Copernicus Global Drought Observatory (GDO): http://edo.jrc.ec.europa.eu/gdo © European Commission, 2019.

- 1 -

GDO INDICATOR FACTSHEET

Soil Moisture Anomaly (SMA) This Factsheet provides a detailed technical description of the indicator Soil Moisture Anomaly (SMA), which is implemented in the Copernicus Global Drought Observatory (GDO), and used for detecting and monitoring agricultural drought conditions. The variable of the hydrological cycle upon which the SMA indicator in GDO is based, as well as the indicator’s temporal and spatial scales and geographic coverage, are summarized below. Examples of the SMA indicator are shown in Figure 1.

Variable Temporal scale Spatial scale Coverage

Soil moisture 10 days (= 1 dekad)

0.1 degree Global

Figure 1: Examples of the Soil Moisture Anomaly (SMA) in GDO, highlighting the conditions of water

stress (negative anomalies) in South Africa, during the Cape Town drought at the end of 2017.

30 Sep 2017

31 Dec 2017

Copernicus Global Drought Observatory (GDO): http://edo.jrc.ec.europa.eu/gdo © European Commission, 2019.

- 2 -

The Soil Moisture Anomaly (SMA) indicator that is implemented in the Copernicus Global Drought Observatory (GDO) is used for determining the start and duration of agricultural drought conditions, which arise when soil moisture availability to plants drops to such a level that it adversely affects crop yield, and hence, agricultural production. The SMA indicator in GDO is derived from anomalies of estimated soil moisture (or soil water) content - which are produced as an ensemble of three datasets: the JRC’s in-house LISFLOOD hydrological model (de Roo et al., 2000), the MODIS-derived land surface temperature (Wan and Li, 1997) and ESA combined active/passive microwave skin soil moisture (Liu et al., 2012). Details on the use of these datasets for drought detection can be find in Cammalleri et al. (2017).

Soil moisture (or soil water content) is an important variable for plant growth, and - together with precipitation and evapotranspiration - is a basic component of the hydrological cycle. The SMA indicator in GDO is computed as an anomaly from the climatological reference period, and it is updated 3 times a month (after the 10th, the 20th and the last day of the month). The SMA indicator is used to detect and monitor agricultural drought, which is one of three main types of drought that are defined according to the variables of the hydrological cycle (i.e. precipitation; soil moisture; groundwater; streamflow) mostly affected. Meteorological drought is a prolonged period of less than average rainfall in a given region, which generally precedes agricultural drought, when there is reduced crop production due to insufficient soil moisture, and hydrological drought, when there is below-normal water availability in rivers, streams, reservoirs, lakes, or the groundwater table.

The SMA indicator implemented in GDO is computed based on 30-day average anomalies (updated every dekad) of three main datasets: 1) the daily soil water content produced by the JRC’s LISFLOOD hydrological model at 0.1 degree resolution. 2) the 8-day land surface temperature (LST) as derived from the acquisition of the MODIS satellite sensor at 0.05 degree resolution (MOD11C2, collection 6). 3) the dekad combined active/passive skin soil moisture product produce within the ESA Climate Change Initiative (CCI) and distributed by C3S at the spatial resolution of 0.25 degree. The three datasets are spatially interpolated over the same regular global grid of 0.1 degree, and 30-day anomalies are computed separately 3 times per month (corresponding to 36 yearly updates). Each datasets utilizes its own baseline dataset, all of which are referring to the same climatological period (2001-2017). The three anomaly datasets are merged into a single product by means of a weighted average, which weights are derived from a Triple Collocation (TC) analysis (see Cammalleri et al., 2017):

SMA = 𝑤LIS𝑧LIS +𝑤LST𝑧LST +𝑤CCI𝑧CCI where the w terms are the weighting factors and the z terms are the anomalies. LST dataset is converted into a water available index (i.e., high LST correspond to low soil moisture), hence negative anomalies represent dry conditions.

Copernicus Global Drought Observatory (GDO): http://edo.jrc.ec.europa.eu/gdo © European Commission, 2019.

- 3 -

The GDO MapViewer allows visualizing the last available SMA map, as well as the past archive (see Figure 2). These maps provide information on the spatial distribution of soil moisture anomalies, and their evolution over time. Soil moisture content can be used as a direct indicator for determining the start and duration of agricultural drought conditions. In fact, soil moisture content provides an assessment of the soil water availability for plants’ needs. Soil moisture content is also obviously related to plant biomass accumulation (i.e., gross primary production) in many environments (e.g., dry, semi-arid and arid) where water availability is the main limiting factor. The maps of soil moisture anomalies can be used as a “proxy” for the presence of potential drought conditions. Of course, the presence of actual water stress conditions will also depend on the specific plants’ resistance and capacity for water extraction from the soil matrix.

Figure 2: Maps of Soil Moisture Anomaly (SMA) for the third update of June 2018 (last dekad),

produced by the processing chain in the Copernicus Global Drought Observatory (GDO).

Copernicus Global Drought Observatory (GDO): http://edo.jrc.ec.europa.eu/gdo © European Commission, 2019.

- 4 -

Strengths: The 10-day (i.e., dekad) update of the SMA indicator in GDO provides continuous information

on the soil moisture content deviation from the climatology, and on the spatial extension of the area affected by or at risk of drought conditions. Moreover, analysis of a time-series of SMA indicators can be used to estimate the duration and the severity of drought.

The meaningful combination of three datasets (weighted average) should minimize the limitations related to the use of a single estimation method.

Weaknesses: The reliability of SMA estimates relies on the accuracy of the three base products. The LISFLOOD

accuracy at global scale is constrained by the accuracy in model forcing and calibration parameters. The LST data here used are just as proxy of soil moisture conditions, and the microwave estimations may be unreliable over densely vegetated areas.

The Triple Collection (TC) method provides accurate estimates of the weighting factors only under very strict hypotheses.

o Cammalleri, C., Vogt, J.V., Bisselink, B. and A. de Roo. 2017. Comparing soil moisture anomalies

from multiple independent sources over different regions across the globe. Hydrol. Earth Syst. Sci. 21, 6329-6343. https://doi.org/10.5194/hess-21-6329-2017.

o de Roo, A., Wesseling C. and W. van Deursen. 2000. Physically based river basin modelling within a GIS: the LISFLOOD model, Hydrol. Process. 14, 1981–1992. https://doi.org/10.1002/1099-1085(20000815/30)14:11/12<1981::AID-HYP49>3.0.CO;2-F

o Liu, Y.Y., Dorigo, W.A., Parinussa, R.M., de Jeu, R.A.M., Wagner, W., McCabe, M.F., Evans, J.P.

and A.I.J.M. van Dijk. 2012. Trend-preserving blending of passive and active microwave soil moisture retrievals. Remote Sens. Environ. 123, 280-297.

o Wan, Z. and Z.-L. Li, Z. 1997. A physics-based algorithm for retrieving land-surface emissivity

and temperature from EOS/MODIS data. IEEE Trans. Geosci. Remote Sens. 35, 980-996.