Abstract
Under the background of global warming, the loss caused by extreme precipitation is becoming more and more serious. As a bridge between risk and disaster losses, research on vulnerability curves has become a focus in the field of disaster risk. In this study, based on meteorological data, crop physiological data, and field management data, typical waterlogging disaster years were selected as cases. The growth process of maize in the Midwest of Jilin province was simulated using the localized and spatially scaled CERES-Maize model. The disaster intensity index was calculated using a Copula function to construct the waterlogging disaster vulnerability curve. The results show that maize vulnerability to waterlogging, in descending order, occurs during: emergence-jointing, jointing-heading, heading-milk, and milky-maturity stages. These results provide an important basis for mitigating waterlogging disasters, choosing irrigation/drainage periods, and optimizing maize planting arrangements.
Introduction
As global warming intensifies, extreme precipitation events are becoming more frequent, making waterlogging a primary threat to food security. In the Midwest of Jilin Province—a major global maize producer—waterlogging is the second most damaging meteorological disaster after drought. While researchers often study disaster risk, there is a gap in understanding the vulnerability curve: the functional relationship between disaster intensity and actual loss. Traditional assessments often rely on sparse historical yield data. This paper addresses the need for a granular, stage-specific understanding of maize's physiological reaction to excess water throughout its life cycle.
Methodology and Data
The study integrated high-quality datasets including daily meteorological records (1960–2014) from 14 stations, 1:1,000,000 scale land use maps, soil property data from the "Jilin Soil" database, and field observations for the "Zhengdan 958" maize variety.
The authors employed a multi-step modeling approach:
- Crop Simulation: The CERES-Maize model was localized for Jilin and scaled to a 5km x 5km grid, simulating growth under optimal ($S_1$) and waterlogged ($S_2$) conditions.
- Disaster Intensity Index: A Copula function combined waterlogging intensity (SPI) and duration into a single joint index.
- Vulnerability Curve Construction: Vulnerability ($S$) was calculated as the relative loss of physiological indicators (LAI for early stages, grain weight for later stages) between $S_1$ and $S_2$.
The model was validated against actual yield data from 2008–2014, achieving an $R^2$ of 0.5305, which is statistically significant for spatial scaling in agricultural modeling.
Key Results
The study quantified maize vulnerability across four growth stages. The "Emergence-Jointing" stage was identified as the most critical period, where waterlogging can lead to losses exceeding 60% due to stunted plant height and leaf development.
The vulnerability hierarchy is as follows:
- Emergence – Jointing: Highest vulnerability (Loss up to >60%).
- Jointing – Heading: High vulnerability; nutrient competition with weeds increases during wet periods.
- Heading – Milk Stage: Moderate vulnerability (Grain weight loss up to 25%).
- Milky – Maturity: Lowest vulnerability; dry matter accumulation is nearly complete.
Discussion and Conclusion
The research highlights that final yield alone is an insufficient metric for vulnerability. By utilizing Leaf Area Index (LAI) for early stages, the model captures physiological damage that precedes yield loss. The use of Copula functions provides a nuanced view of disaster intensity compared to simple rainfall totals.
While the study focused on the "Zhengdan 958" variety, future work should incorporate more diverse hybrids and complex soil-water-plant interactions to improve model precision across the entire province. These findings offer a quantitative framework for optimizing disaster mitigation in China’s "Golden Maize Belt."
Summary updated