David bio photo

David
García-León

Marie Skłodowska-Curie Research Fellow (Ca' Foscari University and Euro-Mediterranean Center on Climate Change, Venice).


Curriculum Vitae
Email Google Scholar Research Gate Orcid Github Twitter RSS

with Sergio Contreras and Johannes Hunink (FutureWater). Submitted to Journal of Agronomy and Crop Science

Europe in general, and Spain in particular, have experienced drought episodes increasingly over the past decades. These ever more frequent events surely pose threats to food supply security and compromise the stability of the domestic agri-food market. In order to insure agricultural production against those extreme episodes, accurate and efficient agricultural policies should be proposed and measures to adapt production to changing meteorological conditions must be taken. But prior to that, it is required to carefully assess and quantify the effect of water scarcity scenarios on the productivity of our crops, i.e., we should have clear the drought sensitivity of crops. This remains an open empirical question (some positive and some negative effects). Recent studies detect a clear relationship between crop yields and some drought indices Gunst et al. (2015).

The use of statistical methods to approach the relationship between yields and meteorological variables has proliferated during the last decade. This is attributable to the increasing availability and improved quality of observed data, the development of computer tools that can handle big data sets, and the adoption of new observation techniques, like remote sensing. Schlenker et al. (2009 is a prominent example of this stream of literature. Some authors, like Lobell (2010), have even claimed that this approach would beat the explanatory and forecasting performance of traditional crop models.

Regarding the impact of droughts in the real economy, a few theoretical exercises linking natural disasters and economic growth have been developed. Other studies prove a clear link between drought episodes and the occurrence of civil conflicts and disease. However, empirical studies addressing specifically the role of droughts in the evolution of crop yields are scarce in the literature, partly because the development of drought indicators has been relatively recent. One of the few examples is the work by Lobell et al. (2014), applied to the corn belt in the U.S. Midwest. They identify that, despite crop yields have generally increased over the studied period thanks to agronomic changes in plants’ drought tolerance, the sensitivity to droughts of some crop varieties, like maize, is greater now than it was at the beginning of the sample. A similar response of Spanish crops could be expected but many heterogeneities could arise due to different geographic, agronomic, or productive motives. Our study will shed light onto the reactivity of Spanish fields to drought events.

Standardised Precipitation Index (SPI). December, 2015. Source: AEMET.

Also very recently, Stagge et al. (2015), using qualitative user-provided drought impact indicators spanning agriculture, energy and industry, public water supply, and freshwater ecosystem across five European countries. They build logistic regression models and demonstrate the feasibility of using such models to predict drought impact likelihood based on meteorological drought indices. In our case, we will focus specifically on the non-irrigated agricultural sector and we will isolate ourselves from any kind of qualitative indicator, adopting instead, crop yields as the main source of detection of drought impacts.


An obstacle to measuring progress in farmers’ fields has been the lack of accurate field-level data on both environmental conditions and yield performance that span a range of drought conditions and time. We will overcome this obstacle by using the Encuesta sobre Superficies y Rendimientos de los Cultivos (ESYRCE). The ESYRCE survey serves as input to the European Commision’s FADN. ESYRCE is an annual survey conducted by the Spanish Ministry of Agriculture in which field-level data on surface and yields is collected and interpolated to construct a country-gridded data set spanning years 1990 to date. Great spatial and crop disaggregation is key to unveil plausible evidence because some areas are more prone to suffering from drought events due to their geographic and orographic characteristics. At the same time, some crop varieties will bear more efficiently with abrupt changes in average climate patterns. Hence, deploying our study at the finest disaggregation level available is crucial to obtain meaningful results. Spatial heterogeneous responses of economic outcomes in response to weather variables have been previously documented in Ceglar et al. (2016) or García-León (2014) for European regions.

We will measure the degree of sensitivity of certain agronomic variables (yield) to traditional weather variables and drought-specific indicators, as those delivered by the GEISEQ system, (see the Infosequia portal).

According to the IPCC (2014) and other organisms, like the UNCCD, there is a need to integrate both scientific and local knowledge to allow for a better understanding of adaptive capacities and a better anticipation of complex interactions between biophysical and social systems in specific settings. Applied to our case study, another way of capturing changes in the evolution of crop yields in response to droughts rests on the difference between the impacts of climate change projected using the short-run (limited adaptation) and long-run (substantial adaptation) response curves. This difference can be interpreted as the private adaptation potential.


Analogously to the case of drought impacts, there are scarce empirical estimates of private adaptation of farmers to changing climate. Lobell et al. (2014, 2015) for Europe and Huang (2015) for the US are some of the few examples available. The former authors estimate adaptation by jointly estimating both short-run and long-run response functions using time-series and cross-sectional variation in subnational yield and profit data that we will adapt to our example at hand. They also develop statistical tests for Climate Change impacts and adaptation that could be easily accommodated to study the adaptation behaviour in response to droughts.


A non-parametric approach to describe spatio-temporal yield variation and drought impacts at the farm level: An application to the Po river basin (In progress)

with Raúl López, Andrea Toreti and Mateo Zampieri (Joint Research Centre)

The purpose of this research project is to advance in describing the role of remote sensing drought indicators as explanatory factors of rainfed crop yields using empirical crop models. From this research phase, an early warning drought classification system could be extracted.

Data

Weather and Remote Sensing data

  • Meteorological indicators (temperature, precipitation) from weather stations: Local network of stations in the Po Valley, as the network is dense enough and available.

  • fAPAR from Copernicus. Indicator on crop photosynthetic activity.

  • Actual evapotranspiration from MODIS. This product provides global ET at 1km spatial resolution in 8-days composites. The estimation of ET is based on the energy balance approach.

  • Superficial soil moisture from radar imagery. It is available since 1979 up to present and the SSM is estimated through a combination of passive and active radar sensors. The spatial resolution is 0.25 degrees (approx. 25 km).

  • Land Surface Temperature (LST), from thermal imagery. This product is available since 1981.

Yield data. RICA dataset on yield and other agricultural variables over the Po Valley.

Methods

We will adapt the methodology of García-León et al. (2017) to this case study. An empirical crop model is estimated using meteorological and biophysical remote sensing observations

Yield anomalies will be defined as

Sensitivity of yields could be explored by looking at the Drought Environment Index (DEI)

Following Zampieri et al (2017), construct a tailor-made Combined Stress Indicator (CSI) as yield explanatory factor, using non-parametric techniques: LOESS detrending and ridge linear equations.

Given the discontinuity of farms surveillance, it is proposed to work on a longitudinal framework. Working at the strata level (economic dimension of farms) seems also interesting.