Use of soil and climate data to assess the risk 
of agricultural drought for policy support in Europe

<p>Drought events cause multiple impacts on the environment, the society and the economy. Here, we analyse recent major drought events with different metrics using a common framework. The analysis is based on current reanalysis (ERA5, ERA5-Land, MERRA-2) and merged remote-sensing products (ESA-CCI soil moisture, gridded satellite soil moisture from the Copernicus Climate Data Store), focusing on soil moisture (or agricultural) drought. The events are characterised by their severity, magnitude, duration and spatial extent, which are calculated from standardised daily anomalies of surface and root-zone soil moisture. We investigate the ability of the different products to represent the droughts and set the different events in context to each other. The considered products also offer opportunities for drought monitoring since they are available in near-real time.</p><p>All investigated products are able to represent the investigated drought events. Overall, ERA5 and ERA5-Land often show the strongest, and the remote-sensing products often weaker responses based on surface soil moisture. The weaker severities of the events in the remote-sensing products are both related to shorter event durations as well as less pronounced average negative standardised soil moisture anomalies, while the magnitudes (i.e., the minimum of the standardised anomalies over time) are comparable to the reanalysis products. Differing global distributions of long-term trends may explain some differences in the drought responses of the products. Also, the lower penetration depth of microwave remote sensing compared to the top layer of the involved land surface models could explain the partly weaker negative standardized soil moisture anomalies in the remote-sensing products during the investigated events. In the root zone (based on the reanalysis products), the drought events often show prolonged durations, but weaker magnitudes and smaller spatial extents.</p>

Download Full-text

Evaluation of Drought Stress in Cereal through Probabilistic Modelling of Soil Moisture Dynamics

Water ◽

10.3390/w12092592 ◽

2020 ◽

Vol 12 (9) ◽

pp. 2592 ◽

Cited By ~ 1

Author(s):

María del Pilar Jiménez-Donaire ◽

Juan Vicente Giráldez ◽

Tom Vanwalleghem

Keyword(s):

Soil Moisture ◽

Soil Depth ◽

Standardized Precipitation Index ◽

Early Warning Systems ◽

Agricultural Drought ◽

Annual Precipitation ◽

Drought Indices ◽

Climate Data ◽

Management Tools ◽

The Standardized Precipitation Index

The early and accurate detection of drought episodes is crucial for managing agricultural yield losses and planning adequate policy responses. This study aimed to evaluate the potential of two novel indices, static and dynamic plant water stress, for drought detection and yield prediction. The study was conducted in SW Spain (Córdoba province), covering a 13-year period (2001–2014). The calculation of static and dynamic drought indices was derived from previous ecohydrological work but using a probabilistic simulation of soil moisture content, based on a bucket-type soil water balance, and measured climate data. The results show that both indices satisfactorily detected drought periods occurring in 2005, 2006 and 2012. Both their frequency and length correlated well with annual precipitation, declining exponentially and increasing linearly, respectively. Static and dynamic drought stresses were shown to be highly sensitive to soil depth and annual precipitation, with a complex response, as stress can either increase or decrease as a function of soil depth, depending on the annual precipitation. Finally, the results show that both static and dynamic drought stresses outperform traditional indicators such as the Standardized Precipitation Index (SPI)-3 as predictors of crop yield, and the R2 values are around 0.70, compared to 0.40 for the latter. The results from this study highlight the potential of these new indicators for agricultural drought monitoring and management (e.g., as early warning systems, insurance schemes or water management tools).

