Abstract
Tropical disturbances play an important role in the formation of tropical cyclones (TCs), but their climate characteristics are still poorly understood. The present study investigated the characteristics of tropical disturbances over the western North Pacific (WNP) for the period 1965–2020 using the reanalysis datasets. An algorithm based on the threshold of synoptic-scale vorticity and convection was developed to identify and track tropical disturbances. Approximately 116 tropical disturbances were detected primitively in the WNP from June to November each year, and about 13.1% of them subsequently developed into TCs. These developing disturbances contributed to the formation of 80.9% TCs in the WNP. Compared with nondeveloping disturbances, the position of developing disturbances is more eastward and southward, whereas the majority of their lifetimes and propagation speeds are similar. The annual numbers of developing disturbances exhibit significant interannual (interdecadal) variability. The percentage of TCs associated with disturbances also experiences a significant inverse interdecadal relationship with the number of TC genesis in the WNP, indicating that the formation probability of TCs is inconsistent with the transition probability from disturbances to TCs. In addition, the long-term trends of developing and nondeveloping disturbances suggest that global warming might be responsible for more tropical disturbances in the WNP instead of higher transition probability from disturbances to TCs.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Tropical cyclones (TCs) are one of the most severe weather systems causing natural disasters, especially in the western North Pacific (WNP). Over the past 40 years, the intense TC genesis frequency especially in the WNP has increased significantly (e.g., Webster et al. 2005; Mei and Xie 2016). Several studies have projected that the genesis frequency of intense TCs tends to increase in the future climate (e.g., Emanuel et al. 2008; Knutson et al. 2010). However, weak TC genesis frequency has decreased over the past half century (Webster et al. 2005). Thus, it is of great importance to investigate the genesis of TCs, which has not been fully understood up to now. Previous studies have revealed that TC genesis requires favorable large-scale environments. Gray (1968) proposed six necessary but insufficient conditions for TC genesis. These conditions include sufficient Coriolis Force, weak vertical wind shear between 200 and 850-hPa, large low-level relative vorticity, high sea surface temperature (SST), high mid-level relative humidity, and large conditional instability. However, the traditional thermodynamic and dynamic conditions mentioned above cannot fully describe the genesis of TCs. Mainly focusing on these factors, earlier studies have manifested that the largest determiner governing the TC genesis varies in different timescales and regions, and that there are also interannual and interdecadal changes for TC genesis itself (e.g., Camargo et al. 2007; Liu and Chan 2008; Sharmila and Walsh 2017). Some studies (e.g., Sharmila and Walsh 2017) found that the dynamic conditions are more important than the thermodynamic ones over the WNP since the in-situ SST is sufficiently warm for TC genesis.
Many studies (e.g., Zehr 1992; Thorncroft and Hodges 2001; Chen 2006) have noted that a TC generally formed from a tropical disturbance. More TCs tend to occur in the active phase of tropical disturbances (Frank and Roundy 2006; Schreck et al. 2012; Wu and Takahashi 2018). However, only a small fraction of tropical disturbances can develop into TCs (Fu et al. 2012; Hennon et al. 2013), which we named the developing disturbances. The rest of the disturbances are called nondeveloping disturbances. The structural differences in developing and nondeveloping disturbances and their relationship with TC genesis have been investigated by many studies through a composite analysis. Developing disturbances tend to emerge accompanied by stronger cyclonic circulation in the middle-to-lower level (McBride and Raymond 1981; Lee 1989; Gray 1998) and higher moisture content in the lower troposphere (Kerns and Chen 2013; Brammer and Thorncroft 2015), compared to the nondeveloping disturbances. Note that these key characteristics of disturbances associated with TC genesis show differences between different ocean basins (e.g., Fu et al. 2012; Peng et al. 2012; Murakami et al. 2013; Raavi and Walsh 2020). The thermodynamic variables such as water vapor and SST are more important than dynamic variables for the disturbance development and TC genesis in the North Atlantic (Fu et al. 2012). The dynamic variables such as vorticity at the lower level is fundamental for tropical disturbances over the WNP (Peng et al. 2012). Many previous studies have focused on the dependence of the genesis process on developing disturbances and their interaction with the large-scale circulation based on few cases, especially genesis from African easterly waves in the Atlantic (Norquist et al. 1977; Ross et al. 2009, 2012). It is shown that their development is closely related to the African easterly jet and convection over Africa. In the WNP, the enhanced tropical depression (TD) disturbance activity during El Niño may help generate more seed disturbances for TC genesis (Wu et al. 2014; Feng et al. 2020a, b). Thus, one must be cautious that the dominant parameters are different in different basins, and thus specified environmental factors should be considered when focusing on a certain basin. Additionally, few studies have investigated the characteristics of developing disturbances in long-term timescales in the WNP, and how the synoptic-scale disturbances evolve on interannual or longer timescales remains unclear.
