A dataset for monitoring agricultural drought in Europe

Abstract

This paper introduces the Combined Drought Indicator dataset, a collection of raster maps generated by the Copernicus European Drought Observatory for monitoring agricultural drought in Europe. Computationally, the CDI involves three indicators: the Standardized Precipitation Index, the Soil Moisture Anomaly and the Fraction of the Photosyntetically Active Radiation anomaly. These are complemented by the use of crop and snow masks. The CDI dataset has a spatial resolution of 1/24 decimal degrees (∼5 km), a temporal resolution of 10 days and is available from 2012 onward. As drought effects are variegated both in space and time, the CDI provides an effective instrument for assessing the different stages of propagation of agricultural drought and their spatial extent. Furthermore, the CDI maps provide relevant information for those private and public actors (water resource agencies, farmers, land managers and so on) involved both in drought preparedness and planning to mitigate drought impacts. Users can access and download the dataset from the Copernicus European Drought Observatory web portal, where an online mapviewer and clickable maps facilitate its interactive exploration.

Fig. 1

The computation of the CDI involves three input indicators: SPI, SMA and FAPAR anomaly. From a mathematical point of view, the three input indicators are represented in terms of symmetric, bell-shaped distributions with zero mean and unit standard deviation. The CDI monitors if one or more indicators (at the 10-day period T) are below their corresponding reference thresholds. This information is then integrated with the CDI class at the 10-day period T-1 in order to classify each pixel of the study domain according to a seven-category qualitative scale. Note, the diagram shows the SPI calculated on two distinct time scales: ∼30 days (SPI-1) and ∼90 days (SPI-3).

Fig. 2

An example of input indicators (FAPAR anomaly, SMA, SPI-1 and SPI-3) used in computing the Combined Drought Indicator for the second 10-day period of April. Note, the four indicators have dimensionless units.

Fig. 3

10-day periods with maximum snow cover extent in Europe in 2016, 2017, 2021 and 2022. Europe's 2015/2016 winter was characterized by the lack of snow for all of Europe (top-left map). January 2017 was a period of exceptionally cold and snowy weather in Eastern and Central Europe (top-right map). In January 2021 an unusual heavy snowfall in Spain was brought by the Storm Filomena (bottom-left map). In January 2022 the cold weather hit Athens and forced the closure of Istanbul airport (bottom-right map).

Fig. 4

Combined Drought Indicator version 3 (upper panel) and version 4 (lower panel) for Finland, Sweden and Norway during the 1st to the 18th 10-day period in 2020 (January - June).

Fig. 5

Static crop masks for Finland, Sweden and Norway during the 1st to the 18th 10-day period in a year (January - June).

Fig. 6

Dynamic snow masks for Finland, Sweden and Norway during the 1st to the 18th 10-day period in 2020 (January - June).

Fig. 7

Alluvial plot of the Combined Drought Indicator for Finland, Sweden and Norway during the 1st to the 36th 10-dayperiod in 2020 (January - December).

Fig. 8

Combined Drought Indicator for France during the 22nd to the 30th 10-day period in 2020 (August - October).

Fig. 9

Alluvial plot of the Combined Drought Indicator for France during the 1st to the 36th 10-day period in 2020 (January - December).

Fig. 10

Standardized Precipitation Evapotranspiration Index (SPEI-3) from May to August 2022.

Fig. 11

Combined Drought Indicator (CDIv4) from May to August 2022.