AS | 并发干旱威胁小麦和玉米生产，并将在未来扩大作物产量差距

图1：185个研究农业气象站点的海拔、农业区划和分布。农业子区域及其种植的作物：东北平原（NE，春小麦和春玉米）、黄土高原（LP，春玉米、夏玉米和冬小麦）、黄淮海平原（HHH，夏玉米和冬小麦）、北方干旱半干旱区（NA，春玉米、春小麦和冬小麦）、青藏高原（QT，春小麦）和长江以南地区（SY，冬小麦）。

Fig. 1. The elevation, division of agricultural regions and distribution of 185 studied agri-meteorological sites. The agricultural sub-regions and crops grown therein are: Northeast Plain (NE, spring wheat and spring maize), Loess Plateau (LP, spring maize, summer maize and winter wheat), Huang-Huai-Hai plain (HHH, summer maize and winter wheat), Northern Arid and Semi-Arid Region (NA, spring maize, spring wheat and winter wheat), Qinghai-Tibet Plateau (QT, spring wheat) and South of Yangtze River Region (SY, winter wheat).

图2：四种作物从1962年到2100年的营养期（a）、生殖期（b）和整个生长期（c）的变化（以春玉米为例，其他见图S1）。2060年代（d）和2100年（e）在四种SSP情景下的整个生长期。潜在产量（f）、雨养产量（g）和产量差距（h）从1962年到2100年的变化。2060年代（i）和2100年（j）在四种SSP情景下的产量差距。实线表示所有站点的平均值，阴影区域表示所有站点的上下四分位数范围。

Fig. 2. Variations in duration of vegetative (a), reproductive (b), and whole growth stages (c) for four crops from 1962 to 2100 (spring maize as an example, see Fig. S1 for others). Durations of whole growth stages in the 2060s (d) and 2100 (e) under the four SSP scenarios. Variations in potential yield (f), rainfed yield (g), and yield gaps (h) from 1962 to 2100. Yield gaps in the 2060s (i) and 2100 (j) under four SSPs. The solid line indicates the average values of all site and the shaded area indicates the upper and lower interquartile value ranges for all sites.

图3：1962–2021年历史时期和2022–2100年未来时期在SSP2-4.5和SSP5-8.5情景下，四种作物在整个生长期的NSPEI（a–c）和SSMI（d–f）MK趋势（详情见图S6）。

Fig. 3. MK trend of NSPEI (a–c) and SSMI (d–f) during the whole growth period in historical (1962–2021) and future (2022–2100) periods for the four crops under the SSP 2-4.5 and SSP5-8.5 scenarios. (see Figs. S6 for details).

图4：在SSP2-4.5和SSP5-8.5情景下，1962–2021年历史时期和2022–2100年未来时期，春小麦在整个生长期的气象、农业和并发干旱的频率（其他作物见图S7–S8）。

Fig. 4. Frequencies of meteorological, agricultural and concurrent droughts throughout the whole growth period of spring wheat in historical (1962–2021) and future (2022–2100) periods under the SSP 2-4.5 and SSP5-8.5 scenarios. (see Figs. S7–S8 for other crops).

图5：产量差距与NSPEI（a，b）或SSMI（c，d）之间的双变量交叉小波共生光谱。以春小麦在历史和SSP2-4.5情景下的波谱为例（其他作物见图S6和S7）。粗黑色轮廓描绘了相对于红噪声的局部功率的5%显著性水平；影响锥（COI）呈现为较浅的阴影。向右箭头表示同相信号，向左箭头表示反相信号。箭头方向代表相位关系，其中同相为正相关（指向右），反相为负相关（指向左）。反映BWC光谱显著相干性的百分比区域统计（g）。细斜线表示给定作物在给定情景下NSPEI和SSMI的PASC结果。饼图显示了在BWC结果中产量差距与NSPEI或SSMI的较高PASC的干旱指数统计数据。产量差距对NSPEI和SSMI的多变量小波相干光谱（e，f）。与BWC光谱相比，选择指数的MWC光谱中PASC的变化（h）（详情见图S9–S11和表S3–S4）。

