Today, the growing number of distributed sensors and tracking systems are generating overwhelming amounts of high velocity spatio-temporal data. Executing high performance queries on enormous volumes of spatial data, has become a necessity for numerous domains ranging from atmospheric, climate and ocean simulations to signal processing, traffic, and behaviour modelling. As the dimensions and volume of the data grows to massive scales, processing and storage with conventional methods is challenged. Most interestingly though, even most state of the art “Big Data” processing tools fall short in supporting spatiotemporal data needs efficiently, as they lack support for even basic spatial properties and methods (such as spatial indexing and joins). Combining these challenges with real time requirements (such as sub-second query response times required for collision avoidance and anomaly detection) only exacerbates the problem.

To support such applications, the research community has long been exploring methods of data reduction, compression, time-window approaches, parallel processing, distributed storing and many more, while often accepting accuracy and performance trade-offs. This special issue aims to highlight problems originating from real world application fields dealing with spatiotemporal Big Data challenges and invite researchers working towards novel methods for addressing these issues to submit their work. The aim of this special issue publication is to cover novel data science theory and algorithms, data engineering and real world systems architectures, which are aimed at the storage, fusion, processing, learning and ultimately knowledge extraction from real world spatio-temporal datasets.

This SI publication is aimed at researchers, scientists and practitioners with interests that lie at the intersection of data science and large-scale data management problems.

The issue will focus on technologies and solutions related (but not limited) to:

-Spatiotemporal compression and clustering techniques effective for big data processing
-Spatial data mining algorithms and solutions
-Large-scale parallel and distributed implementations for geospatial datasets
-Real-time processing and learning based on spatio-temporal features
-Knowledge discovery implementations from spatiotemporal real world datasets;
-Visual and data analytics, knowledge representation of big geospatial data
-Cloud enabled Big data architectures and real world applications;

CFP available at http://www.journals.elsevier.com/future-generation-computer-systems/call-for-papers/special-issue-on-spatiotemporal-big-data-challenges-approach