Abstract
Owing to the complexity and variability of global climate, the study of extreme events to ensure food security is particularly critical. The standardized precipitation requirement index (SPRI) and chilling injury index (ICi) were introduced using data from agrometeorological stations on the Songliao Plain between 1981 and 2020 to identify the spatial and temporal variability of drought, waterlogging, and low-temperature cold damage during various maize growth periods. Compound drought and low-temperature cold damage events (CDLEs) and compound waterlogging and low-temperature cold damage events (CWLEs) were then identified. To measure the intensity of compound events, the compound drought and low-temperature cold damage magnitude index (CDLMI), and compound waterlogging and low-temperature cold damage magnitude index (CWLMI) were constructed by fitting marginal distributions. Finally, the effects of extreme events of various intensities on maize output were examined. The findings demonstrate that: (1) There were significant differences in the temporal trends of the SPRI and ICi during different maize growth periods. Drought predominated in the middle growth period (MP), waterlogging predominated in the early growth period (EP) and late growth period (LP), and both drought and waterlogging tended to increase in intensity and frequency. The frequency of low-temperature cold damage showed a decreasing trend in all periods. (2) The CDLMI and CWLMI can effectively determine the intensity of CDLEs and CWLEs in the study area; these CDLEs and CWLEs had higher intensity and frequency in the late growth period. (3) Compared to single events, maize relative meteorological yield had a more significant negative correlation with the CDLMI and CWLMI.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
With the intensification of global climate change, the frequency and intensity of extreme weather events caused by changes in hydrothermal resources are increasing (Dhaliwal and Williams 2022). These events, including drought, waterlogging, and high temperature and low-temperature cold damage, have become major abiotic stressors that severely limit global crop growth and productivity (Lobell et al. 2011). If population growth and climate change are considered, by 2050, the total global food demand will increase by 30–62%, putting additional pressure on food security. To meet the increased food demand caused by population growth and reduce the negative effects of increasing climate extremes due to global warming, it is crucial to understand the connection between crop yield and climate variables (Shi et al. 2020; Sun et al. 2023).
According to historical data, droughts are becoming more intense and frequent, which will eventually cause future maize yields of China to decline by approximately 36–39% (Ling et al. 2022; Yang et al. 2022). One of the two main reasons for crop-yield losses in all US agricultural regions is excessive precipitation-induced waterlogging, and late-season precipitation has a particularly negative impact on local peanut and sweet potato production (Eck et al. 2020). Significant changes in atmospheric circulation triggered by global warming have exacerbated the instability of the climate system, and extreme low-temperature cold damages occur (Sun et al. 2023). An example of such an event occurred in late April 2020 in northwestern China and damaged 530,100 ha of crops, of which 154,100 ha suffered crop failure, resulting in a direct economic loss of USD 1.2 billion (Kim et al. 2023). Owing to the complexity and uncertainty of climate change, compound events may have worse effects on agricultural production than individual events, because the magnitude, duration, and frequency of future extreme weather events increase concurrently (Feng et al. 2019; Zscheischler et al. 2020).
In recent years, considerable attention has been paid to the effects of compound events on crop yield, and such research can be divided into three major groups. The first approach uses probabilistic statistical assessment. For example, a meta-Gaussian model based on a multivariate distribution was employed to evaluate the possibility of crop production decline under hot and dry circumstances (Feng et al. 2021). The second approach employs a regression model to establish a function between yield and an index of multiple extreme events. Multiple linear regression models also have been used to analyze the effects of concurrent drought and heat stress on northern winter wheat yields (Chen et al. 2016). Recent research has built compound indices to directly characterize the size of compound events (Wu et al. 2019). Wang et al. (2021) created and graded a compound drought and heat wave magnitude index (CDHMI) to assess the destructive potential of various degrees of compound events. The CDHMI offers an alternative and integrated method based on existing extreme event evaluation indicators, incorporating the duration and intensity of each individual extreme event into a compound event, resulting in a better description of the compound extreme events.
The necessity of perceiving extremes from a compound perspective has been highlighted by the recent attention to compound weather and climate extremes. The occurrence of compound dry heat events worldwide has been widely discussed, but most previous studies have disregarded compound events associated with low-temperature cold damage (De Luca and Donat 2023). With climate warming, the northern boundary of spring maize planting in northeastern China has shown a clear northward shift and eastward expansion. Despite the warming trend, the magnitude of the temperature change in the northeast has not decreased, and regional low-temperature cold damage still occurs (Wang et al. 2015; Zhao et al. 2009). Maize grown across different temperature zones is susceptible to low-temperature cold damage, which results in severe yield loss. Therefore, it is of practical significance to study the impacts of low-temperature cold damage and related compound events on agricultural production.
Based on the daily precipitation and temperature dataset from 1981 to 2020, this study aimed to: (1) characterize the spatial and temporal distribution of drought, waterlogging, and low-temperature cold damage during different maize growth periods on the Songliao Plain over the last 40 years; (2) construct compound event indicators to identify the occurrence patterns of compound drought and low-temperature cold damage events (CDLEs) and compound waterlogging and low-temperature cold damage events (CWLEs) in the study area; and (3) explain the effects of drought, waterlogging, and low-temperature and compound extreme events on maize yield.
2 Study Area and Data Sources
This section summarizes the basic geographic profile of the Songliao Plain corn belt and describes the data sources, including meteorological and maize production data.
2.1 Study Area
This study focused on the Songliao Plain corn belt (40° 25′–48° 40′ N, 118° 40′–128° 00′ E), a significant commercial grain production area in China, which encompasses the majority of Liaoning Province, the central and western parts of Jilin Province, and the southern part of Heilongjiang Province (Fig. 1). Water supply for agriculture is limited in areas where rain-fed agriculture predominates (Dang et al. 2022). This region is susceptible to drought, waterlogging, and low-temperature cold damage because of the uneven geographical and temporal distribution of rainfall caused by the monsoon climate, as well as the uneven geographical and temporal distribution of heat resources caused by terrain differences and cold vortex activity. These extreme events have a direct impact on grain output.
