method by transferring deep neural networks trained in Tanzania to a site in Cameroon with straight distance of over 2600 km, and tested multiple variants of the proposed method. Finally, we successfully detected 1192 missing OSM buildings in Cameroon with the fine-tuned model. The results showed that the proposed method led to a significant improvement in f1-score even with as little as 30 training examples from the testing site. This is a crucial feature of the proposed method as it allows to fine-tune  models to regions where OSM data is scarce.

Fig.1: Building detection results of the transferred model in a selected area in Batcham, Cameroon.

The proposed model can detect individual buildings by generating bounding boxes around each of them. However, further volunteer contributions are still expected with regard to mapping detailed footprints as well as validation. Therefore, this work contributes to the efforts towards machine-assisted humanitarian mapping methods. Future works will focus on better facilitating and supporting the OSM mapping community by estimating amounts of missing buildings or prioritizing unmapped areas.

More details will be presented at the AGILE 2021 conference. Stay tuned and hope to see you there!

Related master thesis:

Previous related work, e.g.:

Am kommenden Dienstag 18.05.21 um 19:15 geht Prof. Dr. Detlef Müller-Mahn von der Universität Bonn in seinem Vortrag bei der Heidelberger Geographischen Gesellschaft HGG auf “Afrikanische Zukünfte nach COVID-19″ ein:
Noch steckt die ganze Welt mitten in der Pandemie, aber schon jetzt stellt sich die Frage, wie es danach weitergeht. In Afrika wirkt COVID-19 als Krisenverstärker mit regional sehr unterschiedlichen Folgen. Der Vortrag stellt erste empirische Befunde aus laufenden Untersuchungen in Äthiopien und Kenia vor und verbindet diese mit Forschungsergebnissen aus dem Sonderforschungsbereich (SFB-TRR 228) „Future Rural Africa“. Im Kern geht es darum, wie verschiedene Zukunftsentwürfe im Spannungsfeld von großräumigen Planungen und neuen Ungewissheiten aufeinandertreffen und dabei die Zukunft zu einem Schlachtfeld der Gegenwart werden lassen.

Den Flyer mit den Vortragsthemen finden Sie unter folgendem Link: https://hgg.urz.uni-heidelberg.de/pdf/hgg_flyer_sose2021.pdf

Die Abendvorträge werden auch in diesem Sommersemester online stattfinden. Zugang hierzu haben Mitglieder der HGG und angemeldete Schulklassen. Der Zugangscode wird den Mitgliedern und Neumitgliedern per Mail oder per Post zugeschickt. Für das Sommersemester 2021 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 auf Nachfrage per Mail.



HGG Logo

The significance of gully filling practices on agricultural production has been only rarely evaluated in literature. In this study, a medium-sized catchment was selected, for being representative of the conventional farming practices on annual crops in the Campiña landscape in Southern Spain. During two wheat campaigns, an analysis of the spatial and temporal evolution of satellite-based vegetation indexes (NDVI, EVI and VH/VV radar polarization ratio) was conducted before and after gully filling practices associated to a rehabilitation project. The development of a linear model of topographic factors along with the calibration of a yield equation using a field survey allowed to determine the expected site-based productivities in the study area. The maximum EVI proved to be the best proxy for wheat yield in a location and was controlled mainly by rainfall and irradiation conditions. A significant yield reduction was found in the vicinity of those gullies affected by soil scraping within the rehabilitation project. Indirect yield losses in the productive land doubled the direct cost of unproductive areas caused by gullying and, when added up to the latter, between 15–25% of the expected yield was found to be lost. Although longer and broader studies are needed to confirm our results, this case study shows that current agricultural practices are not environmentally unsustainable but also economically detrimental. Moreover, it highlights that the impact assessment of gully erosion cannot only be circumscribed to the eroded area but the degradation of gully-adjacent locations must be also considered.

