Global Time Series of Annual Nitrogen Stable Isotope Maps Derived from Landsat: 30m from 1984 

We developed a novel dataset that calculates the annual δ15N globally based on Landsat reflectance from 1984 with planned updates for coming years using Google Earth Engine. The resulting spatio-temporal data include four channels: (i) mean δ15N values; (ii) prediction uncertainty; (iii) the number of Landsat retrievals used; and (iv) a data quality assessment. A web-based application provides code-free access and visualisation to the dataset: https://tern-landscapes.earthengine.app/view/nitrogen-availability.

We compiled the δ15N dataset from five sources globally. We obtained six atmospherically corrected surface reflectance bands from Landsat using Google Earth Engine. We obtained the quality assessment channel (QA) and eliminated data containing clouds or cloud shadows. The peak of growing season reflectance were selected based on greenness.

We fit a random forest model to predict measured δ15N from Landsat reflectance: δ15N ~ ρRed+ρGreen+ρBlue +ρNIR+ρSWIR1+ρSWIR2. We the use the fitted model to predict annual δ15N from 1984.
 

We at TERN acknowledge the Traditional Owners and Custodians throughout Australia, New Zealand and all nations. We honour their profound connections to land, water, biodiversity and culture and pay our respects to their Elders past, present and emerging. 

The dataset aims to provide annual global monitoring of foliar δ15N from 1984. The maps could inform the improvement of nitrogen cycle models and the assessments of the impacts of environmental changes.
 

The details of the methods can be found in, Yang, J., Zhang, H., Guo, Y., Donohue, R. J., McVicar, T. R., Ferrier, S., et al. (2025). Globally mapping the nitrogen stable isotope ratios of terrestrial vegetation from 1984 to 2022. Earth's Future, 13, e2024EF005836. https://doi.org/10.1029/2024EF005836