2.2 Data Sources
The meteorological data were obtained from the China Meteorological Data Network.Footnote 1 Twenty-three stations with complete daily meteorological data from 1981 to 2020 in the Songliao Plain corn belt were selected as the study sites, and the daily meteorological factor datasets of each meteorological station included precipitation, average air temperature, maximum air temperature, minimum air temperature, average humidity, sunshine hours, and wind speed. Standard geographic information data for the study area were downloaded from the National Catalogue Service for Geographic Information.Footnote 2 Maize harvested area (ha) data for the study area in 2020 were obtained from Liu et al. (2022). The average annual maize yield per unit area and growth data for 23 administrative divisions in the Songliao Plain region from 1981 to 2020 were collected, with production data from provincial statistical yearbooks and various meteorological bureaus and growth and development data from the National Meteorological Information Centre.Footnote 3 According to the northeast maize phenological characteristics (Li et al. 2020), maize growth periods in the study area are classified and shown in Table 1.
3 Methods
This section describes the specific research methods used in this study.
3.1 Standardized Precipitation Requirement Index
Many indices have been proposed to characterize drought and waterlogging conditions, of which the standardized precipitation evapotranspiration index (SPEI) is widely used. In this study, the crop coefficients (KC) of the northeast region were incorporated into the calculation of the SPEI, considering that maize requires different amounts of water during different growth periods, to create a standardized precipitation requirement index (SPRI), which characterizes drought and flooding in maize planting areas (Wei et al. 2021). The calculation method is as follows:
where i represents the different growth periods; \({ET}_{C}\) is the crop water requirement (mm); KC is the crop coefficient for maize and reflects the combined effects of soil evaporation and crop transpiration (Table 2); and \({ET}_{0}\) is the daily reference evapotranspiration (mm), which reflects the effect of atmospheric evapotranspiration on crop water requirements in different periods and regions. ET0 was calculated using the Penman-Monteith method recommended by the Food and Agriculture Organization of the United Nations (FAO), which is based on physical processes, combines aerodynamic and energetic methods, and has been widely used to calculate the potential evapotranspiration of reference crops in agricultural fields (Chen and Sun 2015; Zhou et al. 2022).
where Δ is the slope of the saturated water pressure-temperature curve, KPa/°C; Rn is the net surface radiation, MJ/(m2 d); G is the soil heat flux, MJ/(m2 d), which is negligible when calculating ET0 on a daily or 10 d scale; T is the mean daily air temperature, °C; U2 is the wind speed at 2 m, m/s; es is the saturated water pressure, KPa; ea is the actual water pressure, KPa; and γ is the wet and dry gauge constant, KPa/°C.
The difference between monthly precipitation and crop water requirement was calculated as follows:
where \({P}_{i}\) is the precipitation at the i month (mm).
Moisture surplus or deficit cumulative sequences at different time scales was established as follows:
where k is the time scale (month); n is the number of calculations.
The difference between precipitation and maize water requirements was normalized, and the log-logistic probability distribution function of the three parameters was used to standardize the probability density and calculate the corresponding SPRI:
where α, β, γ are scale, shape, and position parameters, respectively; \({c}_{0}=2.515517,\, {c}_{1}=0.802853,\,{c}_{2}=0.010328,\,{d}_{1}=1.432788,\,{d}_{2}=0.189269,\,and \, {d}_{3}=0.001308.\)
The specific classification of drought and waterlogging are listed in Table 3.
3.2 Chilling Injury Index
The heat index F(T) considers the suitable, lower, and upper limit temperatures at different growth periods and can effectively characterize the effect of environmental heat conditions on crops. Influenced by many environmental factors, such as latitude, topography, and terrain in the northeast, the heat index distance level percentage was used to construct the chilling injury index (ICi) to characterize the occurrence of low-temperature cold damage at different growth periods (Guo et al. 2009; Cai et al. 2013). The calculation equations are as follows:
where T is the average daily temperature (°C); T0, T1, and T2 are the suitable, lower, and upper limit temperatures of maize in each growth period, respectively (Table 4); \({I}_{Ci}\) is the chilling injury index; \({F\left(t\right)}_{i}\) is the heat index at the i growth period; and \(\overline{F(t)}\) is the average heat index for the station.
Low-temperature cold damage is considered to occur when the ICi is less than − 2%. Cluster analysis was used to classify the calculated ICi (Table 5), and we compared the years of low-temperature cold damage and crop yields recorded in the historical data of each site, and adjusted and validated them to ensure the soundness of the low-temperature cold damage indicators.
3.3 Compound Events Magnitude Index
Adopting the type of compound events proposed by Zscheischler et al. (2020), we focused only on multivariate compound events, which refer to the phenomenon of multiple hazards occurring simultaneously in the same area causing extreme impacts. Two types of compound events were considered: compound drought and low-temperature cold damage events (CDLEs, SPRI < − 0.5 and ICi < − 2%), and compound waterlogging and low-temperature cold damage events (CWLEs, SPRI > 0.5 and ICi < − 2%). To quantify the intensity of compound events, we constructed a compound event amplitude index based on SPRI and ICi, which can eliminate the unit and amplitude differences between different events.