View of the ratio of EVI deviation (REVImax, Eq. 4) between 2017 and 2019 in the study catchment. The zones of maximum variation (dark grey) in the deviation map (on the right) match visually with light-coloured areas affected by scraping in filling operations after 2017 campaign in those gullies not affected by the rehabilitation project (gullies 1 to 3, on the left, Google Earth images). On these left-hand images, the continuous lines correspond to the gully limits before first campaign and the dashed lines, to the areas of maximum deviation on the map.

View of the ratio of EVI deviation (REVImax, Eq. 4) between 2017 and 2019 in the study catchment. The zones of maximum variation (dark grey) in the deviation map (on the right) match visually with light-coloured areas affected by scraping in filling operations after 2017 campaign in those gullies not affected by the rehabilitation project (gullies 1 to 3, on the left, Google Earth images). On these left-hand images, the continuous lines correspond to the gully limits before first campaign and the dashed lines, to the areas of maximum deviation on the map.

Find all details in the full paper (free access for 48 more days):

Castillo, C., Pérez, R, & Orti, M.V., (2021): The impact of recent gully filling practices on wheat yield at the Campiña landscape in Southern Spain. Soil and Tillage Research. Vol. 212 , pp. 1-13.

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.

Find more studies focusing on the challenge of identifying, outlining and analysing gully-affected areas in recent publications:

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.

This study was supported by Project I+D+i 27451 within the Operative Programme Feder Andalucía 2014–2020 with funds from the Andalusian Government (Consejería de Economía y Conocimiento) and FEDER funds.

We are pleased that our research paper “Deriving indicators for Points of Interest and analyzing mixed activities in urban areas” (Ullah, Lautenbach, Zipf 2021) has been accepted at the 12th International Symposium on the Digital Earth (ISDE) scheduled within the GI Week 2021 in Salzburg. The conference will take place from July 5th to July 9th 2021.

We are looking forward to discuss with you following aspects at the conference:

The support of a pedestrian friendly city with a variety of services such as eating, education, health-care, shopping and personal services has become an important goal for urban planning, local authorities, families and economic groups. Although mixed activities play an important role in urban planning projects, measuring mixed activities has been problematic due to the lack of appropriate data and measurement approaches. We suggest a new approach that applies concepts from ecology on Points of Interest (POIs) from OpenStreetMap to quantify mixed activities.

In ecology, there are dozens of potential diversity indices, which have been deployed in recent land use studies to measure mixed activities quantitatively. Ecologists however, have highlighted that these indices are not always expressed in units that are intuitive. The limitations of commonly used diversity indices can be avoided by the use of Hill numbers, which represent a mathematically unified family of diversity indices. We used Hill numbers on POIs to measure mixed activities at a quarter level in Frankfurt.

Fig.1: Sample completeness curves and Hill numbers (q=0,1,2) for the five city quarters: Eschersheim, Heddernheim, Kalbach-Riedberg, Niederursel and Praunheim. The grey areas indicate the 95% confidence bands derived from bootstrap.

Results showed that Hill¹ (exponential of Shannon) is an appropriate quantitative measure to describe the diversity of facilities and services by a single numerical value. Based on a base coverage of 94% (grey dottet line) the curves show that Heddernheim has the highest diversity, followed by Eschersheim and Kalbach-Riedberg, while Niederursel has the lowest Hill¹ value.

More information will be presented a at the GI_Week 2021. Looking forward to see you there! Stay tuned for further updates!

Welcome back to our blog post series ohsome Region of the Month where you can find information on potential applications of the ohsome API. If you’re new to this format you can also read some of the former blog posts from this series which you can find here (about street networks), here (about railway networks) or here (about forests).

This month, we’ll look at another ”evolution-animation”. It deals with the development of mapped buildings in Nairobi (the capital of Kenya) in general and looks at how many of those buildings have additional information in form of the ”addr:city”-tag as a sort of indicator on how detailed the given information is.


As usual, you can download your boundary data from this website. There were multiple requests sent to the ohsome API which you can find in a snippet as well as the boundary-GeoJSON file itself which you can conveniently download there, too.


Below you can see an animation of the evolution of buildings mapped in Nairobi between the years 2008-2021. The red buildings are the ones that have the additional addr:city-information tagged.

