世界发展银行-Salt-of-the-Earth-_-Quantifying-the-Impact-of-Water-Salinity-on-Global-Agricultural-Productivity_27页_1mb

Summary of "Salt of the Earth: Quantifying the Impact of Water Salinity on Global Agricultural Productivity"

Core Content

This paper, Salt of the Earth, quantifies the global impact of water salinity on agricultural productivity by analyzing the effects of increased electrical conductivity (EC) in surface and irrigation water. The study focuses on the relationship between water salinity and crop yields, using a combination of satellite data, hydrological models, and water quality monitoring data from multiple regions.

Main Viewpoints

• Water Salinity Increase: Salinity in surface waters is rising globally due to factors such as over-extraction of water, poor irrigation management, and sea-level rise.
• Impact on Agricultural Productivity: The study finds that even relatively low levels of salinity can significantly reduce agricultural productivity, with a nearly linear relationship between salinity levels and yield losses.
• Global Food Losses: It is estimated that salinity reduces global agricultural production by 124 trillion kilocalories per year, which is enough to feed over 170 million people annually.
• Regional Variability: The impact of salinity is not uniform across regions. High salinity-induced yield losses are observed in all continents, but particularly in areas with existing malnutrition challenges.
• Scientific Basis: The study references a large body of scientific literature that documents the sensitivity of specific crops to salinity, including vegetables and staple crops like rice.

Key Information

Data Sources

• Water Quality Data:

  • Mekong River Basin Commission (MRC): 121 monitoring stations across four countries (Cambodia, Lao PDR, Thailand, and Vietnam).
  • Central Water Commission (CWC): 425 monitoring stations in India, covering major river basins.
  • GEMStat: A global dataset from the United Nations Global Environmental Monitoring System (UNGEMS), covering 71 countries and 1,124 monitoring stations.

• Agricultural Productivity Data:

  • Net Primary Productivity (NPP) is used as a proxy for agricultural productivity, derived from satellite data (MODIS).
  • NPP is measured for each 0.1-degree gridcell, with a minimum cropland threshold of 30% for inclusion in the analysis.

• Irrigation Data:

  • Irrigation data is sourced from the Food and Agriculture Organization’s Global Map of Irrigated Areas (GMIA) version 5.0.6.
  • Gridcells are restricted to those where irrigation is known to occur, ensuring that the water quality data reflects the conditions of the irrigation water.

• Weather Data:

  • Monthly precipitation and temperature data is obtained from Matsuura and Willmott (2001) at a 0.5-degree gridcell level.

Empirical Strategy

• A regression model is used to estimate the impact of salinity on crop productivity, controlling for climate and geographic factors.

• The model is:
  $$
  \Delta \log \left(N P P _ {i t}\right) = \alpha + \lambda * g \left(W Q _ {i t}\right) + \delta * f \left(\text {c l i m a t e} _ {i t}\right) + \sigma_ {i} + \rho_ {\mathrm {y}} + \theta_ {c} * Y + \varepsilon_ {i t}
  $$
  where $WQ_{it}$ is water quality (measured as EC), and $climate_{it}$ includes temperature and rainfall.

• Hydrological Connectivity:

  • In regional analyses (Mekong and India), gridcells are matched to upstream monitoring stations to ensure the water quality data accurately reflects irrigation water conditions.
  • For the global analysis, a simpler method is used, matching gridcells to the nearest station at a higher elevation to simulate water flow direction.

• Robustness Checks:

  • The analysis is tested using different cropland thresholds (30%, 75%, 90%) and different EC ranges.
  • The results show a stable yield response function across various geographies.

Main Results

• Mekong River Basin:

  • When EC exceeds 100 mS/m, agricultural productivity declines by 5.7–8.2%.
  • Approximately 5% of observations exceed this threshold.

• India:

  • Similar results are observed, with a 5.5–6.6% decline in productivity when EC exceeds 100 mS/m.
  • About 5% of observations in India exceed this threshold.

• Global Analysis:

  • When EC exceeds 100 mS/m, global agricultural yields decline by 11.0–13.5%.
  • The study estimates that 124 trillion kilocalories are lost annually due to salinity, equivalent to the food budget of 170 million people.

Additional Findings

• Regional Sensitivity:

  • Vegetables are generally more sensitive to salinity than crops like rice.
  • Vegetables have a soil salinity tolerance threshold of below 250 mS/m, while irrigation water tolerances are 30–50% lower.

• Crops Affected:

  • For example, sweet peppers become sensitive to salinity at 143 mS/m, with a yield decline slope of 11%.
  • Chickpeas, lentils, and faba beans experience a 50% yield decline when EC in irrigation water exceeds 420, 440, and 520 mS/m, respectively.

• Implications:

  • The study highlights the underappreciated threat of water salinity to global food security.
  • It suggests that salinity-sensitive agricultural systems are particularly vulnerable, especially in regions where water scarcity and malnutrition are already present.

Conclusion

The paper concludes that water salinity is a significant and underappreciated threat to global agricultural productivity. It provides the first global estimate of the impact of salinity on crop yields and highlights the need for more attention to water quality issues in agricultural policy and planning. The findings emphasize the importance of addressing salinity in irrigation systems and managing water resources more sustainably to prevent further losses in food production.