The compound drought and low-temperature cold damage magnitude index (CDLMI) and compound waterlogging and low-temperature cold damage magnitude index (CWLMI) were expressed as:
where ΔD denotes the degree of drought, defined as the absolute value of the difference between the drought indicator SPRI value (\({D}_{r}\)) and the threshold (\({D}_{th}\), − 0.5); ΔW denotes the degree of waterlogging, defined as the difference between the waterlogging indicator SPRI value (\({W}_{r}\)) and the threshold (\({W}_{th}\), 0.5); and ΔL denotes the degree of low-temperature cold damage, defined as the absolute value of the difference between the low-temperature cold damage indicator ICi value (\({L}_{r}\)) and the threshold (\({L}_{th}\), − 2%). \({P}_{\Delta D}\), \({P}_{\Delta W}\), and \({P}_{\Delta L}\) are non-exceeding probabilities obtained by fitting marginal distributions of ΔD, ΔW, and ΔC that fall within [0, 1], with large ΔD, ΔW, and ΔL corresponding to high probabilities. Figure 2 shows a conceptual diagram of the compound index construction. Cluster analysis (k-means clustering) was used to group the compound indices at different growth periods into three categories: light, moderate, and severe (Table 6).
3.4 Mann-Kendall Test
The Mann-Kendall (MK) test is a nonparametric test recommended by the World Meteorological Organization and has been widely used. It does not need to follow a certain distribution or be affected by outliers. It can be used for mutation detection and testing of trends in data series. The sequence (UF) and inverse (UB) statistic of the time series data can indicate trend—when UF > 0, the series shows an upward trend and vice versa. When the critical value (0.05 significance) is exceeded, it indicates a significant upward or downward trend of the sequence. When the UF and UB curves intersect with the critical value, the time corresponding to the intersection point is the time of the beginning of the mutation. Specific calculations can be seen in Zhao et al. (2023).
3.5 Frequency of Extreme Events
We focused on the frequency of single extreme events and compound events within different maize growth periods in the study area from 1981 to 2020 by dividing the number of occurrences of each event by the entire period. The frequency of extreme events (F) was calculated as (Yang et al. 2022):
where \({F}_{i}\) is the frequency of extreme events in i growth period, n is the number of years of extreme events, and N is the total number of years recorded.
3.6 Methods of Yield Separation
Meteorological yields \({(Y}_{m})\) are typically used to assess crop yield and weather disaster loss. In this study, the analysis was conducted using the relative meteorological yield (\({Y}_{R}\)), also known as the fluctuating yield, which mainly reflects yield fluctuations caused by climate change (Zhang et al. 2020). The formula used is as follows:
where \({Y}_{m}\) is the meteorological yield (kg hm−2); \(Y\) is the actual yield (kg hm−2) and \({Y}_{t}\) is the trend yield (kg hm−2); δ is the stochastic yield, which is generally ignored; and \({Y}_{R}\) is the relative meteorological yield.
4 Results
This section presents the results of our study, and includes the spatiotemporal characteristics of drought, waterlogging, and low-temperature cold damage, the spatiotemporal characteristics of compound drought and low-temperature damage events (CDLEs) and compound waterlogging and low-temperature damage events (CWLEs), and the impact of extreme and compound events on maize yields.
4.1 Spatiotemporal Characteristics of Drought, Waterlogging, and Low-Temperature Cold Damage
In this section, the temporal variation in drought, waterlogging, and cold damage in different maize growth periods is presented, and then the spatial patterns of various extreme events with different severities in different growth periods are analyzed.
4.1.1 Temporal Characteristics
The MK test was used to detect the mean values of SPRI and ICi in the study area, which helped to identify the trends and changes of drought, flooding, and low-temperature and cold damage within different growth periods.
The SPRI appeared to fluctuate dramatically in the early growth period (EP) (Fig. 3a1), with the MK test indicating the presence of multiple mutation points from 1990 to 2010 and a nonsignificant upward trend after 2008, with the sequence statistic of the time series data (UF) not exceeding the confidence level of 0.05. In the middle growth period (MP) (Fig. 3b1), SPRI changes were relatively stable, with the study area generally showing a tendency towards drought since a single mutation in 2013. All stations except Daan experienced moderate and severe droughts in 2020, with the UF exceeding the confidence level. In the late growth period (LP) (Fig. 3c1), the SPRI trend changed abruptly three times in 1984, 1987, and 2020, although none of these trends changed significantly. After 2016, waterlogging occurred more frequently and with progressively greater severity, which was attributed to the significantly higher rainfall in August in recent years. Meteorological hazard records showed that flooding occurred in most of the areas. In the whole growth period (WP) (Fig. 3d1), the SPRI showed a decreasing trend after 1996, with a drought phase during the maize growing season in 1999–2007, with two abrupt changes in 2018–2020. The Songliao Plain has mainly shown a water deficit during the MP in recent years, with waterlogging occurring mostly in the EP and LP.
In the EP (Fig. 3a2), ICi fluctuated dramatically, with multiple mutation points over the 40-year period. ICi showed a decreasing trend after 1994, which was moderated after reaching a significant level in 2001. After 2008, eight mutations in ICi were observed. Low-temperature cold damage was rare during the MP (Fig. 3b2), which was related to the abundance of heat resources in the northeast in July. As shown in Fig. 3c2, the LP was the period of frequent low-temperature cold damage, with ICi increasing from 2001 to 2020 compared to 1981–2000, and the upward trend of ICi changed significantly after 2009, indicating an increase in the intensity of low-temperature cold damage. This period is an important stage of kernel filling, and special attention should be paid to low-temperature cold damage to mitigate potential threats to maize yield. The ICi showed a decreasing and then increasing trend during the WP (Fig. 3d2), with several mutations exceeding the significance level after 2000. Although low-temperature cold damage was mitigated throughout the growing season, government departments and growers should focus on low-temperature cold damage during different maize growth periods.
