Tuesday, December 4, 2018

LIVID: Characterising the variability of natural resources in long EO image time series


Background
Analysing the temporal dimension is becoming increasingly important in Earth observation (EO) and is required by certain application domains. Remote-sensing-based change detection (DC) was conceptually never limited to bi-temporal analysis (i.e. image differencing) although most change detection analyses have been conducted this way in the past. The availability of data sources with high acquisition frequencies, such as Sentinel-2 (e.g. on average every 5 days), or with a long historic archive, such as Landsat (e.g. over 40 years) now make more complex analysis of temporal dynamics possible.
The LIVID (Long Image Time-Series Variability Indicator Description) approach applies a fully automated, object-oriented workflow to calculate the overall variability of content for uniform objects of any shape and size between multiple images. This type of temporal analysis produces a single map of index values based on user-defined weights of their category changes over time, thus indicating regions with high relative variability or stability. The calculated index is based on semantically enriched, data-derived categories and user-defined weights pertaining to natural resource depletion.

The information about the variability of natural resources over a longer period of time is extremely important for many applications such as nature conservation and agriculture. It might be also used as pre-analysis to identify more specific areas that are worth investigating further using additional time series analysis methods or selecting ground reference points. It might also provide additional evidence in combination with other tools and interpretation of temporal changes.

Short description
LIVID is currently being developed at Z_GIS as an eCognition app. The task would be to develop the application further and apply it as a use-case to an example area/topic.

Suggested reading
Braun, Andreas, and Volker Hochschild. 2017. “A SAR-Based Index for Landscape Changes in African Savannas.” Remote Sensing 9 (4): 359. doi:10.3390/rs9040359
Sudmanns, Martin; Augustin, Hannah; Tiede, Dirk. 2018. LIVID”. Open Science Framework Project. https://osf.io/24ycm/

Start: Anytime

Prerequisites/qualification
eCognition/OBIA ++
Remote Sensing +

Analysis-Ready-Data: Comparison of different pre-processing routines in the big Earth data context


Background
Due to the massive volume, variety, and velocity of the Earth observation (EO) data offered by the Copernicus programme, the term Analysis-Ready-Data (ARD) is emerging when talking about big Earth data. This term refers to data that has been pre-processed to a certain degree such that they can be used directly as input to analyses without further pre-processing. It aims to shift the burden of routine, basic pre-processing from users to Earth observation data providers so that users can focus on their domain-specific applications rather than issues of calibration, topographic correction, etc. However, there is still no commonly agreed definition, understanding or standard of what ARD is or can/will be. Several stakeholders are developing their own ARD products (e.g. Planet, UGSS with Landsat). At Z_GIS we aim to push requirements of ARD further to include not only the provision of data, but providing it together with automatically generated, generic information layers in the context of the Sen2Cube.at project (http://sen2cube.at).



General workflow for generating information products from observations. ARD are concerned with the four first steps (data acquisition, conversion to radiance, TOA reflectance and Surface reflectance) allowing then to analyse data and generate time-series (from Giuliani et al. 2017)

Short description
The tasks include:
  • conducting a state-of-the-art assessment of existing ARD
  • providing a comparison metric for different definitions
  • selecting suitable approaches for ARD in the context of the Z_GIS project Sen2Cube.at
  • reviewing, selecting or developing appropriate quality indicators for data and workflows
  • (optional) implementing ARD generation in an automated workflow 
Suggested reading
Giuliani, G., Chatenoux, B., De Bono, A., Rodila, D., Richard, J.P., Allenbach, K., Dao, H. and Peduzzi, P., 2017. Building an earth observations data cube: Lessons learned from the Swiss data cube (SDC) on generating analysis ready data (ARD). Big Earth Data, 1(1-2), pp.100-117.
Tiede, D., Baraldi, A., Sudmanns, M., Belgiu, M. and Lang, S., 2017. Architecture and prototypical implementation of a semantic querying system for big Earth observation image bases. European journal of remote sensing, 50(1), pp.452-463.
Start: Anytime

Prerequisites/qualification

Remote Sensing ++
programming + (optional)

Comparing OSM and OGD road graph data

Suggested by: Martin Loidl, Bernhard Zagel

Short description: Since summer 2018, the Austrian, authoritative road graph (GIP) is available as
OGD with a complete coverage. This road graph serves as common reference for all road related data. Thus, the geometry is complemented with an extensive list of attributes. 
However, the crowd-sourced pendant, OpenStreetMap (OSM), enjoys big popularity. It is the basis for multiple services and plays a significant role in research.
GIP and OSM are not only different in their origin, but also in their data model, data quality and suitability for several application purposes. A benchmarking protocol for these two road graph sources is still missing. The aim of this master thesis research is to develop such a protocol with a focus on spatial and attributive data quality as well as suitability for a catalog of applications.

