TL;DR
This study analyzes historical precipitation data for Xinyang, China, using Standardized Precipitation Indices (SPIs) and wavelet analysis to identify drought and flood cycles. The findings reveal distinct cyclical patterns at various time scales, suggesting a higher frequency of floods than droughts and predicting a potential shift towards drought conditions after 2021, necessitating proactive drought prevention measures.
What Problem Does This Paper Address?
Accurate identification and prediction of drought and flood events are crucial for effective local management, especially in regions like Xinyang, China, which are susceptible to these extreme weather phenomena. These events significantly impact agriculture, socio-economics, and ecological systems. While various drought indices exist, understanding the cyclical nature and temporal evolution of these events across different time scales is essential for developing robust mitigation strategies. This paper aims to fill this gap by providing a detailed analysis of drought and flood cycles in Xinyang using advanced statistical and time-series analysis techniques.
Methods
The study employs a multi-pronged approach:
- Standardized Precipitation Index (SPI): Precipitation data is transformed into SPI values across three different time scales: 3-month (SPI3), 6-month (SPI6), and 9-month (SPI9). SPI is a widely used index that quantifies drought and flood conditions based on precipitation deficits or excesses relative to a long-term average. The paper details the calculation of SPI, which involves fitting precipitation data to a probability distribution and then normalizing it.
- Morlet Wavelet Analysis: This technique is used to analyze the time-series data of the SPI values to identify periodicities and cyclical patterns. Wavelet analysis is particularly adept at capturing non-stationary signals and revealing how the frequency content of a signal changes over time. The Morlet wavelet is chosen for its suitability in analyzing time-frequency domain characteristics. The study generates wavelet real part, modal, and modulus squared contour maps to visualize these variations.
- Mann-Kendall (M-K) Test: This non-parametric statistical test is applied to detect trends and abrupt changes in the precipitation time series. It is robust and does not assume a specific distribution for the data, making it suitable for hydrological time series analysis. The M-K test helps identify significant shifts in precipitation patterns over time.
Datasets
The study utilizes monthly precipitation data from the Xinyang meteorological station, covering a 45-year period from 1961 to 2015. This data was obtained from the China Meteorological Administration (CMA). No new benchmark datasets were introduced.
Key Insights and Ideas
- Multi-Scale Cyclical Nature: Drought and flood events in Xinyang are not random but exhibit distinct cyclical patterns that vary significantly depending on the SPI time scale. This highlights the importance of considering different temporal perspectives when analyzing hydrological extremes.
- Dominant Cycles Identified: The wavelet variance analysis revealed key cyclical periods for drought and flood events. The 7-year cycle appears to be dominant, followed by 4-year and 13-year cycles, with an 18-year cycle being less dominant. This provides a quantitative understanding of the underlying periodicities driving these events.
- Flood Dominance and Future Shift: The analysis suggests that floods have historically been more frequent than droughts in Xinyang. However, the study predicts a potential shift towards more drought-prone conditions after 2021. This insight is critical for resource managers to re-prioritize their preparedness strategies.
- Abrupt Change Detection: The M-K test identified a significant abrupt change in precipitation patterns starting around 1996, with a subsequent trend towards decrease, and a more pronounced increase in significance from 2005 to 2015. This indicates a notable shift in the region's climatic behavior.
Key Results
- SPI Analysis: The SPI3, SPI6, and SPI9 time series revealed sharp fluctuations in drought indices during specific periods (e.g., 1962-1967, 1979-1984, 1999-2003), often transitioning from flooding to drought conditions.
- Wavelet Analysis:
- SPI3 showed cyclical variations at 14–18 years, 8–13 years, and 3–7 years, with the 18–27-year scale having the strongest energy but weakening after 1998.
- SPI6 exhibited less obvious temporal changes, with a similar pattern to SPI3 but with a weakening periodicity from 1998–2015.
- SPI9 identified significant periodicities at 14–18 years and 8–13 years, with the latter showing more stable periodicity.
- Wavelet Variance: Peaks in wavelet variance indicated dominant cycles of approximately 4a, 7a, 13a, and 18a across the SPI scales, with the 7a cycle being the most dominant.
- M-K Test: The analysis indicated a trend of decreasing precipitation from the 1970s to the 21st century, with an abrupt change detected around 1996.
Limitations and Future Work
- Data Availability: The study acknowledges that the availability of the latest meteorological data was limited, preventing the analysis from extending to the most recent years.
- Future Research Directions: The authors suggest that future research should leverage remote sensing data to acquire up-to-date monitoring information and to accurately estimate drought conditions. Combining classical methods like SPI with new technologies like remote sensing is proposed as a future research direction for regional drought monitoring.
Summary updated