4.1.2 Spatial Characteristics
The frequencies of different intensities of drought, waterlogging, and low-temperature cold damage during each reproductive period were calculated from the SPRI and ICi thresholds in Table 3 and Table 5.
Light drought dominated, followed by moderate and severe droughts during the different growth periods in the study area (Fig. 4). In the EP, droughts occurred most frequently at Fushun and Changtu stations (35%). Severe droughts occurred in 42% of the region during the MP, mainly in the northwest. In the LP, light droughts were concentrated in the eastern part of the study area, whereas moderate and severe droughts were more frequent in the northwest. The western study area, particularly the northwestern part, experienced more severe drought.
Waterlogging showed different spatial distribution patterns from drought and the affected areas were mainly concentrated in the eastern and southern parts of the Songliao Plain (Fig. 5). In the EP, light waterlogging was mainly concentrated in the northeast, with higher frequencies at Shuangyang, Changchun, and Fuyu stations (25–27.5%). In the MP, light and moderate flooding were mainly concentrated in the east and northwest, respectively. In the LP, light and moderate waterlogging occurred more frequently at high latitudes, whereas severe waterlogging occurred mainly in the southern part of the study area, with the highest frequency observed in Xinmin station (15%). The frequency of waterlogging in the Songliao Plain region from 1981 to 2020 was highest in the LP, followed by the EP and MP.
The frequency of low-temperature cold damage showed significantly different spatial distribution patterns in different growth periods (Fig. 6). In the EP, the frequency of moderate and severe low-temperature cold damage showed a decreasing trend from northwest to southeast. In the MP, light low-temperature cold damage was dominant and moderate to severe low-temperature cold damage rarely occurred. Of the 56% low-temperature cold damage that occurred during the LP, moderate and severe low-temperature cold damage occurred in the southern part of the Songliao Plain. The wide range and high frequency of low-temperature cold damage in the study area are partly due to the low ICi selection thresholds.
4.2 Spatiotemporal Characteristics of Compound Drought and Low-Temperature Cold Damage Events (CDLEs) and Compound Waterlogging and Low-Temperature Cold Damage Events (CWLEs)
In this section, the applicability of the CDLMI and CWLMI in the study area is analyzed, as well as the temporal variations and spatial distributions of CDLEs and CWLEs during different growth periods of maize.
4.2.1 Applicability of the Compound Drought and Low-Temperature Cold Damage Magnitude Index (CDLMI) and Compound Waterlogging and Low-Temperature Cold Damage Magnitude Index (CWLMI)
Measuring the intensity of compound events by comparing the intensity of a single extreme event is difficult, but we effectively achieved this by constructing compound indices (CDLMI and CWLMI) based on ΔD (or ΔW) and ΔL.
Figure 7 shows the average CDLMI and CWLMI values in different growth periods from 1981 to 2020. In the EP (Fig. 7a1), the greatest CDLMI value (0.056) was recorded in 1997 and correlated with a relatively high ΔD (0.47) and the highest L (0.14), whereas the second highest CDLMI (0.052) was recorded in 2007 and correlated with a higher ΔD (0.751) and a lower ΔL (0.072). This shows that, compared to 1997, the CDLE in 2007 had more drought damage and less low-temperature cold damage than in 1997. In the LP (Fig. 7c2), the highest ΔL was found in 1982 and the highest ΔW was found in 2020, which were significantly different, corresponding to CWLMI values of (0.00796) and (0.01926), respectively.
4.2.2 Temporal Characteristics of Compound Drought and Low-Temperature Cold Damage Events (CDLEs) and Compound Waterlogging and Low-Temperature Cold Damage Events (CWLEs)
Indicators created for the compound events were used to measure the intensities of the CDLEs and CWLEs at various growth periods in the study region (Fig. 7). In the EP (Figs. 7a1–7a2), CDLMI was higher in 1997 (0.056), 2000 (0.047), and 2007 (0.052); CWLMI was higher in 2014 (0.053) and 2015 (0.023), and large ΔD, ΔW, and ΔL corresponded to high CDLMI and CWLMI values. In the MP (Figs. 7b1–7b2), the two higher CDLMI values were in 1982 and 1997 (0.002 and 0.0006), both corresponded to large ΔD. Higher CWLMI values in 1986 and 1991 (both 0.011) were due to large ΔW, and stronger droughts and waterlogging were the main reasons for the higher intensities of CDLEs and CWLEs, respectively. Compound events had low intensity during the MP because of the low probability of low-temperature cold damage and small ΔL. In the LP (Figs. 7c1–7c2), CDLMI and CWLMI values were generally high during 1981–2000, and after 2000, CDLMI was the highest in 2009 (0.052), and CWLMI fluctuated between 0 and 0.02 during 2010–2020.
4.2.3 Spatial Characteristics of Compound Drought and Low-Temperature Cold Damage Events (CDLEs) and Compound Waterlogging and Low-Temperature Cold Damage Events (CWLEs)
The spatial distribution of the frequencies of CDLEs and CWLEs of different intensities at different growth periods in the study area from 1981 to 2020 is plotted in Figs. 8 and 9. The CDLEs during different growth periods mainly occurred in the western part of the study area, especially in the northwest, including Zhangwu, Yingkou, Tongyu, Changtu, and Qianguo stations, with the severity showing a decreasing trend from northwest to southeast, and temporally concentrated in the EP and LP and very rarely in the MP (Fig. 8). During the MP (Fig. 8b1–b3), CDLEs occurred only at Shuangyang, Liaoyuan, and Nong’an stations, with the most frequent and severe occurrence in Shuangyang. In marked contrast to CDLEs, CWLEs were concentrated in the eastern part of the study area during different growth periods, and both the frequency and intensity of occurrence were highest in the LP, followed by the EP, and lowest in the MP (Fig. 9). In the EP (Figs. 9a1–9a3), moderate and severe CWLEs were concentrated in the northwest as well as at Harbin station, while light CWLEs were concentrated in the east. Light and moderate CWLEs predominated in the MP (Figs. 9b1–9b3). Of the 47% CWLEs that occurred during the LP, severity decreased as latitude increased, and severe CWLEs were concentrated in the southeast (Figs. 9c1–9c3). Moderate and severe CWLEs increased in frequency and shifted from the northwest to the southeast as the growing season progressed.