Lacking in-situ observations, few earlier studies focused on characteristics of tropical disturbances on interannual and longer timescales. Using reanalysis or satellite datasets to detect tropical disturbances has provided a good chance to explore the climate variability of tropical disturbance activity and their relationship with TC genesis. Several methods for tracking tropical disturbances have been used in previous studies. Most of them have identified tropical disturbances by their vortex structure characteristics in the middle-to-lower troposphere (Carlson 1969; Thorncroft and Rowell 1998; Kerns and Zipser 2009; Wang et al. 2012; Peng et al. 2012). Some studies (Peng et al. 2012; Wang and Hankes 2014; Raavi and Walsh 2020) further considered other features such as lifetime and propagation speed when identifying the disturbances, with the weak and short-lived disturbances excluded. For example, Thorncroft and Hodges (2001) defined the developing disturbances using three criteria, including the threshold of the relative vorticity in a 5° square box greater than the artificially-set threshold of 0.5 × 10–5 s−1, the lifetime of more than two days, and the westward distance of more than 10° longitudes. With the development of satellite cloud observations, convection clouds can help detect tropical disturbances as stated in several studies. Chen et al. (2008) checked the effectiveness of the vortex-based method by detecting the outgoing longwave radiation (OLR) around the disturbances. Zawislak (2020) used Tropical Rainfall Measuring Mission (TRMM)-derived precipitation to analyze the difference between developing and nondeveloping disturbances during 1998–2015. Although these studies have been devoted to identifying and comparing the characteristics of developing and nondeveloping disturbances, their results are largely sensitive to the choice of tracking methodology and corresponding thresholds. Moreover, the feature of climate variability of tropical disturbances and its relationship with tropical cyclogenesis remain unclear.
This study attempts to characterize developing and nondeveloping disturbances in the WNP for the period 1965–2020 and to describe and diagnose the effects of variability of tropical disturbances on tropical cyclogenesis in the WNP. The objective is to advance our diagnosis of the developing disturbances that contribute to the climate variability of tropical cyclogenesis in the WNP. The data and the identification of developing and nondeveloping disturbances are described in Sect. 2. Section 3 presents the climatological characteristics of developing and nondeveloping disturbances. Section 4 describes the interannual and interdecadal variability of disturbances, followed by the summary in Sect. 5.
2 Data and methodology
2.1 Data
We use the Joint Typhoon Warning Center (JTWC) best track dataset from 1965 to 2020 to describe the TC location and intensity (available online at http://www.metoc.navy.mil/jtwc/). We only consider the TCs that reach tropical storm (TS) intensity with the maximum sustained surface wind speed greater than 17 m s−1. This study defines the TC genesis position as the corresponding location when a TC first appears in its record. Our study area is confined from the equator to 30°N and from 120°E to 180°. We do not examine the disturbances in the South China Sea (SCS) because the characteristics of SCS TCs are different from those over the WNP (Wu et al. 2020). Besides, we only investigate the tropical disturbances from June to November since these 6 months are the WNP TC active periods.
The tropical disturbances are examined by the fifth generation European Center for Medium Weather Forecasting (ECMWF) reanalysis data (ERA5; Hersbach et al. 2020; https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era5/), at 6-hourly interval (at 0000, 0600, 1200, and 1800 UTC) with a horizontal resolution of 0.25° × 0.25° from 1965 to 2020. 850-hPa wind data are used to extract the vortex of the tropical disturbances. The National Oceanic and Atmospheric Administration (NOAA) Climate Data Record (CDR) of daily OLR data during 1979–2020 are used to represent the convection (Lee et al. 2007; https://www.ncei.noaa.gov/products/climate-data-records/outgoing-longwave-radiation-daily/). Since there were no satellite OLR data before 1979, the mean top net longwave radiation data in ERA5 during 1965–1978 are adopted. Note that the difference between the two kinds of OLR datasets is slight.
