This dataset contains all the necessary data to repeat the analyses conducted in the article 'The impact of data quality filtering of opportunistic citizen science data on species distribution model performance'. Per square kilometer in Flanders, the csv-file contains (i) the model predictor values, (ii) the filtered and unfiltered presence-only occurrences of the 255 considered species used for model training and (iii) the presences and absences per species used for model testing. Details on how the data was collected and how the dataset was formed can be found in the article's Methods and Materials and Supplementary Information.

Please read the ReadMe text file for a description of the dataset.