The frequency and intensity of CDLEs and CWLEs were similar at all stations in the study area, with total counts of 393 and 390, respectively; moderate and severe CDLEs accounted for 42% and 8% of all CDLEs, respectively, and moderate and severe CWLEs accounted for 40% and 5% of all CWLEs, respectively.
4.3 Impact of Extreme and Compound Events on Maize Yields
Given the wide spatial and temporal variability of individual and compound extreme events, it is expected that there will be significant differences in their impact on crop yields. Figure 10 shows the visualization of the Pearson correlation of each indicator with the relative meteorological yield (YR) for different growth periods of each meteorological station.
In the EP and MP, the SPRI and YR were positively correlated at most stations, with significant positive correlations at stations in the northern part of the study area, where maize yields were significantly affected by drought. Negatively correlated stations were mainly at stations in the southeastern part of the Songliao Plain (p > 0.05), where waterlogging had a slight impact on yield losses. In the LP, SPRI and YR were weakly negatively correlated at most stations, except at Shuangyang, where the maize yield was significantly affected by drought. Maize yield in the study area was more susceptible to drought than waterlogging and was most severe during the MP, when the frequency and intensity of drought were higher. ICi and YR were significantly and positively correlated during the EP and LP at most stations, particularly in the southeast.
The CDLMI and YR were significantly and negatively correlated in both the early and late growth periods at most stations in the northern part of the study area. The CWLMI and YR were significantly and negatively correlated mainly in the LP at stations in the southern part of the study area, whereas during the MP, CWLEs significantly affected maize yield in Fushun. Maize yield responded differently to CDLEs and CWLEs at different growth periods. Compared with single drought, flooding, and low temperature and cold events, the correlation between YR and the two compound indices (CDLMI and CWLMI) was stronger (p < 0.05), and maize yield reduction was significantly affected by the compound events, especially CDLEs.
5 Discussion
In this section, we further analyze the spatial patterns and interdecadal variability of the compound events, as well as their impact on maize yields, and discuss the limitations of our research.
5.1 Spatial Patterns and Interdecadal Variability of Compound Events
The frequency and intensity of the compound drought and low-temperature cold damage events (CDLEs) and compound waterlogging and low-temperature cold damage events (CWLEs) in the Songliao Plain region fluctuated markedly from decade to decade. From 1991 to 2020, the frequency of CDLEs in different growth periods showed a decreasing trend in different interdecadal periods, while the frequency of CWLEs in the EP and LP showed an increasing trend. The changing characteristics of compound events may be related to precipitation-temperature dependence as well as single events. There is evidence of enhanced negative soil moisture-temperature dependence in the Northeast (Zhang et al. 2022). This negative correlation can be explained by the fact that increased rainfall is associated with an increase in soil moisture and latent heat fluxes, and therefore a decrease in sensible heat at the surface, which leads to a near-surface temperature decrease (Berg et al. 2015) that can trigger CWLEs. From the 1980s to the 2010s, the frequency of droughts in the study area was higher than that of waterlogging during both the EP and LP. The frequency, severity, and extent of droughts increased across the decades during the MP, whereas the opposite was true for waterlogging. Historical observations and model simulations have confirmed that the frequency of global drought increases under the influence of reduced precipitation and increased evapotranspiration (Yu et al. 2023). The frequency and intensity of waterlogging in the EP and LP increased significantly from the 2010s to 2020s, which is consistent with the results of other researchers (Guan et al. 2021). Although the number of rainfall days has decreased significantly in China, the intensity of precipitation has increased significantly, and heavy rainfall in the Northeast has mostly been concentrated on a few heavy precipitation days in recent years (Song et al. 2015). Influenced by climate warming, the trend of increasing heat resources in the Northeast is obvious (Zhao et al. 2015), and the frequency, intensity, and extent of low-temperature cold damage have shown a decreasing trend after the 1990s in both the EP and MP. Extreme events in China varies across regions and time series, and the frequency of compound dry/warm and wet/cold extremes generally decreases in most areas, which is slightly different from the results obtained in this study. This may be related to the larger study area and time scales of the other studies. The significant reduction in compound dry and cold extremes in the Northeast is consistent with our findings (Wu et al. 2019).
The frequencies of CDLEs and CWLEs were extremely low during the MP (Figs. 8 and 9). The heat resources in the study area varied during different growth periods, with mean daily F(t) of 81.98, 89.08, and 78.77 in the EP, MP, and LP, respectively. In this climatic context, low temperatures mainly occur in the EP and LP, whereas extreme temperature events in the MP are dominated by extremely high temperatures (Guo et al. 2023). The low frequency of low-temperature cold damage in the MP was a result of the climatic conditions in the study area. Although the frequency of drought and waterlogging in the MP is high, when considering their concurrence with low-temperature cold damage, their frequency is low.