In addition, the Pacific Decadal Oscillation (PDO), Interdecadal Pacific Oscillation (IPO, available online at https://psl.noaa.gov/data/climateindices/list/), and El Niño Modoki index (EMI, Ashok et al. 2007) are used in this paper. The Lanczos filtering technique (Duchon 1979) is applied to obtain the synoptic-scale signals of the 3–8-day band. The Student’s t-test is applied to estimate the statistical significance.
2.2 Identifying and tracking developing and nondeveloping disturbances
Previous studies (e.g., Wang et al. 2012; Peng et al. 2012; Raavi and Walsh 2020) have mentioned that the low-level (e.g., 850-hPa and 700-hPa) relative vorticity is one of the most critical variables that can catch the disturbances. As is known to us, tropical disturbances are often accompanied by convective activities and vortex coupling. As Zawislak (2020) illustrated, the vorticity maxima are limited in separating the developing and nondeveloping disturbances. Convection near the disturbance center is favorable for predicting tropical cyclogenesis. In this study, we used the 6-hourly 850-hPa wind and daily OLR data to identify and track tropical disturbances in the WNP with the following definitions given:
-
1)
We used a 4.5° × 4.5° square box to represent the disturbance cyclonic circulation region. There must be a relative vorticity local maximum at 850-hPa in the center of the box.
-
2)
There must be a 3–8-day band-pass filtered relative vorticity local maximum averaged in a quarter of the grid points within the box exceeding 0.6 × 10–5 s−1 at 850-hPa.
-
3)
The local minimum in 3–8-day band-pass filtered OLR averaged in a quarter of the grid points within the box must be less than − 5 W m−2, and the minimum OLR value in the box must be less than − 12 W m−2.
-
4)
The distance between the center of the disturbance detected according to the above-mentioned steps and its next time (6 h) center must not exceed 4.5° latitude and longitude. If the tropical disturbance can eventually develop into a TC with a TC record in the JTWC best-track data, the disturbance is considered as a developing disturbance and its trajectory is correspondingly established. If a disturbance lasts more than 48 h but cannot further develop into TCs, it is defined as a nondeveloping disturbance. The reliability of the methodology is tested and verified in Sect. 4.3.
Figure 1 presents two examples of tropical disturbances in 2016 to illustrate the skill of the method in identifying and tracking the developing and nondeveloping disturbances. The developing disturbance case formed at 0000UTC on 10 October, with an increase in the vorticity and the convection occurring in the center (Fig. 1a, b). Such TD-type structure developed and moved west-northwestward, forming a tropical cyclone in three days later at 0600UTC on 13 October (Fig. 1c–e). The nondeveloping disturbance is similar to the developing disturbance at the early stage of disturbance formation (0000UTC on 6 September). Although the vorticity and convection increased in the early stage, its vorticity was smaller at all times compared to the developing disturbance (Fig. 1g–i). Before the day of dissipation, the nondeveloping disturbance had almost no closed circulation (Fig. 1j, k). After dissipation, the convection around the disturbance also gradually disappeared (Fig. 1l).
To further assess the method skill for identifying disturbances, we examined the characteristics of tropical disturbances each year. Figure 2a and b show the tracks of developing and nondeveloping disturbances in 2016 as an example. The number of developing disturbances identified in 2016 is 21, accounting for 87.5% (14.6%) of the TCs (total disturbances) in that year. Also, the developing cases tend to occur between 140°E and 170°E and further move northward in the WNP compared to nondeveloping cases in 2016 (Fig. 2a, b).
3 Climatological characteristics of developing and nondeveloping disturbances
6496 tropical disturbances from June to November during 1965–2020 were identified based on the above-proposed algorithm. The average numbers of developing and nondeveloping disturbances per year are 15 and 101, respectively. Only a small portion (13.1%) of tropical disturbances can develop into TCs, which is similar to the results over the North Atlantic (Frank 1970; Fu et al. 2012). Approximately 20% of the tropical disturbances can develop into TCs over the eastern Pacific and Atlantic. Additionally, 80.9% of TCs form within the developing disturbances, similar to the results from Wu and Takahashi (2018).
