印度洋偶极子（Indian Ocean Dipole，IOD）是印度洋固有的一种气候现象，通常发生在北半球秋季，会影响印度洋周边国家和地区乃至我国的天气状况。IOD同时受到厄尔尼诺与南方涛动（El Niño-Southern Oscillation，ENSO）和印度洋内部变率的影响。因增暖中心和气候影响的差异，ENSO被分为两种主要类型，即东太平洋型（Eastern Pacific，EP）和中太平洋型（Central Pacific，CP）。然而，当前研究并未充分讨论两种类型ENSO对IOD的影响机制，也未量化两类ENSO和内部变率的影响程度。在此，以上因素对IOD的贡献可以通过一种新型联合线性回归方法进行估算。该方法分离了两类ENSO与内部变率的影响，结果表明印度洋内部变率是IOD海温变化主要来源，占比可达60%以上；ENSO的总贡献占约三分之一，其中以CP型为主，而EP型则倾向于在极端事件中影响IOD。ENSO与内部变率对IOD的作用机制不同：ENSO主要通过Walker环流影响印度洋风场，但因为EP型和CP型的增暖核心不同，两者影响程度有所差异；内部变率则倾向于通过印度洋内部的海洋过程造成海温异常进而引起IOD。此外，由于厄尔尼诺生命期更长，与之共发的IOD正事件有更大几率转化为下一年春季的印度洋海盆尺度增暖，其中ENSO总贡献超过70%。虽然内部变率对该转化并无显著统计关系，但强的IOD正事件仍有机会触发海盆尺度增暖。以上结果有助于提高对气候模态以及跨海盆相互作用的认识。

The Indian Ocean Dipole (IOD) is an intrinsic climate mode in the Indian Ocean, typically occurring during the boreal fall, influencing weather and climate in surrounding regions and even China. It is affected by both the El Niño-Southern Oscillation (ENSO) and internal variability within the Indian Ocean. However, the quantification of two types of ENSO, namely the Eastern Pacific (EP) and the Central Pacific (CP), and internal variability remains unclear. Here, a binary combined linear regression method can be used to separate and estimate the contributions of these three factors. The results show that internal variability is the primary source of IOD sea surface temperature (SST) changes, accounting for more than 60% of the variance. The contribution of ENSO is about one-third, predominantly driven by the CP ENSO, whereas the EP type tends to influence the IOD during extreme events. Their influencing mechanisms are different: ENSO primarily influences the Indian Ocean wind field via the Walker circulation, of which efficiency depends on the location of the warm core with EP and CP types accordingly. In comparison, internal variability tends to cause SST anomalies through oceanic processes that facilitate IOD. Due to the longer lifetime of El Niño events, the co-occurring positive IOD has a higher chance of transforming into the Indian Ocean basin-wide warming in the following spring, in which ENSO contributes more than 70%. Although internal variability does not show a significant statistical relationship with this transition, a strong positive IOD still has the potential to induce the following basin-wide warming. These findings would improve understanding of the climate modes and inter-basin interactions.