You can find an example for the output in the snippet and even another one with instructions on how you can create an animation like this one yourself.

The very first building in Nairobi was mapped in September 2008. About a week later, more buildings joined in. After these first regional mapping activities not much happened until about December 2009, an occurrence that might be linked to an import of Map Kibera data between November 2009 and January 2010. Furthermore, one can observe a more frequent mapping activity, especially in terms of new added areas, from that point onwards. A period of very high mapping activity can be found during the second half of 2011, for which unfortunately no potential explanation could be found. The first buildings that actually have the additional ”addr:city”-information added occur in December 2012, which could be motivated by an increase in coverage of Bing-Aerial Images for Nairobi in April 2012. However, the progress remains rather small until  spring 2016 when bigger parts start getting added. A trend which only continues in the following year around the time when the Youth Mappers launched at Kenyan Universities.


Below is a graphic of density-development of buildings in Nairobi over the years:

In accord with the animation there is no notable density-development until about later 2008 whatsoever. After that, one can observe an overall increasing trend, which again might be related to activities, of the Map Kibera Trust with stronger jumps in density between July and August of 2011, which was already displayed in the animation too, in September 2013, as well as in January 2018 from which the latter values show a stronger increasing tendency. A Mapathon in connection to the 7th International Open Data Day took place in March 2018 and could be a potential reason for the higher values around that time either.


Here you can see the monthly user-counts of the building-tag for Nairobi over the years with the average being 12 users/month:

As well as in the graphic for density, one can see an overall positive trend in users interacting with the building-tag in Nairobi, yet the main developments striking the eye of the observer would be three distinct peaks in user-counts. One is in August 2015, the second in 2019 and the third and last in November 2020. For the first, unfortunately no potential explanation could be found, although mapathons within the local community might be an explanation. The second might be related to an Open Data Mapathon that took place at the beginning of March 2019 and could therefore be responsible for the higher numbers until the first of April 2019. Starting in October 2020 the Youth Mappers Kenya have made an effort trying to map Kenyan Health facilities, so there could be a slight, although temporarily delayed, connection. Furthermore, the graphic shows a period where there has been well above average user-activity between January 2018 until October 2018. These relatively high numbers might be related to the attempt of mapping Nairobi’s Slums for better healthcare or to the Open Data Day Mapathon that was mentioned earlier.


This last graphic shows you the ratio of those buildings, that have the additional ”addr:city” information given to the total of building=*-objects:

Unlike displayed in the other graphics, one can see that there has not been any buildings with the additional ”addr:city”-information until December 2012, therefore there is no ratio-data for the time span of about 5 years available. Since the progress remains quite small (see animation) until early 2016, a potential reason for the development might be an individual effort or smaller local projects, but that is of course only an assumption. The values are generally pretty low yet they do show a positive trend with an exceptionally strong increase in May 2018. After that, one can observe another slight decrease again which itself might be related to the strong increase of overall building-density during that time.

We can conclude that Nairobi has come a long way and can show a very positive development over the years when it comes to OpenStreetMap-data coverage. When looking at the trends of the different graphics and the apparently high activity of the local mapping community, one can assume that there will be more and further detailed mapping going on in the future.

Thanks for reading this month’s blog post on our Region of the Month Nairobi, Kenya. Stay tuned for more blog posts in the coming months!

Background info: the aim of the ohsome OpenStreetMap History Data Analytics Platform is to make OpenStreetMap’s full-history data more easily accessible for various kinds of OSM data analytics tasks, such as data quality analysis, on a regional, country-wide, or global scale. The ohsome API is one of its components, providing free and easy access to some of the functionalities of the ohsome platform via HTTP requests. Some intro can be found here:

It is our great pleasure to inform about the new “Robert and Christine Danziger Scholarships for doing a PhD in Geoinformatics at Heidelberg University.

Please make yourself familiar with the research topics at GIScience Heidelberg and HeiGIT (e.g. projects, papers) when suggesting potential projects for your PhD research to potential supervisors.