Figure 2c and d show the climatological distributions of the occurrence frequencies of annual developing and nondeveloping disturbances in the WNP for 1965–2020. The climatological mean position of the developing disturbances is around 130°–160°E, 5°–20°N, and the occurrence frequency maximum center is slightly southeastward compared with that of the TC genesis in the WNP. The spatial distribution of the nondeveloping disturbances is centered within 150°–175°E, 2°–10°N with some nondeveloping disturbances along the east coast of the Philippines, which is more eastward and southward than that of the developing disturbances. The high values and features of the standard deviations of disturbances are similar to their climatological mean, indicating that the disturbance activity in the WNP might exhibit strong variability.
Figure 3a and b show the probability density and cumulative probability distribution for the percentage of the lifetimes of developing and nondeveloping disturbances. The average lifetimes of developing and nondeveloping disturbances are 2.3 and 3.4 days, respectively. In general, the developing disturbances can last no more than 11.5 days, and more than 70% of developing disturbances can maintain within 3 days. For nondeveloping disturbances, the lifetime of disturbances is similar to that of developing cases, with a lifetime of shorter than 3.5 days that can explain almost 70% of nondeveloping disturbances (Fig. 3b). Generally, the lifetimes of major tropical disturbances are no more than 3 days. This average lifetime of developing disturbances is similar to the finding of Zehr (1992), which is 2.1 days. However, this is inconsistent with the result in Hennon et al. (2013), which shows that the developing disturbances have a shorter lifetime, and about 60% of developing disturbances maintain within 24 h.
The average phase speeds of developing (nondeveloping) disturbances derived from vortex are compared in Fig. 3c to distinguish their features within 72 h prior to TC genesis (disturbance dissipation). The phase velocity of developing (nondeveloping) disturbances are similar in the different stages, the average propagation speed is 5.4 m s−1 (5.2 m s−1), which is consistent with the previous results of Fu et al. (2012). Around 24 h before dissipation, the speed of nondeveloping disturbances slightly decreases.
The activities of developing and nondeveloping disturbances also show seasonal variability. Both of them are most active from July to October, with the peak in August and September, which is similar to the seasonal cycle of the TC number (Fig. 4a). The percentage of developing disturbances to all tropical disturbances is large during July–October, and 14.8% of tropical disturbances can finally turn into TCs (Fig. 4b). About 3.5 disturbances can develop into TCs in August and September each year, accounting for 22.8% and 22.7% of the annual TC genesis number, respectively. This coherent seasonal variability suggests that the similar large-scale environment related to TC genesis may also contribute to the development of the developing and nondeveloping disturbances.
4 Interannual and interdecadal variability
4.1 Interannual variability
Figure 5a shows the annual numbers of TCs and disturbances in the WNP from 1965 to 2020. There are significant interannual and interdecadal variations in both developing and nondeveloping disturbances. The standard deviations of interannual variation (deviation from 5-year running mean) of developing and nondeveloping disturbances are 1.5 and 7.1, respectively. Their corresponding standard deviations of interdecadal variation (5-year running mean) are 2.0 and 5.9, respectively. The interannual variation of developing disturbances is closely related to that of the TC genesis number (Fig. 5c), with a correlation coefficient of 0.7, exceeding the 95% confidence level. This is likely because the TC formation in the WNP is mainly attributed to developing disturbances with a contribution rate of 80.9% (Fig. 5b). Although there were significant interdecadal variations in the rate of TCs associated with disturbances, the 15-year running correlation between developing disturbances and the TC genesis number is consistently significant positive, with almost all correlation coefficients exceeding the 90% confidence level in different running periods (Fig. 5c). The interannual variability of the annual number of developing disturbances shows inconsistency with that of nondeveloping disturbances (Fig. 5a, b). The interannual correlation coefficient between developing and nondeveloping disturbances during 1965–2020 is merely − 0.19, and their 15-year running correlation is insignificant in almost all sliding periods (Fig. 5c).
