On 24 August, OOI Data Lead Jeff Glatstein, Axiom Data Science Designer Brian Stone, and Axiom Data Science Coder Luke Campbell gave a preview of upcoming additions to Data Explorer that will help users access glider data. The presenters sought input from OOI’s user community to improve the platform to ensure it meets data users’ needs when it goes live in September 2021. You can see the demonstration of the upcoming Data Explorer changes in the video below and hear suggestions from OOI’s data users community.[embed]https://vimeo.com/592362218[/embed]
Hydrographer Leah McRaven (PO WHOI) from the US OSNAP team provided the following CTD resources to help researchers and others better how she and the Irminger Sea Array team are working with the near real-time data being provided by CTD sampling from the R/V Neil Armstrong:
There are four main sources considered in this list:
- Seabird Electronics is one of the most commonly used manufacturers of shipboard CTD systems. Their CTDs allow for integration of instruments from several other manufactures.
- The Global Ocean Ship-Based Hydrographic Investigations Program (GO-SHIP) provides decadal resolution of the changes in inventories of heat, freshwater, carbon, oxygen, nutrients and transient tracers, with global measurements of the highest required accuracy to detect these changes. Their program has documented several methods and practices that are critical to high-accuracy hydrography, which are relevant to many CTD data users.
- The California Cooperative Oceanic Fisheries Investigations (CalCOFI) are a unique partnership of the California Department of Fish & Wildlife, NOAA Fisheries Service and Scripps Institution of Oceanography. CalCOFI conducts quarterly cruises off southern & central California, collecting a suite of hydrographic and biological data on station and underway. CalCOFI has made great effort to document methods that are helpful to those collecting hydrographic measurements near coastal regions.
- University-National Oceanographic Laboratory System (UNOLS) is an organization of 58 academic institutions and National Laboratories involved in oceanographic research and joined for the purpose of coordinating oceanographic ships’ schedules and research facilities.
Instrument care and use
Seabird training module on how sensor care and calibrations impact data: https://www.seabird.com/cms-portals/seabird_com/cms/documents/training/Module9_GettingHighestAccuracyData.pdf
Data acquisition and processing
Notes on CTD/O2 Data Acquisition and Processing Using Seabird Hardware and Software: https://www.go-ship.org/Manual/McTaggart_et_al_CTD.pdf
CalCOFI Seabird processing: https://calcofi.org/about-calcofi/methods/119-ctd-methods/330-ctd-data-processing-protocol.html
Seabird CTD processing training material: https://www.seabird.com/training-materials-download
Within this material, discussion on dynamic errors and how to address them in data processing: https://www.seabird.com/cms-portals/seabird_com/cms/documents/training/Module11_AdvancedDataProcessing.pdf
General overview documents and resources
GOSHIP hydrography manual: https://www.go-ship.org/HydroMan.html
CalCOFI CTD general practices: https://www.calcofi.org/references/methods/64-ctd-general-practices.html
As part of the ongoing the Ocean Observatories Initiative (OOI) effort to improve data quality, OOI is implementing Quality Assurance of Real-Time Oceanographic Data (QARTOD) tests on an instrument-by-instrument basis. Led by the United States Integrated Ocean Observing System (U.S. IOOS), the QARTOD effort draws on the ocean observing community to provide manuals, which outline and identify tests to evaluate data quality by variable and instrument type. Currently, OOI is focused on implementing the Gross Range and Climatology Tests for the variables associated with CTD, pH, and pCO2 sensors. Over the coming months tests will be applied to data collected by pressure sensors, bio-optical sensors, and dissolved oxygen sensors. Ultimately, where and when appropriate, QARTOD tests will be applied to the relevant variables for all OOI sensors.
The Gross Range test aims to identify data that fall outside either the sensor measurement range or is a statistical outlier. OOI identifies failed/bad data with a threshold value based on the calibration range for a given sensor. We also calculate suspicious/interesting data thresholds as the mean ± 3 standard deviations based on the historical OOI data for the variable at a deployed location. As implemented by OOI, the Gross Range test identifies data that either fall outside of the sensor calibration range, and is thus “bad”, or data that are statistical outliers based on the historic OOI data for that location.
