This is a repository for data import development of the Soil Spectroscopy for Global Good project.

The OSSL is distributed through API, MongoDB, and as static files hosted on a Google Cloud storage bucket.

The full documentation about the database and access options are available in the OSSL manual.

Importing new datasets

• Define a dataset code following the format in manual project codes and create a new folder inside the dataset/ folder;
• Document import steps in the README.Rmd by producing four standard tables: soilsite, soillab, mir, and visnir. Follow this example;
• Follow the table formats described in the example or summarized in ossl_level0_names_soilsite, ossl_level0_names_soillab, ossl_level0_names_mir, ossl_level0_names_soilsite;
• Make sure to run a quality analysis similar to the other datasets. This includes the assessment of the analytical, spatial, and spectral consistency;
• We are currently running this task on an internal server, so you can host your dataset on a public repository (preferably Zenodo) or share privately (not preferred, soilspec4gg@woodwellclimate.org) and provide your importing steps in this github repository.
• The final integration by joining and binding all the datasets is described in main dataset/ folder and run on our internal server.

In order to maximize transparency, for now, we have decided to produce two different levels for the OSSL database. Level 0 takes into account the original analytical methods employed in each dataset but tries to initially fit them to our reference lists: KSSL Guidance – Laboratory Methods and Manuals (the method/procedures codes are archived in this file) or some ISO standard. If a reference method does not fall in any previous method, then we create a new variable sharing at least a common property name and unit. The final harmonization takes place in the OSSL Level 1, where those common properties sharing different methods can be converted to a target method using some publicly available transformation rule, or in the worst scenario, they are naively bind or kept separated to produce its specific model in the OSSL engine. All the implementations are documented in this github repository.

NOTE: for more advanced users of the soil spectral libraries we advise to contact the original data producers, especially to get help with using, extending, and improving the original SSL data. Note: we do NOT provide support with issues that my arise with the original soil spectral libraries.

The OSSL database

The OSSL includes the USDA NRCS NCSS-KSSL MIR & VNIR libraries (labeled KSSL), ICRAF-ISRIC MIR & VNIR libraries (labeled ICRAF_ISRIC), AFSIS I & AFSIS II MIR libraries (labeled AFIS1 & AFSIS2), ESDAC LUCAS VNIR library (labeled LUCAS), ETH Congo Basin MIR library (labeled CASSL), the Scion Research NZ MIR library (labeled Garrett), the University of Zurich permafrost MIR library (labeled Schiedung), and the Serbian SSL from University of Novi Sad (labeled Serbia). The KSSL database represents a snapshot from July 2022.

• Spectral diversity
  

• MIR locations
  

• VisNIR locations
  

Other tools and repositories

We are fitting globally-applicable prediction models to enable free and open production of primary soil data. The output models are available via the S3 cloud-service and/or API under CC-BY license. If you fit your own models and/or if you are interested in contributing to this project please contact us and help us make better open soil data for global good.

If you follow some minimum requirements e.g. Docker image specifications; open data license with a DOI as in https://doi.org/10.5281/zenodo.5759693), we can host your models so other users can make use of your models and cite/attribute them in the literature. For more info on how to contribute refer to: https://soilspectroscopy.github.io/ossl-manual/.

If you notice a bug or artifact please register an issue and we will try to fix asap.

• OSSL documentation: https://soilspectroscopy.github.io/ossl-manual/;
• OSSL Explorer: https://explorer.soilspectroscopy.org;
• OSSL Engine: https://engine.soilspectroscopy.org;
• Model fitting repository: https://github.com/soilspectroscopy/ossl-models;