b) Continuing Projects -> Current status and project results to date -> 2) Technical Data Integration

Based on the characteristics of the surface, midwater and benthic data sets and the analysis that had been done to date during the science discussions, in 2015 the architecture in Figure 1 was proposed as a means to curate, integrate and serve the various time series data.  As of the writing of this proposal, the boxes that are solid red have been done or are stable and being augmented as data analysis continues.  The boxes in dashed red are in the early stages of development with more work to be completed in during the rest of 2016.

Figure 1 would go here
[Caption - Diagram of proposed integrated data architecture for ITS project]
Software was developed to take in the specific listing of data that is applicable to the science questions being asked in the ITS meetings and loaded it into a MongoDB database.  MongoDB was chosen to to its ability to embed aggregation and data processing directly into the data storage.  Access APIs were then developed to allow HTTP requests from clients to access this combined data set.  Several external data sets were also discussed during the ITS meetings and the loading software was written to include those data in the integrated data set as well (MEI for example).  Current work is focused on building appropriate aggregation and analysis algorithms to provide more specific views of the raw data and on a visualization application to explore the data sets (see Figure 2)

Figure 2 would go here
[Caption - Early stages of the data visualization and exploration interface to the integrated data]

c) New and Continuing Projects: Plans for 2017 -> Engineering/technology development

In 2017, we are proposing to continue the work on the integrated data storage access and visualization software by focusing on:
	1. Integration of additional external data sets.  During our discussions, many other data sets have been discussed (mooring, ROVCTD, etc.).  This work would write additional software to include that data in the integrated data storage.
	2. Work will continue to expand the embedded aggregation and analysis in the MongoDB database itself, but there will be additional work to design maintainable ways to integrate other data analysis packages like Matlab and R.  This work will also be happening in both the CANON/ODSS work and the ESP project so it will be highly leveraged work.
	3. During 2016, some of the data analysis and visualization produced by Monique have expanded beyond the capabilities of the HighCharts package being used in the currently deployed web application.  This work proposes investigating and replacing the HighCharts package with the D3 package which has vastly greater charting and visualization capabilities.
	4. The last section of work is to build a mechanism to allow users to integrate and visualize their own data sets along side the Integrated Time Series data sets.
	
Total hours: 250

The green boxes in Figure 3 are the areas of proposed work.

Figure 3 would go here
[Caption - Integrated Time Series data architecture with four highlighted areas of proposed work (1-4)]