The Climatology Test is a variation on the Gross Range Test, modifying the relevant suspicious/interesting data thresholds for each calendar-month by accounting for seasonal cycles. The OOI time series are short (<8 years) relative to the World Meteorological Organization (WMO) recommended 30-year climatology reference period. To help ensure quality, we calculate seasonal cycles for a given variable using harmonic analysis, a method that is less susceptible to spurious values that can arise either from data gaps, measurement errors or from the presence of real, but anomalous, geophysical conditions in the available record. First, we group the data by calendar-month (e.g. January, February, …, December) and calculate the average for each month. Then, we apply the monthly-averaged-data with a two-cycle (annual plus semiannual) harmonic model. Each harmonic is determined using a least-squares fit – a procedure that minimizes the sum of the squares of the differences between the data points and the curve to be fit. This produces a “climatological” fit for each calendar-month.
Next, we calculate the standard deviation for each calendar-month from the grouped observations for the month. The thresholds for suspicious/interesting data are set as the climatological-fit ± 3 standard deviations. Occasionally, data gaps may mean that there are no historical observations for a given calendar-month. In these instances, we linearly interpolate the threshold from the nearest months. For sensors mounted on profiler moorings or vehicles, we first divide the data into subsets using standardized depth bins to account for differences in seasonality and variability at different depths in the water column. The resulting test identifies data that fall outside of typical seasonal variability determined from the historic OOI data for that location.Read More
To provide context and comparison for data collected by OOI instrumentation, OOI collects and disseminates data collected by shipboard underway sensors and from water samples from CTD casts. Shipboard underway data can be accessed by using username and password ‘guest’ on the OOI Alfresco Document Management System, organized by cruise. Each cruise folder contains a Ship Data folder in the format provided by the ship operators and a Water Sampling subfolder. The Water Sampling subfolder includes scanned and digitized versions of the CTD logs, as well as, discrete water sample analyses in the formats provided by the labs which conducted the analyses.[caption id="attachment_20259" align="alignleft" width="199"] Collecting water samples from the CTD rosette on the Pioneer 8 cruise aboard the R/V Neil Armstrong. ©WHOI.[/caption]
To make these data more easily accessible to the science community, we have developed a common template to provide a full set of discrete water sample data from a cruise. These “Discrete_Sample_Summary” spreadsheets include the details for each Niskin bottle fired on a CTD cast, the CTD instrument rosette data from the time of bottle closure, and the water sample data and quality flags based on World Ocean Circulation Experiment (WOCE) standards.
These CSV files with common data formats can easily be read and manipulated in MATLAB, Python, or other computing programs and languages. Because water analysis data are received at different times from different labs, these spreadsheets are updated as data become available. An accompanying README file contains version history, general notes, and a description of the quality flags. The original spreadsheets from labs, which may contain additional data and methodology, will also be posted.
An example of how to read and use this discrete sample data can be found in this Jupyter notebook. Discrete_Sample_Summary spreadsheets have been posted for the Regional Cabled Array cruises 6-10, the Coastal Endurance Array cruises 1-13, and the Global Irminger Sea Array cruises 1-6. We will continue to work on completing spreadsheets for past cruises as well as cruises going forward.[caption id="attachment_20261" align="aligncenter" width="640"] Comparison of dissolved oxygen data on the Washington Shelf Surface Mooring with water sampling data from Endurance Cruise 13. Data from Deployment 10 and Deployment 11 are plotted together, and overlap during 5-7 July.[/caption] Read More
It all started with an email blast advertising a workshop offering opportunities to develop new lessons using OOI data. By happenstance, the workshop was right down the street from Kathleen Browne’s office at Rider University in Lawrenceville, New Jersey, making it convenient for her to attend. Sponsored by the National Science Foundation-funded OOI Ocean Data Labs at Rutgers University of New Jersey, the workshop’s focus on “data explorations” meshed nicely with Browne’s long held involvement in advancing science education plans for both higher ed and K-12 teachers. Browne teaches oceanography and served as academic director of Rider’s Science Education and Literacy Center, which works to advance K-12 and college math and science instructors’ use of inquiry-based instruction.
