CAFE Project

Summary, Science Questions, and Progress

Summary

Cyclic, though irregular, ENSO events are large-scale ocean-atmosphere perturbations in the Pacific Ocean known to have associated climate effects in many regions around the world. These climatic teleconnections, in turn, affect the productivity of marine and agricultural ecosystems. For example, the drought of 1988, during a La-Niña year, is estimated to have cost the U.S. Midwest $39 billion in damages.  Recent advances in satellite remote sensing and ocean-atmosphere modeling offer the possibility of forecasting region-wide seasonal events related to the El Niño-Southern Oscillation (ENSO), providing potential to improve agricultural planning and fishery management in the regions affected. The overall goal of the CAFE project is to help realize that potential and to discover effective modes of such forecasts.

To achieve this goal the project has developed a coordinated methodological framework for linking observational and modeling tools used to predict ENSO, its climatic manifestations around the world, and its biophysical and socioeconomic impacts. This will include an in-depth analysis of the complete remote sensing record of the recent manifestations of ENSO events. The interdisciplinary team consists of oceanographers, atmospheric physicists, marine biologists and agro-ecologists.

The CAFE project is analyzing case studies to serve as prototypes for the characterization of El Niño-La Niña cycles henceforth. The primary fishery study region will be the tropical Pacific Ocean. Specific agricultural regions of interest are the U.S. Corn Belt and Southern Great Plains; the Nordeste region in Brazil; Southeast Brazil/Uruguay/Cerrados; Zimbabwe/South Africa; Mediterranean and Western Indonesia.

Science Questions

The current phase of the project addresses two main questions:

• How well do linked prediction systems reproduce observed agricultural and marine ecosystems responses with perfect predictions during the 1997-2000 ENSO cycle?
• How can current individual modeling systems, methodological approaches and their linkages be improved to develop optimized linked prediction systems?

The specific science questions for each of the project components include:

Linked Prediction System

• How well do perfect predictions of agriculturally important climate variables allow agricultural models to reproduce observed yield variations during the 1997-2000 ENSO cycle?
• How well do weather generators and mesoscale climate models using observed monthly mean anomalies in agricultural models reproduce observed agricultural production changes?
• How well do the GISS general circulation model (GCM) and regional model monthly climate anomalies and weather generators in agricultural models reproduce observed agricultural production changes, given:
• Observed sea surface temperatures (SSTs) everywhere?
• Observed SSTs in the Pacific Ocean and climatological SSTs elsewhere?
• Predicted SSTs in the Pacific Ocean and climatological SSTs elsewhere?

Remote Sensing

• How well do remote sensing indicators (NDVI from AVHRR, MODIS, ASTER data) characterize agricultural responses to the ENSO cycle?
• How do remote sensing data (TOPEX/Poseidon altimeter data, SSM/I, NSCAT, and QuickScat) help to improve ENSO model prediction skill?
• How do remote sensing data (MSU, AMSU) help to evaluate GCM model development in regard to improving GCM predictions of ENSO climate teleconnections over agricultural regions?

Climate Teleconnections

• How does GCM simulation of ENSO regional climate teleconnections improve with increased vertical and horizontal resolution?
• What elements contribute to uncertainty in GCM regional climate predictions, e.g., what is the nature and stability of the first and second moment statistics, the frequency distribution of signal and noise, the skill in characterizing seasonality, and the differences between the El Niño and La Niña events of the regional climate simulations?
• What are the nature of any systematic errors?
• How many GCM ensemble members are optimal for prediction of agricultural outcomes?
• Does averaging results from two different GCMs enhance the predictability of ENSO's impacts on agriculture?
• How do weather generators and mesoscale climate models compare in skill at downscaling GCM results for use in agricultural model simulations?

Marine Ecosystems/Fisheries

• What were the interactions between the forcing, the ocean circulation response, and the biology that led to the extreme swings in the tropical Pacific ecosystem during 1997-1998?
• How did the Catch Per Unit Effort (CPUE) of the tuna fishery in the tropical Pacific during 1997-1998 compare with previous El Niño events?
• Is there a significant relation between changes in CPUE, foraging, ocean color, primary and/or secondary production in the tropical Pacific?
• How did the Pacific tuna fisheries react to the 1997-1998 El Niño/La Niña? Were they aware of forecast information? If so, did it guide their planning? What climatic/oceanic indicators are deemed important?
• How well do the forecasted winds reproduce the surface chlorophyll response in the coupled ecosystem model and how well the does the surface chlorophyll from the model and from SeaWiFS correlate with each other and with the fisheries data?

Terrestrial Vegetation/Agriculture

• Were regional agricultural outcomes of the 1997-98 El Niño as expected, both in terms of productivity and economics, based on historical climate-agricultural responses?
• What is the comparative strength of agricultural productivity and economic responses to El Niño and La Niña events?
• What do crop model simulations predict as effective adaptation strategies and what adaptation strategies were actually taken in the study regions?
• How were ENSO predictions utilized in agricultural planning in the study regions; what were the consequences, and how could predictions have been better utilized, given improved understanding of the linked prediction system?

Progress

The project team is conducting a coordinated analysis of the 1997-2000 El Niño/La Niña, examining the predictions, oceanic manifestations, climate teleconnections, marine and terrestrial biological responses, and fishery and agricultural ecosystem outcomes.

• The ENSO prediction component is concentrating on the use of satellite data (TOPEX and QuickSCAT) to improve various stages of ENSO prediction technology: model initialization, process simulation, and data assimilation.
• We have used modeling studies to assess how well the GISS GCM can predict climate parameters of consequence for agricultural output. Five simulations using observed SSTs have so far been performed with a 4°x5° (lat × long) model with 12 vertical layers, for the decade of the 1990s; an additional simulation has been made for a 2°x2.5° model with 32 vertical layers for the same time period, as have four additional runs for the critical 1996-1999 period. A ten-year run has also been completed using a 4°x5° model with 53 vertical layers. For finer-scale resolutions, the GISS/CCSR 15-layer regional climate model (RCM) has been used to downscale NCEP reanalysis fields and model output over South America, beginning with Feb. 20, 1997 (Results are posted on an ftp site hosted by the Experimental Climate Prediction Center at Scripps, UCSD).
• The agricultural team has developed a set of steps to facilitate regional analysis of ENSO effects on agriculture. A key step is the assessment of the temporal and spatial patterns of ENSO effects through analysis of satellite data (AVHRR NDVI and EOS products). We are now focusing on the utilization of ENSO forecasts for commercial and subsistence agriculture in Uruguay and Nordeste, Brazil, 1996-2000.
• The fisheries component is assessing the utility of linked satellite data, ENSO predictions, GCMs, ocean models, and marine ecology models in ecosystem process studies.

CAFE Home * Summary * Team
Components: ENSO, Climate, Marine, Vegetation
Education * Instruments * Acquisitions * Publications