Fig. 5. Bivariate cross-wavelet coalescent spectra between yield gap and NSPEI (a, b) or SSMI (c, d). The wave spectra of spring wheat in historical and SSP2-4.5 are used as examples (see Figs. S6 and S7 for other crops). Thick black contours depict the 5% significance level of local power relative to red noise; the cone of influence (COI) is presented as a lighter shade. Right-pointing arrows indicate in-phase signals while left-pointing arrows indicate anti-phase signals. The phase relationship is represented by the arrow direction, where the in-phase is a positive correlation (point to the right) and the anti-phase is a negative correlation (point to the left). The statistics of the percent area of significant coherence reflected by BWC spectra (g). Thin diagonal line indicates the PASC results of the NSPEI and SSMI corresponding to a given crop in a given scenario. The pie charts show statistics for the drought index with a higher PASC in the BWC results for yield difference and NSPEI or SSMI. The multiple wavelet coherence spectra of the yield gaps for NSPEI and SSMI (e, f). Changed PASCs in MWC spectra relative to BWC spectra with the selected index (h) (see Figs. S9–S11 and Table S3–S4 for details). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

图6：在1962–2021年和2022–2100年SSP2-4.5和SSP5-8.5情景下，不同NSPEI和SSMI组合下的产量差距（a–c，以春小麦为例，其他作物见图S12）。在1962–2021年和2022–2100年SSP2-4.5和SSP5-8.5情景下，不同并发干/湿水平下NSPEI和SSMI识别的产量差距（d–f，其他作物见图S13。N：无干旱；SD：轻度干旱；MD：中度干旱；HD：重度干旱；SW：轻度湿；MW：中度湿；HW：重度湿）。

Fig. 6. Scatters of yield gap under different combinations of NSPEI and SSMI in 1962–2021 and 2022–2100 under SSP2-4.5 and SSP5-8.5 scenarios (a–c, using spring wheat as an example, see Fig. S12 for other crops). Yield gaps in different concurrent dry/wet levels identified by NSPEI and SSMI in 1962–2021 and 2022–2100 under SSP2-4.5 and SSP5-8.5 scenarios (d–f, see Fig. S13 for other crops. N: no drought; SD: slight drought; MD: moderate drought; HD: heavy drought; SW: slight wet, MW: moderate wet; HW: heavy wet.).

图7：在1962–2021年和2022–2100年SSP2-4.5和SSP5-8.5情景下，轻度、中度和重度单一和并发干旱下的产量差距（其他情景见图S14）。

Fig. 7. Yield gaps under slight, moderate, and heavy levels of single and concurrent droughts in 1962–2021 and 2022–2100 under SSP2-4.5 and SSP5-8.5 scenarios (see Fig. S14 for other scenarios).

图8：在1962–2021年（a，d）和2022–2100年SSP2-4.5（b，e）和SSP5-8.5（c，f）情景下，并发中度干旱导致的春玉米、夏玉米、春小麦和冬小麦的产量差距的空间分布。

Fig. 8. Spatial distribution of yield gaps due to concurrent moderate droughts for spring maize, summer maize, spring wheat and winter wheat in 1962–2021 (a, d) and 2022–2100 under SSP2-4.5 (b, e) and SSP5-8.5 (c, f) scenarios.

图9：在1962–2021年（a）和2022–2100年SSP2-4.5（b）和SSP5-8.5（c）情景下，单一类型气象（NSPEI < -0.5，SSMI > -0.5）或农业干旱（SSMI < -0.5，NSPEI > -0.5）条件下的产量差距（其他作物见图S15）。

Fig. 9. Yield gaps in single-type meteorological (NSPEI < −0.5, SSMI > −0.5) or agricultral drought (SSMI < −0.5, NSPEI > −0.5) conditions in 1962–2021(a) and 2022–2100 under SSP2-4.5 (b) and SSP5-8.5 (c) scenarios (see Fig. S15 for other crops).

图10：在2022–2100年SSP2-4.5（b）和SSP5-8.5（c）情景下，春季和夏季玉米（a–d）以及春季和冬季小麦（e–h）的NSPEI和SSMI相对权重。

Fig. 10. Relative weights of NSPEI and SSMI for spring and summer maize (a–d) and spring and winter wheat (e–h) in 2022–2100 under SSP2-4.5 (b) and SSP5-8.5 (c) scenarios.

引用：Hou, M. et al. Concurrent drought threatens wheat and maize production and will widen crop yield gaps in the future. Agricultural Systems 220, 104056 (2024).