Download Full-text

Agricultural drought analyses for temperate Australia using a biophysical pasture model. 2. Relationship between rainfall and the start and end of predicted droughts

Australian Journal of Agricultural Research ◽

10.1071/ar08076 ◽

2008 ◽

Vol 59 (11) ◽

pp. 1061 ◽

Cited By ~ 4

Author(s):

G. M. Lodge ◽

I. R. Johnson

Keyword(s):

New South ◽

Summer Rainfall ◽

Growing Season ◽

Agricultural Drought ◽

Wagga Wagga ◽

Climate Data ◽

Winter Rainfall ◽

Rainfall Occurrence ◽

South Wales ◽

The Relationship

This paper reports relationships between predicted soil water content (SWC) on the first day of the month (SWCFOM, mm of water) and previous monthly rainfall for 100 years of daily climate data (1905–2005) at four sites (Albany, Western Australia; Hamilton, Victoria; and Wagga Wagga and Barraba, New South Wales). Overall, predicted SWCFOM was correlated (P < 0.05) with rainfall in the previous one, two, or three months. However, the proportion of variation in SWCFOM that could be attributed to its regression on previous rainfall was variable and the relationship tended to improve when individual months were examined. At the three winter-rainfall sites (Albany, Hamilton, and Wagga Wagga), there was a reasonably good relationship between the start of a predicted drought and the end of the growing season and also between the end of a predicted drought and the occurrence of break-of-season. However, for the summer-rainfall dominant site at Barraba, rainfall occurrence was less seasonally defined and there was no clear relationship. While analysis of historical rainfall data for the months in which predicted agricultural droughts started or ended provided some useful insights, it was concluded that it would probably be more instructive to model SWC outcomes for a range of future rainfall scenarios and then examine their likelihood of occurrence using rainfall percentiles.

Download Full-text

Agricultural drought projection in Ethiopia from 1981 to 2050: Using coordinated regional climate downscaling experiment climate data for Africa

Extreme Hydrology and Climate Variability ◽

10.1016/b978-0-12-815998-9.00024-5 ◽

2019 ◽

pp. 311-323 ◽

Cited By ~ 1

Author(s):

D.T. Degefie ◽

Jemal Seid ◽

Berhan Gessesse ◽

Tulu Besha Bedada

Keyword(s):

Regional Climate ◽

Agricultural Drought ◽

Climate Data ◽

Climate Downscaling

Download Full-text

An Integrated Method for Identifying Present Status and Risk of Drought in Bangladesh

Remote Sensing ◽

10.3390/rs12172686 ◽

2020 ◽

Vol 12 (17) ◽

pp. 2686

Author(s):

Md. Sarker ◽

Nichol Janet ◽

Siti Mansor ◽

Baharin Ahmad ◽

Shamsuddin Shahid ◽

...

Keyword(s):

Remote Sensing ◽

Potential Risk ◽

Monsoon Season ◽

Meteorological Drought ◽

Agricultural Drought ◽

Anthropogenic Activity ◽

Drought Risk ◽

Field Observations ◽

Climate Data ◽

Drought Conditions

The occurrence and severity of agricultural droughts may not be dependent upon climatic variables alone. Rather increasingly, drought is affected by human interventions such as irrigation. Anthropogenic activity has introduced uncertainty in the assessment of current drought and future drought risk in many parts of the world; neither climatic nor remote sensing data alone are able to assess drought conditions effectively. In response, we present a simple approach to assess drought by combining a remote sensing-based drought index, the Temperature Vegetation Dryness Index (TVDI), climate data (i.e., rainfall and temperature), and field observations to evaluate recent drought conditions in northwestern Bangladesh (NWB). Applying this approach, we gained five insights: (i) the TVDI successfully indicated the drought conditions of NWB and agrees with field observations, (ii) the integrated use of TVDI and climate data (such as rainfall and temperature) provides the best understanding of the difference between meteorological drought and droughts resulting from surface moisture conditions, (iii) the TVDI results agree with rainfall data (r2 = 0.40 in March and r2 = 46 in April) in a part of the study area (NWB) where irrigation is not available, (iv) the TVDI can be used along with climate data to predict the potential risk of drought, and (v) while meteorological drought exists due to low rainfall and high temperature in this NWB in pre-monsoon season, because of widespread irrigation practices, meteorological drought is unable to trigger agricultural drought over most parts of the study area. The findings imply that there is a potential risk of drought in NWB, since any disruption of irrigation water supply could trigger a severe agricultural drought over the whole region. This is similar to what is currently observed over a small part of NWB.

Download Full-text