Short facts:

The Robert and Christine Danizger Scholarship is awarded to very good up-and-coming doctoral students from Africa (primarily from Ghana or another country in West and Central Africa) who are planning a doctorate at Heidelberg University in the fields of Geoinformatics (or Political Science) and who need help with funding their doctorate.

Learn more about everything you need to consider when applying, on this page.


  • The amount of the scholarship for doctoral students is 1,200 € per month.
  • The award period is generally 2 semesters. An extension is possible on request and if all requirements are met (academic performance, financial hardship, availability of funds) for a maximum period of 2 semesters.


You can apply for a scholarship from 1 May to 15 July, 2021. Please send your application documents by email to danziger_stipendium@uni-heidelberg.de.


Please submit the following application documents during the application period:

  • Signed application form
  • Current performance records: University diploma that qualifies for a doctorate (original certificate) including the final grade as well as an official translation in German or English and if available, an official conversion of the diploma to the German grading system.
    If your university diploma is not available by the end of the application period, it is sufficient to submit a provisional certificate from the university confirming your academic performances, so far (ideally with a grade of all examinations, so far), that states that you expect to complete your course of study by the start of the semester following the application period. Please note that you have to submit your missing certificate by the start of the lecture period.
  • Confirmation of acceptance of the doctorate (ps: currently either Prof. Zipf or Prof. Höfle can accept PhD students in Geoinformatics at Heidelberg University)
  • Curriculum vitae in table form
  • Letter of motivation with reference to the topic of the doctorate
  • Recommendation letter from a university instructor in the specific field
  • Proof of citizenship
  • Self-disclosure of revenue and expenditure (part of the application form)
  • Evidence confirming the financial situation (pay slip of the parents, own wage slips etc.)


Many years of friendship and cooperation connect us to Ghana and, more recently, to other parts of West and Central Africa as well. With the scholarship, we want to support talented young people from African countries wishing to come to Heidelberg University to start and successfully complete a Master’s degree or a doctorate. Where financial resources are limited, a scholarship can be crucial. Programs of study in the fields of Geography, Political Science and Sociology are particularly important to us.” – Christine and Robert Danziger

Geoinformatics Heidelberg University

Geoinformatics Heidelberg University

We do thank Christine and Robert Danziger cordially for this wonderful opportunity!

The project “25 Mapathons goes into the next phase. Based on the experience of until today 17 OSM mapathons with German Red Cross (GRC) chapter and an extensive literature review on the success of mapathons, the “25 Mapathons” team has identified several points of possible improvement in the way how mapathons are conducted in general:

A common problem of mapathons is, that although many new people with no prior contact to OpenStreetMap (OSM) are taking part in the event, just a tiny fraction of those people keep mapping afterwards.

Therefore, we developed a new mapathon concept with the aim to increase user retention rates. The new concept shall increase intrinsic and extrinsic motivation and break down existing barriers and challenges for beginners.
Among others, one change to the conventional way mapathons are conducted is the extended duration of the event and the mapping period in order to tackle problems like time pressure or insufficient feedback. The variables of interest, such as motivation, previous experience and feedback are captured by questionnaires for both organizers and participants. The results of the two-part mapathons in terms of mapping activities will then be compared to common one-part mapathons.
This concept was developed in close coordination with the Missing Maps community. The “25 Mapathons” team is very thankful for the extensive feedback by members of the community.

To test the concept of two-part mapathons, the “25 Mapathons” team prepared a test run in which mappers, mapping beginners and organizations with experiences in hosting mapathons can participate. In order to conduct comparable 2-part mapathons, the organizers will be provided with an exact timetable, a powerpoint presentation and the necessary questionnaires. Questionnaires will be also provided for organizations conducting conventional mapathons. The material is available in English and German.

If you and your organization are interested in participating in the experiment – either as organizers or participants - please contact S.Lohr@uni-heidelberg.de.