Previous studies found that the interannual variability of TC genesis frequency in the WNP is generally associated with several modes of tropical climate variability (e.g., Chen and Tam 2010; Jin et al. 2013; Zhan et al. 2017; Gao et al. 2018; Yong and Chen 2019). For instance, El Niño-Southern Oscillation (ENSO) can regulate the location of TC genesis over the WNP, and the Central-Pacific El Niño (CP-El Niño) events can lead to more TCs moving toward East Asia (Jin et al. 2013). The TC frequency positively and significantly correlates with EMI (Chen and Tam 2010). Additionally, in the positive phase of the Pacific Meridional Mode (PMM), the warm SSTA in the subtropical eastern Pacific can influence the TC activity in the WNP through the Gill response (Zhan et al. 2017; Gao et al. 2018). The reduction of TC genesis frequency over the WNP after the late 1990s is significantly correlated to the PDO/IPO negative phase (Zhao et al. 2018; Yong and Chen 2019). Table 1 shows the correlations of tropical disturbances and TCs with ENSO and PMM indices. The numbers of developing disturbances and TC genesis show insignificant correlations with Niño-3.4 index, while the correlation is significant between the number of nondeveloping disturbances and the Niño-3.4 index. There are significant positive correlations of the EMI and PMM indices with the number of TC genesis (developing disturbances), with the correlation coefficients of 0.44 and 0.35 (0.42 and 0.39), respectively. However, the correlation between the number of nondeveloping disturbances and the EMI (PMM) index is negative (insignificantly positive). These results indicate that the developing disturbances exhibit a similar relationship with climate variability to TC genesis, while the nondeveloping disturbances seem to be independent of them.
Such an interannual relationship also exhibits interdecadal changes. Figure 5d shows the 15-year running correlations of EMI and PMM indices with tropical disturbances and TCs. The EMI and PMM indices show significant positive correlations with TCs since the 1980s. For developing disturbances, the EMI index shows significant positive correlations since 1987. These significant changes might be influenced by the intensified impact of the central Pacific warming on the monsoon trough over the WNP since the early 1980s (Zhao et al. 2018). The eastward extension of the monsoon trough induced by the warmer ocean surface in the tropical central Pacific led to increased tropical wave activity by accelerated wave-mean flow interaction (Wu et al. 2015a, b). Although the nondeveloping disturbances had relatively weaker relationships with the EMI and PMM indices for most epochs, their moving correlation coefficients decreased around the 1980s (Fig. 5d). Such changes can be explained by the inverse interdecadal relationship between the numbers of developing and nondeveloping disturbances.
4.2 Interdecadal variability
In addition to the interannual variability mentioned above, the numbers of both tropical disturbances and TCs exhibit large interdecadal variability (Fig. 5a). Note that their interdecadal variability is tangled with long-term trends. Both developing disturbances and TCs show a slightly decreasing trend during the period 1965–2020, while the nondeveloping disturbances exhibit a significant increasing trend. This difference suggests that global warming has led to more tropical disturbances in the WNP but has not contributed to the transition from disturbances to TCs (Fig. 5b). On an interdecadal timescale, the number of developing disturbances shows a significant positive correlation with the number of TCs in the WNP after removing their linear trend, with a relative maximum in the late 1980s to early 1990s and a minimum around the late 1990s. Such variability of TCs shows sudden interdecadal changes around 1987 and 1997 (Fig. 6a). For developing disturbances, the interdecadal changes appear around 1987, 1992, and 2000 (Fig. 6a). These changes match well with the interdecadal changes of WNP TC numbers in previous studies (He et al. 2015; Liu et al. 2019). The number of nondeveloping disturbances shows an inverse interdecadal variability, and the relevant sudden interdecadal changes appear around 1978 and the early 1990s (Fig. 6a). The correlation coefficient between nondeveloping and developing disturbances is − 0.66, which exceeds the 95% confidence level. As is displayed in Fig. 6b, the percentage of developing disturbances to total TCs also varies in different decades, the reason for which will be further discussed in the next sub-section. The TC activity in the WNP is strongly influenced by the PDO and IPO on an interdecadal timescale. Table 1 shows the relationship among the tropical disturbances, TCs, and climate factors. It is indicated that there is no significant correlation between the number of region-averaged tropical disturbances in the WNP with the PDO and IPO. To obtain a sense of the spatial feature in the relationship of the disturbances with PDO and IPO, correlation charts of developing and nondeveloping disturbances with respect to the time series of the PDO and IPO are computed. The PDO and TC genesis frequency show significant positive correlations in most of the tropical basin (120°–180°E, 5°–15°N) (Fig. 7a). The positive correlation suggests that the PDO shows a significant consistent effect on TCs (Table 1). The correlation between the developing disturbances and the PDO is significantly positive (negative) in the major southeast (northwest) part of 155°E in the WNP, and shows a significant positive relationship in the tropical WNP (162°–175°E, 0°–20°N) for the nondeveloping disturbances (Fig. 7b, c). The spatial distributions of correlation coefficients of IPO with TCs and disturbances are similar to those of PDO (Fig. 7d–f). The spatial inconsistency of the influence of PDO and IPO on TCs leads to their weak interdecadal correlations with TCs.