“Since I have been involved in a whole suite of science education lesson plans and I teach oceanography, it seemed like a unique and worthwhile opportunity, “ she explained.
Little did Browne expect that a five-day workshop would lead to her becoming an integral part of the OOI Ocean Data Labs network. She and Gabriela Smalley, a colleague in Rider’s Geological, Environmental and Marine Sciences Department (which Browne chairs) and colleagues from Maine Maritime Academy (Lauren Sahl), University of Kentucky (Rebecca Freeman), Southern Maine Community College (Carol White), and Rutgers (Sage Lichtenwaler) created an OOI data exploration on ocean anoxic events. Browne and Smalley have since also contributed to a new lab manual created by the Ocean Data Labs. Browne created labs in geology, while Smalley contributed labs on primary production and biology.
The pandemic has changed the way Browne is considering classroom instruction. “I’m in the midst of rethinking how I teach labs, particularly given all I’ve learned in these days of remote instruction. It’s hard to teach an oceanography lab remotely when students are clearly looking for something different than what we can do remotely. I imagine, for example, we might use the intro labs in the lab notebook in the beginning of the semester because we are finding that students really need focused guidance on describing patterns in data sets before they can make sense of the results.
“Students struggle with understanding data. But, the need for data analysis is not limited to science. It’s important in politics, health care, and most aspects of life today. We live in a world where an enormous amount of data is presented to us daily. It’s not that everyone needs to process that data, but they do need to at least make sense of the data visualizations that might be offered. It’s my hope that my students can do that so they might be able to recognize when something doesn’t make sense. I’d like my students to be better consumers of data, and skeptics of data so they are not misled.”
Browne‘s interest in the business of students struggling to describe patterns in data has prompted a project to figure out the best instructional ways to help students develop data literacy skills to improve their scientific explanation skills. She teamed with Smalley, Andrea Drewes, a learning scientist in Rider’s Graduate Education, Leadership and Counseling department, and Sage Lichtenwaler, a lead investigator and data wrangler with Ocean Data Labs, on a recently funded NSF project. The project will assess students’ data and science literacy skills — that is understanding the scientific principles involved. The project will use a variety of techniques and resources, including the Ocean Data Lab’s data explorations.
“Our approach will be to help students develop their data literacy skills such as the ability to read a graph correctly and identify and articulate in detail the patterns and trends that they see in the data. With those skills refined we will help to help students learn the science in our courses so they are able to explain something they learned about the data and the science using scientific evidence and reasoning.”
The effectiveness of this approach will be tested in two different classroom settings. One will serve as a control, while the other will integrate tools and instruction to help students work with data visualizations. Ultimately, the team will assess the learning gains in the control and intervention classes. Successful techniques identified will be shared broadly with science instructors, learning scientists, and other educational researchers throughout the nation to improve data literacy among U.S. students.
Browne and her team will begin the formal data collection this fall and expect to report their findings in 2023. “My work with OOI data and the Ocean Data Labs has opened up new avenues of exploration for me, and ultimately will help students’ abilities to use data and science literacy skills to contribute as active citizens to pressing ocean-related issues in the world.”Read More
The Ocean Observatories Initiative (OOI) includes sensors that measure key biogeochemical properties (pH, pCO2, bio-optics, nitrate, dissolved oxygen) on both moored and mobile autonomous platforms across arrays in the Atlantic, Pacific and Southern Oceans. These sensors provide enormous potential to support the oceanographic community in studying a wide range of important interdisciplinary questions. However, OOI biogeochemical sensor data have thus far been underutilized by the oceanographic community, as the application of these rich data streams to quantify biogeochemical fluxes and answer many questions of scientific interest (e.g., rates of air-sea CO2 flux, productivity, and export; comparison across sites; monitoring of long-term changes) require effective calibration and validation, including post-deployment human-in-the-loop processing. To broaden the use of OOI biogeochemical sensor data and increase community capacity to produce analysis-ready data products, we have acquired NSF support to bring together scientists with expertise in biogeochemical sensor calibration and analysis from both within and beyond the current OOI user community to develop guidelines and best practices for using OOI biogeochemical sensor data. These recommendations will be collated in a published white paper that will be shared with the broader oceanographic community to build data user capacity and enable new scientific applications of OOI biogeochemical sensor data.