References, suggested reading:
  • GIP at OGD portal https://www.data.gv.at/katalog/dataset/3fefc838-791d-4dde-975b-a4131a54e7c5.
  • GRASER, A., STRAUB, M. & DRAGASCHNIG, M. 2013. Towards an Open Source Analysis Toolbox for Street Network Comparison: Indicators, Tools and Results of a Comparison of OSM and the Official Austrian Reference Graph. Transactions in GIS, 510–526.
  • HELBICH, M., AMELUNXEN, C., NEIS, P. & ZIPF, A. 2012. Comparative Spatial Analysis of Positional Accuracy of OpenStreetMap and Proprietary Geodata In: JEKEL, T., CAR, A., STROBL, J. & GRIESEBNER, G. (eds.) GI_Forum 2012: Geovisualization, Society and Learning. Salzburg: Wichmann, VDE Verlag.
  • BARRON, C., NEIS, P. & ZIPF, A. 2014. A Comprehensive Framework for Intrinsic OpenStreetMap Quality Analysis. Transactions in GIS, 877–895.
Related to project: Bicycle Observatory (https://bicycle-observatory.zgis.at)

Start/finish by: Anytime

Prerequisites/qualifications: Profound skills in data management and processing as well as in spatial network analysis.

Health effects of active commuting

Suggested by: Martin Loidl, Bernhard Zagel

Short description: In the past 2.5 years, we have investigated the health effects of active commuting in the context of an interdisciplinary, award winning research project (GISMO). For this, we implemented a 2:1 randomized, clinical intervention study, in which we measured the physical and psycho-emotional effects of changing from car commuting to active modes (cycling and walking in conjunction with public transport). All 73 subjects underwent medical pre- and post examinations and completed mobility questionnaires. In order to determine the mobility "dose", subjects were required to document their commuting behaviour in mobility diaries and to where GPS trackers for 2x2 weeks within the intervention period of one year.
The extensive data set that results from this study serves as a rich resource for all kind of research questions at the intersection of geography, health and mobility. We are happy to share it and fuel your own research in this context.

References, suggested reading:
  • CELIS-MORALES, C. A., LYALL, D. M., WELSH, P., ANDERSON, J., STEELL, L., GUO, Y., MALDONADO, R., MACKAY, D. F., PELL, J. P., SATTAR, N. & GILL, J. M. R. 2017. Association between active commuting and incident cardiovascular disease, cancer, and mortality: prospective cohort study. BMJ, 357.
  • PAGE, N. C. & NILSSON, V. O. 2017. Active Commuting: Workplace Health Promotion for Improved Employee Well-Being and Organizational Behavior. Frontiers in Psychology, 7.
  • SALLIS, J. F., CERIN, E., CONWAY, T. L., ADAMS, M. A., FRANK, L. D., PRATT, M., SALVO, D., SCHIPPERIJN, J., SMITH, G., CAIN, K. L., DAVEY, R., KERR, J., LAI, P.-C., MITÁŠ, J., REIS, R., SARMIENTO, O. L., SCHOFIELD, G., TROELSEN, J., VAN DYCK, D., DE BOURDEAUDHUIJ, I. & OWEN, N. 2016. Physical activity in relation to urban environments in 14 cities worldwide: a cross-sectional study. The Lancet, 387, 2207-2217.
Related to project: GISMO (https://gismoproject.com)

Start/finish by: Anytime

Prerequisites/qualifications: Profound skills in data management and processing, spatial analysis and statistical knowledge.

Spatial factors for mobility behaviour

Suggested by: Martin Loidl, Bernhard Zagel

Short description: Mobility is spatial by its very nature. Being mobile implies moving from an origin to a destination. Consequently, mobility is highly dependent from the spatial environment. Key geographical concepts, such as distance, accessibility and topology, are crucial for understanding mobility.
We have conducted the latest mobility survey (2018) at the University of Salzburg. Among 30+ questions on mobility behavior, we asked for the zip code of the home place and the university building that is the destination of the majority of commuting trips. We are now interested in the correlation of spatial factors and the mode choice of university members.
Additional to the data from the mobility survey, we can offer raw data from the latest nation-wide mobility survey. Relevant spatial data that represent the spatial environment are mostly available as OGD and can be accessed via the national data portal.