5.2 Impact of Compound Events on Maize Yield
Extreme agrometeorological events occur more frequently during the growing season because of global warming and have a significant effect on crop growth (Zhang et al. 2021; Yang et al. 2023). Shorter photosynthesis and seed filling processes can result from precipitation and temperature conditions that vary with climate change, which in turn can alter the number of growing days and the phenological phase of crops (Wang et al. 2022). During the pre-growth period of maize, a short-term water deficit causes delayed leaf tip emergence and reduced leaf area, whereas a long-term water deficit results in shortened leaves and internodes (NeSmith and Ritchie 1992). Waterlogging causes roots to remain hypoxic, which impedes stem morphogenesis and affects nutrient uptake and dry matter partitioning in maize over the kernel (Shao et al. 2023). Exposure to cold stress during maize seed germination and seedling growth delays tasselling and flowering (Birch et al. 1998; Sanchez et al. 2014). At this time, the CDLEs and CWLEs mainly affect maize reproductive growth. Between nodulation and staminate extraction, a water deficit causes the closure of leaf stomata and degradation of chlorophyll pigments, leading to a rapid decrease in leaf photosynthetic and transpiration rates. Waterlogging reduces the N concentration in the main stem and inhibits root respiration and nutrient uptake (Otie et al. 2019; Huang et al. 2022). Low-temperatures during the large trumpet period may lead to partial pollen abortion, resulting in reduced maize nodulation and yield reduction (Feng et al. 2013). When drought (or waterlogging) and low temperature occur together, they can directly affect the nutritional growth of maize. The late reproductive period is the period when maize stem and leaf growth and kernel filling are clearly dominant, and the duration of kernel filling is a major part of the maize growing season. Moisture deprivation can lead to reduced pollen fertilization and failure of synchronous pollination. Frequent waterlogging can lead to mold and sprouting of maize, increased susceptibility to lodging, preventing harvesting, resulting in reduced yield and quality (Hu et al. 2023). Maize is very sensitive to low temperature during the kernel filling period, with temperatures below 13.5 °C leading to shorter kernel filling times and reduced kernel filling rates (Sanchez et al. 2014; Li et al. 2021). During this period, the CDLE and CWLE significantly affect the firmness and cob weight of maize.
Excessive rainfall significantly reduces maize yields in colder regions, which, combined with poor soil drainage, is exacerbated under conditions of high preseason soil water storage (Li et al. 2019). Our results show that maize yields in the southern part of the study area were more susceptible to CWLEs during the LP, when the frequency and severity of CWLEs were higher in the south. The frequency, extent, and intensity of CWLEs in the EP showed an increasing trend after the 2010s. Although in the EP the relative meteorological yield and CWLMI had no large-scale significant negative correlation, relevant authorities should take timely drainage measures to reduce crop yield losses.
During crop growth and development, crops are biologically affected by various stressors and their combinations. The effects of multiple hazards and risks must be considered, and incomplete risk analyses often lead to the underestimation of potential impacts (Zhou et al. 2022; Bi et al. 2023). Most studies have explored the possibility of compound events leading to reduced maize yields; however, the impact of compound events on maize yields at different growth periods has not yet been considered. We constructed the CDLMI and CWLMI that incorporated the characteristics of drought, waterlogging, and low-temperature cold events to help quantify the intensity of CDLEs and CWLEs and their impacts. This complements the study of compound events associated with low temperatures and is an important foundation for understanding compound events and the spatial and temporal variability of hazards and disasters.
This study also has some limitations. In this study, different growth periods of maize were used as the research temporal scale, ignoring the evolutionary characteristics of short-term extreme events. Further expansion of the time scale to a few days is needed to assess short-term events and impacts. Drivers and mechanisms of change of different compound events need to be analyzed, as well as contribution of the geology/topography (altitude, slope, landform type) to different hazards, which can help provide technical support for extreme event risk management.
6 Conclusion
The main objectives of this study were to analyze the spatial and temporal distribution characteristics of drought, waterlogging, and low-temperature cold damage during different growth periods of maize and to construct a compound magnitude index based on available meteorological indicators to quantify the magnitude of compound events during various maize growth periods. The following conclusions were drawn:
-
(1)
From 2000 to 2020, the trend of different extreme events in the Songliao Plain region was evident. In the early growth period (EP), both the standardized precipitation requirement index (SPRI) and the chilling injury index (ICi) tended towards waterlogging and low-temperature cold damage; drought remained the main meteorological hazard in both the EP and the middle growth period (MP), with 42% of the severe droughts occurring in the MP; waterlogging and low-temperature cold damage in the late growth period (LP) required special attention.
-
(2)
The compound drought and low-temperature cold damage events (CDLEs) and compound waterlogging and low-temperature cold damage events (CWLEs) occurred mainly in the EP and LP, whereas in the MP, CWLEs were chiefly confined in the study area’s northern districts. From 1991 to 2000, the frequency of CDLEs tended to decrease with advancing interdecadal change in the EP and LP, and the frequency of CWLEs tended to increase.
-
(3)
At 73% of the meteorological stations a high correlation between maize yield and drought and low-temperature cold injury was observed. The compound drought and low temperature cold damage magnitude index (CDLMI) had a strong negative correlation with the relative meteorological yield (YR) of EP as did the compound waterlogging and low temperature cold damage magnitude index (CWLMI) with the YR of the LP, implying that the occurrence of CDLEs and CWLEs during these two periods largely determined maize yield. Maize yield reduction was more significantly (p < 0.05) affected by compound events compared to single events.
References
Berg, A., B.R. Lintner, K. Findell, S.I. Seneviratne, B. van den Hurk, A. Ducharne, F. Chéruy, and S. Hagemann et al. 2015. Interannual coupling between summertime surface temperature and precipitation over land: Processes and implications for climate change. Journal of Climate 28(3): 1308–1328.
Bi, W., M. Li, B. Weng, D. Yan, Z. Dong, J. Feng, and H. Wang. 2023. Drought-flood abrupt alteration events over China. Science of The Total Environment 875: Article 162529.
