Abstract
The Indian Ocean Dipole (IOD) is a prominent mode of climate variability in the tropical Indian Ocean (IO). It exerts a significant influence on biological activities in this region. To elucidate the biological response to the IOD, previous research has introduced the Biological Dipole Mode Index (BDMI). However, the delineation of the region by the BDMI has limitations in capturing IOD-induced chlorophyll variations in the IO. By analyzing observational data and historical simulations from a Coupled Model Intercomparison Project (CMIP) model, this study shows that chlorophyll levels in the IO exhibit a dipole pattern in response to IOD. During the developing and mature phases of the IOD, we observe a substantial decrease in chlorophyll in the south-southwest of India, contrasting with a pronounced increase in the southeast of the IO. This response is attributed to anomalous southeasterly winds induced by IOD, which enhance nutrient upwelling in the southeastern IO and suppress it in the south-southwest of India, resulting in corresponding changes in surface chlorophyll blooms. Based on this finding, we propose a new Biological Dipole Index that more robustly explains the surface chlorophyll response to IOD in the tropical IO. This study highlights the profound influence of IOD on oceanic chlorophyll and underscores the importance of a more comprehensive understanding of the associated biophysical interactions.
Introduction
Marine phytoplankton play a key role in global biogeochemical cycles (Field et al 1998), contributing to the marine food web (Richardson and Schoeman 2004) and regulating the global climate and carbon cycle (Murtugudde et al 2002, Gregg et al 2003). Previous studies have reported that surface chlorophyll-a (hereafter chlorophyll) concentrations are an important indicator of phytoplankton productivity (Field et al 1998). The western Indian Ocean (IO) has been found to be one of the most productive regions in the tropical oceans (Lee et al 2005), and therefore many studies have conducted to understand the associated dynamics of biophysical interactions (Goes et al 2005, Wiggert et al 2005). For example, Roxy et al (2016) investigated the changes in chlorophyll concentration in the western IO and reported an alarming decline in phytoplankton during summer. Further they showed that the enhanced ocean stratification driven by the rapid warming in the IO is a key factor for the declining chlorophyll in the IO. In addition, the biological productivity exhibits a pronounced seasonal cycle in the IO as a result of monsoon variability, which modulates upwelling, entrainment and convective mixing in the upper ocean (McCreary et al 1996, Kone et al 2009, Vijith et al 2016). On the other hand, the seasonal evolution of surface chlorophyll in the IO has been found to be significantly modulated by tropical climate variability, such as El Nino-Southern Oscillation (ENSO, Behrenfeldet al 2006, Wiggert et al 2009, Currie et al 2013, Racault et al 2017) and Indian Ocean Dipole (IOD, Brewin et al 2012, Park and Kug 2014). Thus, the chlorophyll in the IO is robustly influenced by the local and remote forcings.
The IOD is a prominent and dynamic climate phenomenon that exerts a profound influence on weather and climate patterns throughout the IO region (Saji et al 1999). This interannual air-sea coupled climate mode is characterized by a dipole pattern in sea surface temperature (SST) anomalies between the western and eastern equatorial IO (Saji et al 1999). IOD typically manifests in two phases: the positive phase is characterized by warmer SST anomalies in the western IO (10° S-10° N, 50° E-70° E) and cooler SST anomalies in the eastern IO (10° S-0°, 90° E-110° E), and vice versa for the negative phase. These SST anomalies trigger variations in the atmospheric circulation, leading to shifts in the position of the Walker circulation, changes in monsoon patterns and altered precipitation regimes in the affected regions (Sajiet al 1999, Webster et al 1999). Thus, the IOD is a crucial component of the Earth’s climate system (Saji and Yamagata 2003), affecting not only regional climate variability, but also having far-reaching effects on weather systems (Chan et al 2008), agriculture (Ashok and Saji 2007), ecosystems (Marchant et al2006, Park and Kug 2014) and socio-economic conditions (Feng et al 2022) in countries bordering the IO.