4.3 Sensitivity analysis of the disturbance identifying and tracking algorithm
The annual percentage of developing disturbances to TCs also exhibits a significant interdecadal variation in the WNP, with similar features to the numbers of TCs and the developing disturbances (Fig. 5b). To examine the dependence of disturbance variability on identifying conditions, the percentage of developing disturbances to TCs in different decades is shown in Fig. 6b. When more (less) TCs form in the WNP during the period 1988–1997 (1998–2020), the percentage of TCs associated with disturbances is 67.6% (90.0%) and significantly below (above) normal (80.9%). In other words, more TCs tend to be accompanied by decreased disturbance-to-TC transition probability on the interdecadal timescale in the WNP. This suggests that the favorable environments for TC genesis do not imply favorable conditions for the transition of disturbances to TCs, which even might be out-phase in the interdecadal timescale.
Why is there such an interdecadal variation in the transition probability of disturbances to TCs and whether the definition of identifying disturbances causes this interdecadal variation? To answer these questions, Fig. 8 shows the composite structure and spatial distribution of disturbances (3–8-day) concerning the disturbance vorticity maximum one day prior to the TC genesis during different decades. For TC genesis that is attributed to disturbances (Fig. 8a–d), the cyclonic vortex center of disturbances is in the southeast region of the TC genesis point one day prior to genesis. Both convection and vorticity show northeast-to-southwest tilted structures and their centers are located around the vortex center for three decades. There are weaker convections and vorticity of disturbances during the period 1988–1997 than those during other periods. For TC genesis without precursory disturbances (Fig. 8e–h), a similar structure is observed but its vorticity is relatively weak and its convection is disorganized, especially during the period 1988–1997. The weak signal of the disturbances may cause the identified transition of disturbances to be below normal during the period 1988–1997 (Fig. 8c, g).
To further measure the dependence of the percentage of the variability of tropical disturbances and the percentage of developing disturbances to TCs on the choice of vorticity and convection thresholds, we computed the numbers of the developing and nondeveloping disturbances during the period 1965–2020 based on a range of reasonable thresholds of vorticity and convection. Figure 9a, b shows the percentage and number of developing disturbances as a function of the vorticity (OLR) threshold. For a high threshold (2 × 10–5 s−1 for vorticity or − 30 W m−2 for OLR), more than a third of TCs have no precursor disturbances. Both the percentage and number of developing disturbances have increased with decreasing threshold of vorticity (OLR), and they show a slower increase when the threshold of vorticity (OLR) is below (over) 1.1 × 10–5 s−1 (− 12 W m−2). The number of nondeveloping disturbances would rapidly increase when the vorticity and convection thresholds decrease (Fig. 9c, d). These results suggest that the decreased vorticity and convection thresholds favor the growth of disturbance numbers, but the percentage of developing disturbances to TCs has a slower growth rate as a function of threshold values of vorticity and convection when the threshold is below a certain value. It is interesting to note that similar changes occur during the period 1988–1997. The percentage of developing disturbances to TCs during the period 1988–1997 obviously changes in small thresholds (Fig. 9). In other words, the developing disturbance variability is not sensitive to a range of reasonable thresholds.
5 Summary
In this study, the characteristics of tropical disturbances over the WNP from June to November during 1965–2020 are investigated through the statistical analysis based on an identifying and tracking algorithm of tropical disturbances. The tropical disturbances occur in nearly all regions of the tropical WNP. Approximately 116 tropical disturbances develop in the WNP each year, and about 13.1% of them can develop into TCs. About 80.9% of TC geneses originate from tropical disturbances. The climatological mean position of developing disturbances is around 130°–160°E, 5°–20°N, and the maximum area is northwest tilted near the 148°–150°E, 10°–12°N. Also, the position of maximum occurrence frequency is slightly southeastward than that of TC genesis. Compared with developing disturbances, the spatial distribution of nondeveloping disturbances is more eastward and southward, which is centered at 150°–175°E, 2°–10°N. The average lifespans of developing and nondeveloping disturbances are 2.3 and 3.4 days, respectively. The lifetime of the majority of (70%) developing disturbances is no more than 3 days while that of nondeveloping disturbances is about 3.5 days. The average propagation speed of developing (nondeveloping) disturbances within 72 h prior to TC genesis (disturbance dissipation) is 5.4 m s−1 (5.2 m s−1). The developing and nondeveloping disturbances are more active from July to October, with a peak in August and September, similar to changes in TC numbers.