Birch, C.J., G.L. Hammer, and K.G. Rickert. 1998. Temperature and photoperiod sensitivity of development in five cultivars of maize (Zea mays L.) from emergence to tassel initiation. Field Crops Research 55(1): 93–107.
Cai, J., C. Wang, and J. Zhang. 2013. An assessment of the drought disaster and chilling injury hazard for maize in the various growth stages in northeast China. Acta Meteorologica Sinica 71(5): 976–986.
Chen, H., and J. Sun. 2015. Changes in drought characteristics over China using the Standardized Precipitation Evapotranspiration Index. Journal of Climate 28(13): 5430–5447.
Chen, Y., Z. Zhang, P. Wang, X. Song, X. Wei, and F. Tao. 2016. Identifying the impact of multi-hazards on crop yield—A case for heat stress and dry stress on winter wheat yield in northern China. European Journal of Agronomy 73: 55–63.
Dang, Y., L. Qin, L. Huang, J. Wang, B. Li, and H. He. 2022. Water footprint of rain-fed maize in different growth stages and associated climatic driving forces in Northeast China. Agricultural Water Management 263: Article 107463.
De Luca, P., and M.G. Donat. 2023. Projected changes in hot, dry, and compound hot-dry extremes over global land regions. Geophysical Research Letters 50(13): Article e2022GL102493.
Dhaliwal, D.S., and M.M. Williams. 2022. Evidence of sweet corn yield losses from rising temperatures. Scientific Reports 12(1): Article 18218.
Eck, M.A., A.R. Murray, A.R. Ward, and C.E. Konrad. 2020. Influence of growing season temperature and precipitation anomalies on crop yield in the southeastern United States. Agricultural and Forest Meteorology 291: Article 108053.
Feng, S., Z. Hao, X. Zhang, and F. Hao. 2019. Probabilistic evaluation of the impact of compound dry-hot events on global maize yields. Science of The Total Environment 689: 1228–1234.
Feng, S., Z. Hao, X. Zhang, and F. Hao. 2021. Changes in climate-crop yield relationships affect risks of crop yield reduction. Agricultural and Forest Meteorology 304–305: Article 108401.
Feng, R., J. Wu, R. Ji, Y. Zhang, L. Wang, H. Wang, and W. Yu. 2013. Analysis of growth parameters and yield of spring maize under chilling stress. Agricultural Research in the Arid Areas 31(1): 183–187.
Guan, X., Y. Zang, Y. Meng, Y. Liu, H. Lv, and D. Yan. 2021. Study on spatiotemporal distribution characteristics of flood and drought disaster impacts on agriculture in China. International Journal of Disaster Risk Reduction 64: Article 102504.
Guo, Y., J. Zhang, K. Li, H. Aru, Z. Feng, X. Liu, and Z. Tong. 2023. Quantifying hazard of drought and heat compound extreme events during maize (Zea mays L.) growing season using Magnitude Index and Copula. Weather and Climate Extremes 40: Article 100566.
Guo, J., L. Zhuang, and Y. Chen. 2009. Study on forecasting methods of corn heat index in Northeastern China (I)——Heat index and corn yield. Journal of Catastrophology 24(4): 6–10 (in Chinese).
Hu, J., W. Yu, P. Liu, B. Zhao, J. Zhang, and B. Ren. 2023. Responses of canopy functionality, crop growth and grain yield of summer maize to shading, waterlogging, and their combination stress at different crop stages. European Journal of Agronomy 144: Article 126761.
Huang, C., W. Zhang, H. Wang, Y. Gao, S. Ma, A. Qin, Z. Liu, et al. 2022. Effects of waterlogging at different stages on growth and ear quality of waxy maize. Agricultural Water Management 266: Article 107603.
Kim, G.-U., H. Oh, Y.S. Kim, J.-H. Son, and J.-Y. Jeong. 2023. Causes for an extreme cold condition over Northeast Asia during April 2020. Scientific Reports 13(1): Article 3315.
Li, Y., K. Guan, G.D. Schnitkey, E. DeLucia, and B. Peng. 2019. Excessive rainfall leads to maize yield loss of a comparable magnitude to extreme drought in the United States. Global Change Biology 25(7): 2325–2337.
Li, X.F., Z.B. Guo, H.X. Zhu, P. Wang, L.J. Gong, L.X. Jiang, and H.Y. Zhao. 2020. Time-series characteristics of drought and flood in spring soybean growing season and its effect on soybean yield in Heilongjiang Province. China. The Journal of Applied Ecology 31(4): 1223–1232.
Li, Z., Z. Zhang, J. Zhang, Y. Luo, and L. Zhang. 2021. A new framework to quantify maize production risk from chilling injury in Northeast China. Climate Risk Management 32: Article 100299.
Ling, M., X. Guo, X. Shi, and H. Han. 2022. Temporal and spatial evolution of drought in Haihe River Basin from 1960 to 2020. Ecological Indicators 138: Article 108809.
Liu, W., Z. Li, Y. Li, T. Ye, S. Chen, and Y. Liu. 2022. Heterogeneous impacts of excessive wetness on maize yields in China: Evidence from statistical yields and process-based crop models. Agricultural and Forest Meteorology 327: Article 109205.
Lobell, D.B., W. Schlenker, and J. Costa-Roberts. 2011. Climate trends and global crop production since 1980. Science 333(6042): 616–620.
NeSmith, D.S., and J.T. Ritchie. 1992. Short- and long-term responses of corn to a pre-anthesis soil water deficit. Agronomy Journal 84(1): 107–113.
Otie, V., A. Ping, I. Udo, and E. Eneji. 2019. Brassinolide effects on maize (Zea mays L.) growth and yield under waterlogged conditions. Journal of Plant Nutrition 42(8): 954–969.
