Impact of Open Landsat Data on Science

Organisator Laboratory of Geo-information Science and Remote Sensing

di 10 maart 2015 09:00 tot 09:30

Locatie Lumen, building number 100
Droevendaalsesteeg 3a
6708 PB Wageningen
+31 317 481 700
Zaal/kamer 2

by Puja Mishra (Nepal)


The tendency of open data has been grown exponentially all over the world in the last decade. Data provides the indication or evidence for the published body of scientific progress and knowledge (journals). The transparency, reusability, availability and reproducibility of scientific data can increase the scientific process towards the societal benefit. In order to increase transparency, reusability, participation and/or government efficiency, governments around the world start to pay attention to the opening of their data. In line with those trends, U. S. Geological Survey announced in 2008 the opening of Landsat data to make all the Landsat data archive publicly and freely available. After this open data policy, users are allowed to access, freely download and use Landsat imagery products. It is very likely that this initiative have had a significant impact on society, especially on science. However, there is no comprehensive study found that can assess the impact of open Landsat data on science.

In this context, this research aims at assessing the impact of open Landsat data on science. Based on literature review, the possible research methods are investigated to meet the assessment requirements of Landsat data archive. Finally, a fitness for use criteria is used to select the most suitable research method to measure the impact of open Landsat data on science. This assessment uses Systematic Literature Review (SLR) method which focuses on the six indicators to measure the effects of open Landsat: 1) number of publication in which Landsat data used per year; 2) number of publication in different document type; 3) number of publication in different languages; 4) number of publication that use Landsat data per country per year; 5) number of publication in which Landsat data used in different disciplines per year; and 6) number of images used in publication. According to these indicators, search query is constructed to extract the scientific publications in Scopus digital data source from 1972 to 2014. Research tendency was investigated by statically analysing the distribution of publication year, authors, authors with affiliation, title, source title, citation, published country, document type, discipline and language type. After opening Landsat archive from 2008, the result shows that the trend of Landsat use has significantly increased with increasing number of articles (69.43%) and books (0.65%). While production of the conference papers (29.06%) and reviews (0.86%) were relatively reduced almost by half. English language of those journals took the majority of all the publication however after free data distribution policy the publications number of publications in other national languages have slightly increased to languages such as Arabic, Chinese, Korean, Persian, Thai, Ukrainian, Turkish, Spain, Portuguese, Italian, and Finnish, Estonia, Serbian. Similarly, the result shows that after Landsat opening archive the Asian region (mainly China) became the biggest user of Landsat data. Overall, it has also been found that agricultural, ecological/ecosystem science/management; forest science/management, water resources, land use/land cover, emergency/disaster management/hazard insurance, urban planning and development were dominant application areas of Landsat. Before opening Landsat archive, the scientists were limited to the use of data for large areas. It has been observed that the use of multiple Landsat images using time series and near real time analysis for earth observation and monitoring in large areas have increased significantly. To test the search result of Scopus using SLR method, 3% of Landsat publications (501 in total) from 2002 to 2014 were randomly read by abstract, title and keywords. The validation accuracy of five indicators number of publication in which Landsat data used per year; number of publication in different document type; number of publication in different languages; number of publication that use Landsat data per country per year; number of image used in publication were recorded with the highest accuracy (100%), whereas Landsat publication in different discipline was found to have lowest overall accuracy (68.17%). The results suggest that SLR method can help the researchers to assess the overview of global impact of open data on science.

Keywords: Open data; Landsat; science; impact assessment; systematic literature review.