This activity was initially planned as a small workshop in conjunction with the 2021 Ocean Carbon & Biogeochemistry (OCB) summer workshop (June 2021 in Woods Hole, MA). Given the uncertainties related to the pandemic, we would like to identify potentially interested participants early and query their preferred level(s) and mechanism(s) of engagement with this activity to help inform our planning efforts. We anticipate that the development of best practices for different biogeochemical variable sets will require several months of commitment, likely a combination of online engagement and participation in an in-person workshop that will either take place in 2021 or 2022. We envision that members of the oceanographic community may have varied levels of interest in such an activity (e.g., development of detailed OOI sensor data processing guidelines vs. broader scientific applications of OOI biogeochemical data streams). If you have any interest in potentially participating in this activity at any level, please fill out this Google form. Responses received by mid-October are greatly appreciated.
Additional information and questions about this activity can be directed to Hilary Palevsky (firstname.lastname@example.org), Sophie Clayton (email@example.com), and Heather Benway (firstname.lastname@example.org).
On 16 September at 4 pm Eastern, Ocean Data Labs will kick off it fall webinar series: Ocean Data Labs Plus, a webinar series for Community College and University Professors teaching oceanography or geosciences courses. The series opener will be “New You Can Use,” hosted by the OOI Data Labs Project Team and special guests.
The webinar will explore how the Data Labs Project can support your efforts to introduce big data into your undergraduate courses. Join the Team to find out about newly-developed interactive online data-focused activities that are grounded in learning science – and consider how to effectively incorporate them into your courses. Check out the Ocean Data Lab’s online collection of data explorations and data nuggets in advance, and bring your questions and ideas. Each webinar will last about 60-75 minutes and is meant to be more of an interactive discussion.
To help make OOI data more accessible, useable, and easily integrated into research and classrooms, the OOI data team has spent the last year developing a new tool that will allow users to discover the data required to meet their needs. The new “Data Explorer” has been undergoing user testing for the past three months and will be ready for broad distribution in early October.
Data Explorer will allow users to search and download cabled, uncabled, and recovered data, compare datasets across regions and disciplines, generate and share custom data views, and download full data sets using ERDDAP.
“We are really excited about the launch of Data Explorer version 1,” explained Jeffrey Glatstein, head of OOI’s cyberinfrastructure, who guided a team of data and visualization experts in developing the tool. “It is an excellent tool that offers a variety of ways for researchers, educators – and others interested in ocean observations—to use OOI data to answer their ocean observation data questions. What we’ve learned over the test period is that additional refinements will be needed, but we first want input from the community, to help identify and prioritize the most important ones. Subsequent versions will build upon this first release to make the Data Explorer the primary tool to explore OOI data.”
Data Explorer contains physical, chemical, geological, and biological ocean observation data collected in near real time. Glatstein and the OOI Data Team worked with Axiom Data Science to develop a system that is both powerful yet user friendly.
Stay tuned. The launch is set for the beginning of October 2020.
The OOI Ocean Data Labs team is looking for instructors of introductory oceanography courses to “test drive” a collection of new online laboratories that focus on important oceanographic themes and topics using OOI data.
They are seeking a pilot implementation team of 14-16 faculty to implement two labs with students this Fall. They are offering a $750 stipend, which includes a training webinar, detailed feedback and evaluation, and a wrap-up session. The implementation must be completed by December 2020.
Beginning in January 2020, a team of faculty contributors compiled a sequence of OOI Data Labs into an online laboratory manual. It includes topics in biological, chemical, physical and geological oceanography for use in typical Introductory Oceanography courses. The manual is a collection of eight lab exercises, with built-in assessments, and accompanying instructor guides.