Sanchez, B., A. Rasmussen, and J.R. Porter. 2014. Temperatures and the growth and development of maize and rice: A review. Global Change Biology 20(2): 408–417.
Shao, J., P. Liu, B. Zhao, J. Zhang, X. Zhao, and B. Ren. 2023. Combined effects of high temperature and waterlogging on yield and stem development of summer maize. The Crop Journal 11(2): 651–660.
Shi, J., L. Cui, and Z. Tian. 2020. Spatial and temporal distribution and trend in flood and drought disasters in East China. Environmental Research 185: Article 109406.
Song, X., S. Song, W. Sun, X. Mu, S. Wang, J. Li, and Y. Li. 2015. Recent changes in extreme precipitation and drought over the Songhua River Basin, China, during 1960–2013. Atmospheric Research 157: 137–152.
Sun, J., T. Liu, S. Xie, J. Xiao, L. Huang, Z. Wan, and K. Zhong. 2023. Will extreme temperature events emerge earlier under global warming? Atmospheric Research 288: Article 106745.
Wang, P., L. Han, G. Zhou, and H. Liang. 2015. The effects of climate warming on different cultivars of spring maize under climate warming in Northeast China and the adaptation countermeasures. Journal of Natural Resources 30(8): 1343–1355 (in Chinese).
Wang, T., N. Li, Y. Li, H. Lin, N. Yao, X. Chen, D. Li Liu, Q. Yu, and H. Feng. 2022. Impact of climate variability on grain yields of spring and summer maize. Computers and Electronics in Agriculture 199: Article 107101.
Wang, W., Y. Zhang, B. Guo, M. Ji, and Y. Xu. 2021. Compound droughts and heat waves over the Huai River Basin of China: From a perspective of the magnitude index. Journal of Hydrometeorology 22(11): 3107–3119.
Wei, S., K. Li, J. Zhang, Y. Yang, C. Liu, and C. Wang. 2021. Hazard Assessment of peanut drought and flood disasters in Huang-Huai-Hai Region. Journal of Applied Meteorolgical Science 32(5): 629–640 (in Chinese).
Wu, X., Z. Hao, F. Hao, and X. Zhang. 2019. Variations of compound precipitation and temperature extremes in China during 1961–2014. Science of The Total Environment 663: 731–737.
Yang, Y., K. Li, S. Wei, S. Guga, J. Zhang, and C. Wang. 2022. Spatial-temporal distribution characteristics and hazard assessment of millet drought disaster in Northern China under climate change. Agricultural Water Management 272(C): Article 107849.
Yang, W., L. Zhang, and Y. Gao. 2023. Drought and flood risk assessment for rainfed agriculture based on Copula-Bayesian conditional probabilities. Ecological Indicators 146: Article 109812.
Yu, E., D. Liu, J. Yang, J. Sun, L. Yu, and M.P. King. 2023. Future climate change for major agricultural zones in China as projected by CORDEX-EA-II, CMIP5 and CMIP6 ensembles. Atmospheric Research 288: Article 106731.
Zhang, Y., Z. Hao, S. Feng, X. Zhang, and F. Hao. 2022. Changes and driving factors of compound agricultural droughts and hot events in eastern China. Agricultural Water Management 263: Article 107485.
Zhang, F., X. Yang, S. Sun, J. Gao, Z. Liu, Z. Zhang, and T. Liu. 2021. A spatiotemporal analysis of extreme agrometeorological events during selected growth stages of maize (Zea mays L.) from 1960 to 2017 in Northeast China. Theoretical and Applied Climatology 143(3): 943–955.
Zhang, J., Z. Zhang, and F. Tao. 2020. Rainfall-related weather indices for three main crops in China. International Journal of Disaster Risk Science 11(4): 466–483.
Zhao, Y., Y. Guo, R. Wang, K. Li, G. Rong, J. Zhang, and C. Zhao. 2023. Characteristics of drought, low temperature, and concurrent events of maize in Songliao Plain. International Journal of Climatology 43(7): 3041–3071.
Zhao, J., J. Mu, and J. Guo. 2015. Response of agricultural thermal resources equal to and more than 10°C to climate change over the past 50 years in northeast China. Journal of Natural Disasters 24(3): 190–198 (in Chinese).
Zhao, J.-F., X.-G. Yang, and Z.-J. Liu. 2009. Influence of climate warming on serious low temperature and cold damage and cultivation pattern of spring maize in Northeast China. Acta Ecologica Sinica 29(12): 6544–6551.
Zhou, Y., P. Zhou, J. Jin, C. Wu, Y. Cui, Y. Zhang, and F. Tong. 2022. Drought identification based on Palmer drought severity index and return period analysis of drought characteristics in Huaibei Plain China. Environmental Research 212: Article 113163.
Zscheischler, J., O. Martius, S. Westra, E. Bevacqua, C. Raymond, R.M. Horton, B. van den Hurk, and A. AghaHouchak et al. 2020. A typology of compound weather and climate events. Nature Reviews Earth & Environment 1(7): 333–347.
Acknowledgments
This study was supported by the National K&D Program of China (2022YFD2300201), the National Natural Science Foundation of China (U21A2040), the Major Science and Technology Program of Jilin Province (YDZJ202303CGZH023), the National Natural Science Foundation of China (42077443), the Science and Technology Development Planning of Jilin Province (20210203153SF), the Key Scientific and Technology Research and Development Program of Jilin Province (20200403065SF), and the Construction Project of the Science and Technology Innovation Center (20210502008ZP).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Chen, D., Guo, Y., Wang, R. et al. Quantifying Multi-hazards and Impacts Over Different Growth Periods of Maize: A Study Based on Index Construction. Int J Disaster Risk Sci 14, 822–839 (2023). https://doi.org/10.1007/s13753-023-00516-8
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13753-023-00516-8