The first opportunity for participation is the final test for the two-part Mapathon on the 05.5.2021 (part 1​) and on the 11.5.2021 (part 2). We would love to welcome you at the event!
(Registration via Eventbrite: https://www.eventbrite.de/e/missing-maps-mapathon-for-beginners-2-part-event-tickets-150635327319)

We will keep you updated over the course of the experiment and hope to be able to support the OSM and Missing Maps community with valuable findings!


Seit kurzem nutzt die Stadt Mannheim openrouteservice (ORS) von HeiGIT für das Routing von Fußgängern, Fahrräder, Autos, LKW und Rollstuhlfahrern im Online-Stadtplan für die breite Öffentlichkeit, d.h. die Bürger und Besucher von Mannheim. Sie können es hier verwenden:


Dies zeigt, wie openrouteservice leicht in verschiedenste Clients und Apps integriert werden kann. Openrouteservice verwendet die freien und offenen Daten von OpenStreetMap (OSM) und bietet eine Vielzahl von frei nutzbaren API-Funktionen mit weltweiter Abdeckung. Ein Ökosystem aus verschiedenen open source Bibliotheken und Clients, z.B. für Javascript, Python, R, QGIS und das Web usw. erleichtern die Nutzung für viele Fragestellungen und ermöglichen die Integration auch in Ihre Anwendung.

Die Funktionen der ORS API-Endpoints umfasst unter anderem

  • Routing mit Anweisungen für Fußgänger, Rollstuhlfahrer, Autos, verschiedene Arten von Fahrradprofilen und Schwerfahrzeugprofilen mit vielfältigen Optionen
  • Fahrzeugflotten Optimierung (Traveling Salesman Problemeusw.)
  • Isochronen für Erreichbarkeitsanwendungen
  • Zeit-Distanz-Matrizen
  • Geokodierung und umgekehrte Geokodierung
  • elevation
  • Points of Interest (POI)

openrouteservice im Stadtportal Mannheim
openrouteservice im Stadtportal Mannheim

Bitte beachten Sie, dass das API-Kontingent für die Optimierung von Flotten mittels der openrouteservice optimize Schnittstelle erhöht wurde, um die Logistik während der Corona-Krise zu unterstützen. Sie können auch die interaktive Dokumentation der verschiedenen API-Endpunkte testen.

Laufende Forschungen bei GIScience Heidelberg und HeiGIT zu Routing beschäftigen sich, z.B. mit gesundem, schattigen, ruhigen und grünem Routing oder Anwendungen in der humanitären Hilfe (z. B. Zugang zur Gesundheitsversorgung im globalen Süden ).

Weitere News (Auswahl):

It’s been a little longer than a month since we published the Open Healthcare Access Map. Last week we rolled out all countries of Sub-Saharan Africa. Today all countries of South and Southeast Asia follow.

The web application is available here: https://apps.heigit.org/healthcare_access/.

To get a clearer understanding of healthcare infrastructure data in OpenStreetMap, please consult this blogpost from 2019. With the support of ohsome.org, the evolution of healthcare-related objects in OpenStreetMap is described for each country worldwide. An update of this analysis as well as an integration into the open healthcare access map is planned and will be announced soon.

All available countries in Open Healthcare Access Map

Fig. 1: All available countries in Open Healthcare Access Map

According to WorldPop data, South and Southeast Asia is home to 2.578 billion people. Our analysis shows that 1.801 billion (69.8%) of these live within a 60-minute drive of any healthcare facility. 1.455 million (57.7%) live within 30 minutes and 906 million (35.15%) live within 10 minutes. The following map shows the distribution of accessibility on the hexagon scale (zoom level8) across the two world regions.

Hexagon aggregation of population covered for 10, 30 and 60 minutes  driving distance towards the closest healthcare facility.

Fig. 2: Hexagon aggregation of population covered for 10, 30 and 60 minutes driving distance towards the closest healthcare facility.

Please note that this is still a prototype and feedback on improvements and desired functionalities is very welcome. Once again we highly appreciate your feedback on the app! Please feel free to reach out via mail.

Related work:

Older Posts »