| 05 Aug 2026
High resolution mapping of topsoil properties in the Western Balkans using the LUCAS soil survey
Abstract. The Western Balkans are undergoing rapid socio‑economic and environmental change as national policies converge with those of the European Union (EU). In line with the EU Soil Monitoring Law, a robust, spatially explicit baseline of soil health is required to support both national decision‑making and the region’s EU accession process. This study presents the first EU‑wide comparable, high‑resolution (500 m) digital soil maps for five Western Balkan countries (Albania, Bosnia‑and‑Herzegovina, Montenegro, North Macedonia and Serbia).
A total of 955 field‑sampled points from the Land Use and Cover Area frame Survey (LUCAS) soil campaign were analysed for pH (H₂O and CaCl₂), electrical conductivity, organic carbon, carbonate, phosphorus, total nitrogen, extractable potassium, cation exchange capacity and texture fractions (clay, sand, silt, coarse fragments). Gaussian Process Regression (GPR) was selected because it provides a statistically rigorous framework for (i) integrating the same set of ancillary covariates used in EU‑wide soil mapping, thereby ensuring methodological comparability, and (ii) delivering full predictive‑variance surfaces alongside the mean predictions. Results reveal distinct regional gradients and identify the dominant environmental drivers of soil variability across the Western Balkans.
The generated maps constitute a comprehensive baseline for monitoring soil quality, informing agro‑environmental research, and guiding policy development in the context of EU enlargement.
Competing interests: The authors declare no conflict of interest. They acknowledge that the manuscript was assisted by Genesis (the GPT@JRC evolution) to improve readability and clarity in selected sections. All content was subsequently reviewed, edited, and approved by the authors, who retain full responsibility for the scientific integrity and conclusions of the paper. The core analysis, findings, and arguments remain in the original work of the author(s). The use of this tool did not influence the research findings or their interpretation
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
General Comments
This manuscript details a recent soil mapping effort for the data-sparse Western Balkans, building off previous mapping efforts across the European Union. It is presented in a clear and straightforward manner, and I found it to be generally well written and easy to follow. There are a few places where the authors could provide more context and clarity for readers. In general, I feel that this paper, and its associated data products, will be a solid contribution to the understanding of soil properties in this study region with some minor modifications.
Providing a map with key subregions (Vojvodina Region, for example) that are referenced throughout the results and discussion would help serve as a guide for readers who are unfamiliar with Western Balkan geography as they work through the mapped soil property section of the paper.
I felt that this is a paper where the mapped results fit very well into the discussion. It may be appropriate in this instance to integrate the two into a Results & Discussion section, if permissible by journal standards. Normally, I am against this, but I think it would work well for this particular paper.
The authors highlight that a key strength of their approach is that the resulting maps are “uncertainty-aware”, but very little is mentioned about prediction uncertainty in the section going over predicted soil property maps. Why is this? Some elaboration into regions that performed particularly well or poorly would help strengthen the impact of the paper.
Specific Comments
Introduction: The introduction is very brief. This is not a bad thing, but there is room to expand on the motivation of the paper. For example, what tangible impacts have the referenced Ballabio 2016 and 2019 papers had on land management decisionmaking in the EU?
L85: I recommend closing the introduction with a statement of this work’s importance for the region.
Section 2.1: I think it would help the reader to include a table of the specific soil variables being investigated right as the soil dataset is introduced. A very brief description/importance of each variable would also help.
L109: The authors provide references to works that describe Gaussian Process Regression, but they provide almost no description themselves. Providing 1-2 sentences summarizing GPR would help a lot here.
Table 2: It would be interesting to show the performance metrics of both kernels on this table, highlighting those with best performance. Does one greatly outperform the other for any soil property?
L251: Remove space in west-ern.
L300-303: Adding one to two sentences that summarize the natural processes driving C/N variability would be helpful here.
L419-423: This passage is strangely worded. Consider revising it for clarity.
L463: Is this a biodiversity hotspot because of multiple crops, or because it also supports a variety of native flora?