The author’s have a very comprehensive, interesting data set for eight farms in Austria and Iceland under conventional and organic management. There’s much in the paper and in the raw data that SOIL readers will find valuable going forward, and for this reason I recommend it for publication. I would still like to echo the reviewers concerns regarding the experimental design.
The author’s main goal was to test the effects of these two management practices on ‘soil quality’, but the experimental design is troubling and one of the reviewers described clearly the issues with pseudoreplication, and the limitations for inference. I urge the authors to do what they can to add additional organic and conventional sites to the network. The addition of one additional pair of sites in both countries would be sufficient, with the doubling of current sites ideal. These sites are going to presumably be carried forward a long time, and I’m afraid the replication issue is going to hamper their usefulness to the scientific community. I appreciate the author’s being straightforward with the design issues, explaining clearly what they did, and using a reasonable model, but the long-term impact of this study will be limited by site number.

Statistics section – ‘By including farm as a random nested factor, we were able to test differences in soil type (Iceland) and crop type (Austria)’. I don’t think this is correct since differences in crop and soil type among farms within a country were not tested. I think variance was partitioned by farm, but farm effects were not tested.

The suggested corrections by the reviewers and the topical editor should be followed before the paper can be accepted.