The annual numbers of developing and nondeveloping disturbances exhibit significant interannual and interdecadal variabilities. The interannual variation of the number of developing disturbances is close to that of the number of TCs but inconsistent with the total number of disturbances. The two modes of tropical climate variability CP El Niño and PMM are significantly correlated with the developing disturbances and TCs. On the interdecadal timescale, the number of developing disturbances shows a significant correlation with the number of TCs in the WNP, with a maximum in the late 1980s to early 1990s and a minimum around the late 1990s. The interdecadal variation in the occurrences of nondeveloping disturbances shows an inverse relationship with that of developing disturbances. Such interdecadal variability of disturbances exhibits an insignificantly weak correlation with the PDO and IPO. This can be mainly attributed to the opposing influences of climate indices (PDO and IPO) on disturbances between the zonal region of the WNP. In addition, the long-term trends suggest that global warming has led to more tropical disturbances in the WNP but not the case for disturbance-to-TC transition probability.
The annual percentage of TCs associated with disturbances (transition probability from disturbances to TCs) also exhibits a significant interdecadal variation in the WNP. When more (less) TCs form in the WNP during the period 1988–1997 (1998–2020), the transition probability from disturbances to TCs is significant below (above) normal. The composite structure and spatial distribution of disturbances during different decades suggest that the weak signal of the disturbance may cause the identified transition of disturbances to be below normal. However, the percentage of TCs associated with disturbances during the period 1988–1997 has no obvious change in small thresholds. It is suggested that the developing disturbance variability is not sensitive to a range of reasonable thresholds.
This study presented the main characteristics of the developing and nondeveloping disturbances in the WNP. However, the mechanisms of tropical cyclogenesis in developing disturbances and the contributions of the large-scale environment remain uncertain and need further investigation.
References
Ashok K, Behera SK, Rao SA et al (2007) El Niño Modoki and its possible teleconnection. J Geophys Res 112:C1107. https://doi.org/10.1029/2006JC003798
Brammer A, Thorncroft CD (2015) Variability and evolution of African easterly wave structures and their relationship with tropical cyclogenesis over the eastern Atlantic. Mon Weather Rev 143:4975–4995. https://doi.org/10.1175/MWR-D-15-0106.1
Camargo SJ, Emanuel KA, Sobel AH (2007) Use of a genesis potential index to diagnose ENSO effects on tropical cyclone genesis. J Clim 20:4819–4834. https://doi.org/10.1175/JCLI4282.1
Carlson TN (1969) Some remarks on African disturbances and their progress over the tropical Atlantic. Mon Weather Rev 97:716–726. https://doi.org/10.1175/1520-0493(1969)097%3c0716:sroada%3e2.3.co;2
Chen TC (2006) Characteristics of African easterly waves depicted by ECMWF reanalyses for 1991–2000. Mon Weather Rev 134:3539–3566. https://doi.org/10.1175/MWR3259.1
Chen G, Tam CY (2010) Different impacts of two kinds of Pacific Ocean warming on tropical cyclone frequency over the western North Pacific. Geophys Res Lett 37:1–6. https://doi.org/10.1029/2009GL041708
Chen TC, Wang SY, Yen MC, Clark AJ (2008) Are tropical cyclones less effectively formed by easterly waves in the western North Pacific than in the North Atlantic? Mon Weather Rev 136:4527–4540. https://doi.org/10.1175/2008MWR2149.1
Duchon CE (1979) Lanczos filtering in one and two dimensions. J Appl Meteorol 18:1016–1022
Emanuel K, Sundararajan R, Williams J (2008) Hurricanes and global warming: Results from downscaling IPCC AR4 simulations. Bull Am Meteorol Soc 89:347–367. https://doi.org/10.1175/BAMS-89-3-347
