Climate and Agro-Ecological Variability

The contribution of Theme 2 is a better understanding of regional climate variability and extremes and
the consequences for the agro-ecosystems and environmental services that support the rural populations
in the studied watersheds. This work will advance climate science in the social context of this project
and inform the evaluation of current and future vulnerability and risk.

Modeling climate variability and extremes (Theme 2A) Sauchyn and Villalba:

Various climate model projections (e.g., Kharin and Zweirs, 2005; Tebaldi, et al., 2006; IPCC, 2007) suggest increased interannual hydroclimatic variability and frequency of heavy precipitation events with 21st century greenhouse warming. To understand the potential for shifts in regional hydroclimatic variability and the severity and frequency of extreme events, we will initially establish the nature and forcing of the internal (natural) variability that underlies the trends imposed by a warming atmosphere and oceans. We will analyze all available climate and surface water records in each study area, using appropriate parametric and non-parametric methods for trend detection and determining the probabilities of exceeding specific precipitation and surface water levels. These methods will include the pre-whitening of time series to prevent the false detection of trends in autocorrelated data, tests for shifts in the mean and variance of the probability distribution functions of hydroclimatic data, analysis of the spatial coherence of the identified trends, and the explicit segregation, detection and attribution of trend versus low-frequency variability. We will place particular emphasis on the analysis of hydrometric records (e.g., Masiokas et al., 2010; St. Jacques et al., 2010) given their availability in all regions, the tendency for streamflow to integrate watershed hydrology and climate, the sensitivity of runoff to climate fluctuations, and the importance of surface water supplies in the studied watersheds.

Environmental Services (Theme 2B) Piwowar and Poveda:

The climate sensitivity of the ecosystems and water resources in the study regions is a major determinant of the vulnerability of rural communities and economies given their dependence on the abundance and quality of these environmental services (Santibañez and Santibañez, 2008; Sauchyn and Kulshrestha, 2008). Recent trends in vegetation productivity and the extent of snow and glacier ice cover is a strong signal of climate change and variability. Fluctuations in natural capital, and the link to climate variability, can be measured for the past several decades using instrumental records from monitoring networks and various archives of digital satellite data. Surrogate indicators of environmental services, such as "normals" of vegetation vigour (e.g., Piwowar, 2010), extent of glaciers and snowpack (e.g., Poveda and Pineda, 2009), surface water storage in open (e.g., Tang et al., 2010) and frozen (e.g., Turchenek et al., 2006) states, and deforestation (e.g., Broich et al., 2009) or afforestation (e.g., Le Hegarat-Mascle et al., 2005) will be derived through hypertemporal analyses (McCloy and Lucht, 2003) of the 20-30 year archives of NOAA AVHRR, Landsat MSS/TM/ETM, Nimbus-7 SMMR, DMSP SSM/I, and Aqua AMSR-E satellite imagery. Statistically significant anomalies will indicate the spatial and temporal domains of environmental responses to climate forcing

Modeling Climate Impacts on Agro-Ecosystems (Theme 2C) Santibañez and Kulshrestha:

Theme 2C will examine impacts of climate change and variability on the capacity of the watersheds and ecosystems to sustain agricultural productivity and critical environmental services. Decision makers generally require that researchers translate projected climate change changes into potential impacts (e.g., crop yield and seasonality, irrigation demand) in order to prepare appropriate adaptation strategies at local and regional scales. Most adverse climate impacts to agricultural systems are due to the exposure of crops, water and soil resources to extreme conditions rather than shifts or trends in the means. To evaluate the possible responses of grasslands and crop species to climatic variations, we will apply a dynamic simulation model (SIMPROC) developed by Santibañez (1986; 1998). This model integrates the main ecophysiological relations that operate between the seedling and harvest stages of several cultivated species. Growth simulation is based on levels of solar radiation, leaf interception, temperature and relative satisfaction of evapotranspiration needs. The crop develops from one phenological stage to the next based on accumulation of thermal or development units. There is a subroutine to simulate the water balance of the soil-plant system. To assess variability of grassland production, the SIMPROC-grassland model can simulate the annual curve of dry matter accumulation, on a weekly basis, as a response to climate. The resulting 30 dry matter curves are integrated in a sub-routine that generates the weekly probability of dry matter availability. The SIMPROC model has been validated for agricultural regions of Chile, calibrated in several Latin American countries, and used to simulate crop and grassland productivity under present and future climatic conditions supposing temperature and precipitation variation from GCMs.