In addition to the effects on the physical structure of the IO, there are biological consequences of IOD with potential importance for the marine ecosystem, fishery resources and carbon sequestration (Currieet al 2013). The IOD-induced changes in the surface circulation can lead to upwelling/downwelling and mixing and regulate the supply of nutrients to the surface causing robust changes in surface chlorophyll in the IO (Sarma 2006, Chen et al 2013). However, the relationship between IOD and ocean biogeochemistry is complex and depends on the specific phase of IOD (positive or negative) and their regional effects (Sari et al 2020). Therefore, the biological response to IOD has been widely studied (Vinayachandran et al2002, Chen et al 2013, Currie et al 2013, Sankar et al 2019, Thushara and Vinayachandran 2020, Shi and Wang 2021). For example, Murtugudde et al (1999) reported a strong phytoplankton bloom in the eastern equatorial IO, an area characterized by low climatological productivity, due to the intense positive IOD of 1997. Consistently, similar responses were also observed during the 2019 positive IOD event (Shi and Wang 2021). Although similar blooms were observed during the 2006 event, Wiggert et al (2009) showed that the biological responses to the 1997 event were more intense, suggesting that IOD intensity is an important factor for the intensity of chlorophyll. Furthermore, Wiggert et al (2009) showed that the 1997 and 2006 events caused a decrease in surface chlorophyll in the Arabian Sea (AS), an area characterized by high climatological productivity. On the other hand, several studies have reported an increase in surface chlorophyll in the southeastern AS during negative IOD events due to enhanced Ekman suction and coastal upwelling, as well as a shoaling thermocline in the region (e.g., Thushara and Vinayachandran 2020). Thus, in addition to the differences in chlorophyll intensity, the biological responses to IOD also show a diverse spatial structure.
Given that IOD influences the chlorophyll variability of the tropical IO, Shi and Wang (2021) proposed a biological IOD, and later the authors introduced the biological dipole mode index (BDMI, Shi and Wang 2022) to characterize and quantify IOD-induced changes in surface chlorophyll in the IO. The BDMI is defined as the difference between the average chlorophyll anomaly in the western IO (10° S-10° N, 50° E-70° E) and the eastern IO (10° S-0°, 90° E-110° E), where the chosen region is the same as that used to define the dipole mode index (DMI). However, there are some limitations in the index region that prevent an accurate representation of the chlorophyll variability in response to IOD in the IO. As shown in the composite analysis of positive and negative IODs (Fig. S1), IOD-induced changes in chlorophyll remain weak in the western equatorial IO. This raises questions about the accuracy of using chlorophyll in the western equatorial IO to define the BDMI in previous studies, and therefore possible clarifications are sought to understand the IOD-induced chlorophyll changes (e.g., intensity and spatial pattern) in the tropical IO. Furthermore, understanding these interactions is crucial for understanding the broader ecological and environmental consequences of IOD events over the Indian Ocean. Therefore, in the present study we investigate the biological consequences of IOD.
Data and methods
To investigate the effect of IOD on chlorophyll, we used monthly chlorophyll from the European Space Agency Ocean Color Climate Change Initiative (ESA-OC-CCI v4, Sathyendranath et al 2018). In addition, we used monthly SST data from extended reconstructed SST (ERSSTv5, Boyin et al 2017), 10 m zonal and meridional winds from National Centers for Environmental Prediction (NCEP2), subsurface temperature from NCEP Global Ocean Data Assimilation System (GODAS, Kanamitsu et al 2002), and surface wind stress from European Center for Medium-Range Weather Forecasts (ECMWF) reanalysis (ERA5).
Along with observations, we used historical simulations from 13 CMIP6 coupled general circulation models, including chlorophyll, SST, low-level (850 hPa) winds, subsurface temperature, and nutrients (nitrate and phosphate, Table S1), to provide supporting evidence for the observed relationships. It should be mentioned that these 13 models (Table S1) among CMIP6 models are selected because they only provide full 3-D datasets of ocean biogeochemistry. However, many of the CMIP models fail to reproduce the spatial pattern of climatology in chlorophyll concentrations and interannual variability in the IO (Roxyet al 2016). Therefore, following Roxy et al (2016), we selected a single model to investigate the biophysical response to IOD based on the pattern correlation coefficient (PCC > 0.5) of mean chlorophyll and the interannual variability of chlorophyll (standard deviation) (Fig. S2). The model chosen is the Geophysical Fluid Dynamics Laboratory Earth System Model (GFDL-ESM4, Dunne et al 2020). GFDL-ESM4 simulates the mean climatology of chlorophyll concentrations and the interannual variability of chlorophyll with a relatively small bias compared to the other 12 CMIP6 models (Fig. S2).
In addition, the dipole mode index (DMI) is calculated as the difference between the averaged SST anomaly in the western Indian Ocean (10° S-10° N, 50° E-70° E) and the eastern Indian Ocean (10° S-0°, 90° E-110° E, Saji et al 1999). We also examined the biological dipole mode index (BDMI) introduced by Shi and Wang (2022). The BDMI is defined as the difference of average chlorophyll anomaly in the western Indian Ocean (10° S-10° N, 50° E-70° E) and the eastern Indian Ocean (10° S-0°, 90° E-110° E). Here, all anomalies are calculated by removing the seasonal cycle and the long-term linear trend. The datasets have been analyzed for the period 1997-2020 for the observations and for the period 1850-2014 for the CMIP6 simulations.
Results