References, suggested reading:
  • BUEHLER, R. 2011. Determinants of transport mode choice: a comparison of Germany and the USA. Journal of Transport Geography, 19, 644-657.
  • DALTON, A. M., JONES, A. P., PANTER, J. R. & OGILVIE, D. 2013. Neighbourhood, Route and Workplace-Related Environmental Characteristics Predict Adults' Mode of Travel to Work. PLoS ONE, 8, e67575.
  • SCHEINER, J. 2010. Interrelations between travel mode choice and trip distance: trends in Germany 1976–2002. Journal of Transport Geography, 18, 75-84.
  • YANG, L., HIPP, J. A., ADLAKHA, D., MARX, C. M., TABAK, R. G. & BROWNSON, R. C. 2015. Choice of commuting mode among employees: Do home neighborhood environment, worksite neighborhood environment, and worksite policy and supports matter? Journal of Transport & Health, 2, 212-218.
Related to project: Bicycle Observatory (https://bicycle-observatory.zgis.at)

Start/finish by: Anytime

Prerequisites/qualifications: Profound skills in data management and processing, spatial analysis (network analysis) and statistical knowledge.

Mapping mobility types

Suggested by: Martin Loidl, Bernhard Zagel

Short description: Mobility behaviour is very different, depending on lifestyle, values, stage of life, socio-economic status, location and many more factors.
In this master thesis mobility types should be defined based on a synthesis from literature and mobility surveys. These mobility types should then be linked to descriptive attributes, which can be further linked to geolocated statistical data. The goal of this research is to produce distribution maps of mobility types with a high spatial resolution (e.g. census districts).

We can provide extensive data sets and raw data from mobility surveys. Additionally, results from focus group and expert interviews will be available by Q2 2019.

References, suggested reading:
  • DAMANT-SIROIS, G., GRIMSRUD, M. & EL-GENEIDY, A. M. 2014. What’s your type: a multidimensional cyclist typology. Transportation, 41, 1153-1169.
  • FRANKE, A., ANKE, J., LIßNER, S., SCHAEFER, L.-M., BECKER, T. & PETZOLDT, T. 2018. Are you an ambitious cyclist? Results of the Cyclist Profile Questionnaire in Germany. International Cycling Safety Congress. Barcelona: ISCS. 
  • DILL, J. & MCNEIL, N. 2012. Four types of cyclists? Testing a typology to better understand bicycling behavior and potential. Portland: Oregon Transportation Research and Education Consortium (OTREC).
Related to project: Bicycle Observatory (https://bicycle-observatory.zgis.at)

Start/finish by: Anytime

Prerequisites/qualifications: Profound skills in data management and processing. Statistical knowledge.

Wednesday, August 22, 2018

Reconstruction and analysis of pre-modern infrastructure


Suggested by: Christian Neuwirth

Short description: Remains of canal-shaped landforms in the Upper Austrian municipality of Regau bear witness to historic construction work. So far, no clear evidence for their function could be found. One hypothesis assumes historic hydro-engineering as a response to dry climate conditions. Similar structures could also result from historic road networks, which were overprinted by linear erosion.
The key objective is to verify different hypotheses by the analysis of high-resolution LIDAR data. Your first task will be to reconstruct the pre-modern terrain model. The purged DTM serves as a basis for the reconstruction of historic infrastructures.
You can freely choose from a variety of methods such as 3D modeling and visual interpretation, geomorphometric analysis or spatial simulation modeling to shed light on this unresolved question.
References, suggested reading:
Berlinger, J., 1926. Über Bodendenkmale. Heimatgaue Z. Für Oberösterr. Gesch. 7, 194–201.
Neuwirth, C., D’Oleire-Oltmanns, S., Eisank, C., 2013. A Proposal for Mapping Historic Irrigation Channels to Reveal Insights into Agro-Climatic Systems: A Case Study in Upper Austria. Verlag der Österreichischen Akademie der Wissenschaften.
Schmidt, J., Werther, L., Zielhofer, C., 2018. Shaping pre-modern digital terrain models: The former topography at Charlemagne’s canal construction site. PLOS ONE 13, 1–21. https://doi.org/10.1371/journal.pone.0200167
Start/finish by: Anytime
Prerequisites/qualification: background or interest in history and archeology, experience with a method relevant for this topic (geomorphometic analysis, 3D modeling etc.), basic programming/scripting skills
Please send your questions and further ideas to christian.neuwirth@sbg.ac.at.