Today’s LYD post is by Amy Koshoffer, Science Informationist based at the Geology Math and Physics Library with editorial support from Dr. Eric J. Tepe, Assistant Professor of Biology and Curator of the Margaret H. Fulford Herbarium.
It has been sometime since I stepped over the threshold of my old lab in the Care/Crawley Building. Many changes occurred in the interim including a move to another floor of the building. There are times I miss the bench research and the data I created in my time as a senior research assistant. One of my favorite techniques was microscopy and particularly Electron Microscopy (EM). I remember the multitude of samples processed, the long wait for samples to be ready to image and then finally all the amazing images we captured. Processing samples for EM imagining is a long and sometimes challenging technique. The samples need to be dehydrated and then infiltrated with a resin to stabilize the structures and prevent destruction from the electron beam during viewing. You might not know if a sample has been ideally preserved until you get to the imaging lab and begin to examine the sample. But what joy when the images look amazing with crisp detail and no water holes. So much work and resources went into the sample preservation and acquiring images.
I wonder what will happen to that effort in the years and decades to come. Are there others who might want to use the physical samples and digital images in their own work? Did I do what was needed to make sure that someone could reuse all the data created?
How does one love data? In one way, loving your data means respecting your data legacy. Often, research impact is measured in articles published and grants received. The articles written are the interpretation of data gathered to address a set of research questions. The data behind the articles isn’t used up, and have the potential to continue to contribute to new questions from other researchers. So through the data, a researcher can continue to have long-term impact. This is what is meant by data legacy. Love for data means following best practices in data management, sharing, and preservation that will allow for data reuse. Many key actions for good data management, such as choosing a steward for the data, saving data in open formats, and creating good documentation, have been highlighted in this week’s LYD17 posts and tweets (#lyd17). But what if good, or even any, documentation is not available, and what if the person who created the data is not accessible? In the example of my EM work, the samples and images could have easily been tossed out during the lab move, or deleted from a computer if future lab members cannot identify the samples or images. Data is at risk of being lost or destroyed if not well managed. How can unloved data be rescued from this fate?
Rescuing data that has not been well loved (i.e. well managed) takes detective work and perseverance. Several projects to rescue truly unloved data are described here at the LYD website. Rescuing data also means bringing it into the light of day (i.e. putting data in a place where others can find it), and in a format that is useable. In a time when “to google” means to find, making analog data digital also increases data discovery through portals such as research data repositories. Repositories help researchers find what they are looking for through description (i.e metadata) and other documentation, guidance for preservation best practices, migrating to the most currently used file formats, and through back-ups to prevent data loss because of a storage system failure. Description comes from the curation process which generates quality metadata (information about the data such as who was the creator or when was the data created) about the data being preserved. UC Biology Faculty Eric Tepe, Steven Rogstad, and Theresa Culley are under taking just such a digitization and curation project for the Margaret H. Fulford Herbarium which houses ca. 125,000 physical specimens of vascular plants, bryophytes, lichens, fungi, and algae, as well as supporting documents. These researchers are supported by an NSF grant (ADBC-1410548) to database and digitize label data of the lichen and bryophyte collections held by the herbarium. The specimen labels include information such as genus, species, collector, collection date, and details about specimen location such as geospatial data like GPS coordinates or site characteristics, and partner herbaria. Once digitized, the data is transferred to both our own institutional repository at Scholar@UC (http://scholar.uc.edu) and a national repository for biodiversity data at iDigBio (https://www.idigbio.org/). The goal is to digitize all of the Natural History collections held by the herbarium, and other supporting data, resulting in a fully accessible collection that is available to the public and the greater research community. Efforts such as the Herbarium Digitization Project show us a path to follow to make other kinds of data available to all, and to prevent it from being lost and unloved and, ultimately, in need of rescue.