Applications are due Sunday 9 August 2020. Apply here.Read More
Two weeks. That is the amount of time Janice McDonnell and Sage Lichtenwalner, Co-PIs of the OOI Ocean Data Lab Project, had to create an eight-week intensive, hands-on virtual program for Research Experience for Undergraduates (REU) students, who couldn’t attend their original programs due to COVID-19 restrictions. McDonnell and Lichtenwalner jumped in with both feet and successfully pulled together a program for 16 undergrads from 16 institutions that will wrap up on July 31st. Not only did they have to develop a curriculum in short order, they recruited 17 mentors, who provided one-on-one mentoring for each of the REU students.
In 14 days, McDonnell and Lichtenwalner, working with colleagues at the Rutgers RIOS REU, developed a curriculum, which included two two-hour workshop sessions on Zoom every day for the first two weeks. The workshop was followed by six weeks where students focused on a research project using oceanographic data to answer a scientific question. They were helped by REU leaders and mentors along the way.
The initial workshop focused on students using a “baby data set” before working on something as complex as an OOI dataset or other similarly large dataset. “Initially, we gave them mini research projects that could be done in two weeks. This allowed the students to be collaborative and interactive, while learning Python as a tool using data to answer questions about the ocean,” explained McDonnell.
The key to the program’s success was keeping it engaging. The curriculum mixed up content, approach, and activities to ensure students stay involved on and off the screen. The team used a tactical approach – starting where students are and building upon their level of understanding. They incorporated a lot of different approaches, including think-pair-shares and an activity from the Right Question Institute, another NSF-funded project, which guides people to ask better research questions.
“We were constantly looking for opportunities to be interactive, reflective, and to give students the opportunity to apply their knowledge,” added McDonnell.
Another technique used were Zoom breakout groups to supplement group interaction. These breakout sessions provided students with opportunities to work together, with mentors, and to get to know each other.
Breakout sessions were also used to meet members of a career panel in virtual personalized sessions.“The career panel for this REU cohort was much more diverse than it is typically. The virtual nature of the interaction made it possible for people from all over the country to join in and participate,” said Lichtenwalner. “We facilitated it in a nice round robin sort of way using Zoom’s breakout function. Students chose their top choices, then met with them either individually or with another student. This gave them an incredible opportunity to meet people in careers that they might not otherwise have access to.”
“I’m really proud of our REU,” said McDonnell. “It’s not easy to teach online but good learning can, and does, happen online and we were able to do that. And the collaboration that took place was really the secret sauce for making this all work. NSF program officer Lisa Rom and Science Assistant Rennie Meyers were committed to and worked really hard to find a solution to make the REU program happen in the middle of a pandemic. The students were great and so excited about the opportunity that we put together for them. And, the mentors went above and beyond the call of duty to help make this program work for the students.”
Each week, for example, one mentor, Dr. Jessica Carriere-Garwood, of Rutgers University introduced her students to people in her professional circle at lunch time each week, with the opportunity to talk about what the students are doing and their interests. “That doesn’t always happen in a regular REU. There are a lot of pluses and minuses of being virtual and this was certainly one of the pluses,” added McDonnell.
Ed Dever, one of the mentors from Oregon State University, had this to say about his mentoring experience: “Janice and Sage did a remarkable job spinning this REU up on short notice. They mentored not just the students, but the mentors (well, at least this mentor!). Janice, Sage, and Christine Bean did an amazing job of building a virtual community of students and mentors on the fly. This virtual community provided a unique experience to the students in that the community was much broader than an in-person REU at a single institution. It took all of us out of our comfort zones and helped us grow. Throughout the whole process, Sage patiently guided REU students in technical aspects of using Python to access and analyze data.”
The REU cohort will finish their research projects and present on 30-31 July. Of the 16 students, eight are using data from the OOI.
Citing the success of this first virtual REU, the team’s National Science Foundation (NSF) program officer Rom pondered “Why don’t we do this all the time, even if there isn’t a pandemic?“
Based on the successful experience this summer, NSF’s Division of Ocean Sciences is encouraging REU proposals for virtual REU’s, and especially those that use OOI data. The deadline is August 26th this year. Apply here.Read More