Archive for the ‘Data and statistics’ Category

Looking for data and data repositories? Check out!

During my search for a suitable repository for publishing data and code of our potato trade network analysis, @hauschke and @pampel pointed me to is a global registry of data repositories with the mission to

present repositories for the permanent storage and access of data sets to researchers, funding bodies, publishers and scholarly institutions. In the course of this mission aims to promote a culture of sharing, increased access and better visibility of research data.

Currently, contains over 600 data repositories which are accessible either by browsing through the catalogue or via a search facility. A really useful resource!


Really a Nontraded Commodity? – Datasets and code now available

05/12/2013 4 comments

Figure 1: Trade network for fresh potatoes.The most successful contribution to this blog is (a little to my own surprise) our analysis of the international potato trade network Really a Nontraded Commodity?

For all those who are interested in what’s behind that work, or who want to take the analysis further, or simply want to know how these graphs have been produced we have now made the matrices of bilateral trade in fresh, frozen, and seed potatoes as well as the R code publicly available.

All data comes with a Creative Commons license and is openly accessible at the zenodo repository.

(Thanks, zenodo, for this great service!)

Crop model calibration with yield trial data: Dealing with missing soil data

A previous blog post discussed how to deal with missing weather data when calibrating a crop model with incomplete yield trial data. A further common problem associated with the use of field trial data from crop breeders for the purpose of calibration of the DSSAT-SUBSTOR  potato model is missing or insufficient soil data.

The DSSAT soil water and nutrient routines require information of the soil found at the site, consisting of a broad range of soil physical and chemical parameters taken from different depths of the soil. The soil data provided along with yield trial data, however, often only consists of information on pH, nutrient availability, organic matter content and soil texture, taken from a sample at one single depth. In this situation, how can we obtain a complete soil profile to be used in DSSAT? Read more…

Two interesting sources for soil data and maps

Many biophysical modeling efforts require soil data. There are (at least) two interesting sources that provide such information.

The first one is, an initiative which seeks to create a new digital soil map of the world, applying state-of-the-art soil mapping technologies.

The second one is the European Soil Portal of the Joint Research Center of the European Commission. The European Soil Portal provides access to a very rich compilation of soil related information, among them datasets, soil maps and a map viewing application. The great thing about this portal is that it not only focuses on Europe, but also covers other parts of the planet, like for example Africa.

[Many thanks to Thordis for pointing me to the European Soil Portal]