The rise of our API economy over the last decade has offered developers a flexible and growing array of capabilities to help them build a rich variety of applications. APIs are the new software building blocks for the 21st century.

And the building blocks continue to grow and expand like wildfire. Consider the illustration from ProgrammableWeb below, the WebAPI count for which surpassed 17,000 in 2017 and shows beyond hockey stick growth over the past 12+ years.

But not every gap in the marketplace has been filled. New API requirements are constantly being identified, and vendors across a range of industries are filling those gaps as quickly as possible.

One such challenge has been for those who wish to tap into the vast and fast-growing stores of geospatial-temporal big data. Such a class of data, so named because of its inherent link to place and time, includes a wide variety of satellite and aerial imagery, global-scale data and models (weather, climate, oceans, and the like), georeferenced Internet of Things/sensor networks, and big-event data captured on near-real time platforms like Twitter and GDELT (a data set monitoring media outlets’ output from around the world in real-time).

Though such data is often freely available, its gargantuan size and the complexities associated with its preparation for use can often make it difficult to exploit, especially within the confines of a simple API call. This is especially the case for time-critical applications.

That’s the problem statement…now for IBM’s possible solution: The IBM PAIRS Geoscope.

IBM PAIRS Geoscope continuously and automatically ingests, curates, and integrates geospatial-temporal data from a variety of sources and offers the tidy result as a service. From a single API, developers can leverage this expansive collection of diverse datasets as an information service, a discovery service, or as a tailor analytics service.

Why developers want to tackle the geospatial-temporal data challenge?

Through the use of tools like Google Maps and visualization and mapping tools like ESRI, many of us have become familiar with the explosion of mapping data. But when considering geospatial-temporal data use cases more thoroughly, the challenges and complexities are profound as developers attempt to combine massive datasets that span over space and time.

The data are huge — often terabytes or even petabytes — and the datasets can be difficult to handle, as many of them have not historically been available via APIs. That’s why IBM PAIRS Geoscope is aimed at organizing these datasets. They can help developers create new types of applications that can access these datasets quickly and cleanly, and without the huge ETL load and data normalization processes that would normally be required.

What are the specific benefits of IBM PAIRS Geoscope?

From the developer’s point of view, IBM PAIRS Geoscope presents a number of key differentiators and benefits:

  • Speed or time to insights. Data layers are curated such that IBM PAIRS Geoscope can deliver data and insights by orders of magnitudes faster than compared to conventional data services.
  • Ease of use. Developers don’t have to learn and stitch together multiple APIs. Instead, using IBM PAIRS Geoscope and one API, developers can efficiently and quickly explore and analyze a wide variety of complex geospatial-temporal data.
  • Comprehensiveness. IBM PAIRS Geoscope can automatically ingest, curate, and integrate all forms of geospatial-temporal data, including raster (e.g. satellite imagery), vector (e.g, maps), IoT (e.g. sensors), and social data (e.g. Twitter).
  • Scalability. It can scale to very big data and complex queries compared to conventional systems.

What are some common use cases for the IBM PAIRS Geoscope data?

We’ve only just begun and yet have already received a wide range of inquiries as to specific use cases that would provide optimum leveraging of PAIRS Geoscope data, especially those challenging problems that can’t be easily solved with other analytics techniques. These range from “Where is the best place to go on vacation?” to “Where are the most vulnerable places or biggest problems during an earthquake in a specific region?” to “How can we track the spread of a wildfire.” And we know there are many, many more to come.

How are organizations already using IBM PAIRS Geoscope data?

The service is already being used by clients from across commercial and academic institutions, including and receiving tens of thousands of API requests per day.

By way of example, IHI Corporation, a global engineering, construction and manufacturing company is working with IBM PAIRS Geoscope to develop a new system for improving the accuracy of long-term (30+ days) weather forecast by 30+ percent over other techniques. The team users data from GPS Radio Occultation sensors on satellites, which can yield 3D temperature, pressure, and humidity profiles on the atmosphere. Using this service the data is blended with historical and long-term weather forecast data and machine learning techniques to produce improved weather forecast insights.

The University of Michigan is another example of an organization making effective use of IBM PAIRS Geoscope data, in this case for its solar car race team’s racing strategy. In last year’s Bridgestone World Solar Challenge, an 1,800 mile race across the Australian Outback, the Michigan team used IBM PAIRS Geoscope as an analytics service to create a customer machine learning-based weather prediction model using historical solar and wind data. They used that insight to adjust their race strategy and ultimately came in 2nd place.

How can developers get started with IBM Pairs Geoscope?

Visit the IBM PAIRS Geoscope landing page to give the web interface a test run on some initial use cases of how you can use the service. The corresponding APIs can then serve as a starting point for a wide variety of applications that bring AI techniques to geospatial-temporal data. There, you will also be able to see the full documentation and API access details to better understand more sophisticated query and discovery examples.


More on IBM PAIRS Geoscope:


Examples of datasets already curated and available via IBM PAIRS Geoscope

Agricultural Datasets
California Irrigation Management Information System (CIMIS)
USDA Cropscape Crop-Specific Land Cover Data
USDA Federal Grain Inspection Service (FGIS)
USDA Foreign Agricultural Service (FAS) Export Sales Reporting (ESR)
USDA Foreign Agricultural Service (FAS) Global Agriculture Information Network (GAIN)
USDA Foreign Agricultural Service (FAS) Global Agricultural Trade System (GATS)
USDA Foreign Agricultural Service (FAS) Production, Supply and Distribution (PS&D)
USDA National Agriculture Imagery Program (NAIP)
USDA National Agricultural Statistics Service (NASS)
USDA Natural Resources Conservation Service (NRCS) Soil Survey Geographic database (SSURGO)
USDA World Agricultural Supply and Demand Estimates (WASDE)

Land Elevation Datasets
NASA/USGS Shuttle Radar Topography Mission (SRTM) World Elevation
USGS National Elevation Dataset (NED)

Populace Datasets
European Commission Eurostat Collection
United Nations Statistics Division UNdata Collection
US Census Bureau
World Bank DataBank Collection

Satellite Earth Observation Datasets
European Space Agency (ESA) Sentinel Platform
NASA Moderate Resolution Imaging Spectrometer (MODIS) Platform
USGS Landsat Platform

Weather/Climate Datasets
European Center for Medium-Range Weather Forecasts (ECMWF)
NASA Daily Weather Parameters for North America (DAYMET)
NOAA National Center for Environmental Information (NCEI) Climate Forecast System (CFS)
NOAA National Center for Environmental Information (NCEI) Global Forecast System (GFS)
NOAA National Center for Environmental Information (NCEI) North American Mesoscale Forecast System (NAM)
Northwest Alliance for Computational Science & Engineering (NACSE) PRISM Climate Pattern Model

Join The Discussion

Your email address will not be published. Required fields are marked *