Feng T, Yang XQ, Jia-Yuh YU, Huang R (2020a) Convective coupling in tropical-depression-type waves. Part I: Rainfall characteristics and moisture structure. J Atmos Sci 77:3407–3422. https://doi.org/10.1175/JAS-D-19-0172.1
Feng T, Yang XQ, Wu L, Huang R, Yang D (2020b) How do the monsoon trough and the tropical upper-tropospheric trough affect synoptic-scale waves: A comparative study. J Meteorol Soc Japan 98:735–754. https://doi.org/10.2151/jmsj.2020-037
Frank NL (1970) Atlantic tropical systems of 1969. Mon Weather Rev 98:307–314. https://doi.org/10.1175/1520-0493(1994)122%3c0539:ATSO%3e2.0.CO;2
Frank WM, Roundy PE (2006) The role of tropical waves in tropical cyclogenesis. Mon Weather Rev 134:2397–2417. https://doi.org/10.1175/MWR3204.1
Fu B, Peng MS, Li T, Stevens DE (2012) Developing versus nondeveloping disturbances for tropical cyclone formation. Part II: western North Pacific. Mon Weather Rev 140:1067–1080. https://doi.org/10.1175/2011MWR3618.1
Gao S, Zhu L, Zhang W, Chen Z (2018) Strong modulation of the Pacific Meridional Mode on the occurrence of intense tropical cyclones over the western North Pacific. J Clim 31:7739–7749. https://doi.org/10.1175/JCLI-D-17-0833.1
Gray WM (1968) Global view of the origin of tropical disturbances and storms. Mon Weather Rev 96:669–700. https://doi.org/10.1175/1520-0493(1968)096%3c0669:gvotoo%3e2.0.co;2
Gray WM (1998) The formation of tropical cyclones. Meteorol Atmos Phys 67:37–69. https://doi.org/10.1007/BF01277501
He H, Yang J, Gong D et al (2015) Decadal changes in tropical cyclone activity over the western North Pacific in the late 1990s. Clim Dyn 45:3317–3329. https://doi.org/10.1007/s00382-015-2541-1
Hennon CC, Papin PP, Zarzar CM et al (2013) Tropical cloud cluster climatology, variability, and genesis productivity. J Clim 26:3046–3066. https://doi.org/10.1175/JCLI-D-12-00387.1
Hersbach H, Bell B, Berrisford P et al (2020) The ERA5 global reanalysis. Q J R Meteorol Soc 146:1999–2049. https://doi.org/10.1002/qj.3803
Jin CS, Ho CH, Kim JH et al (2013) Critical role of northern off-equatorial sea surface temperature forcing associated with central Pacific El Niño in more frequent tropical cyclone movements toward East Asia. J Clim 26:2534–2545. https://doi.org/10.1175/JCLI-D-12-00287.1
Kerns BW, Chen SS (2013) Cloud clusters and tropical cyclogenesis: Developing and nondeveloping systems and their large-scale environment. Mon Weather Rev 141:192–210. https://doi.org/10.1175/MWR-D-11-00239.1
Kerns B, Zipser E (2009) Four years of tropical ERA-40 vorticity maxima tracks. Part II: Differences between developing and nondeveloping disturbances. Mon Weather Rev 137:2576–2591. https://doi.org/10.1175/2008MWR2545.1
Knutson TR, McBride JL, Chan J et al (2010) Tropical cyclones and climate change. Nat Geosci 3:157–163. https://doi.org/10.1038/ngeo779
Lee CS (1989) Observational analysis of tropical cyclogenesis in the western North Pacific. Part I: Structural evolution of cloud clusters. J Atmos Sci 46:2580–2598
Lee HT, Gruber A, Ellingson RG, Laszlo I (2007) Development of the HIRS outgoing longwave radiation climate dataset. J Atmos Ocean Technol 24:2029–2047. https://doi.org/10.1175/2007JTECHA989.1
Liu KS, Chan JCL (2008) Interdecadal variability of western North Pacific tropical cyclone tracks. J Clim 21:4464–4476. https://doi.org/10.1175/2008JCLI2207.1
Liu C, Zhang W, Geng X et al (2019) Modulation of tropical cyclones in the southeastern part of western North Pacific by tropical Pacific decadal variability. Clim Dyn 53:4475–4488. https://doi.org/10.1007/s00382-019-04799-w
McBride JL, Raymond Z (1981) Observational analysis of tropical cyclone formation. Part II: Comparison of non-developing versus developing systems. J Atmos Sci 38:1132–1151
Mei W, Xie SP (2016) Intensification of landfalling typhoons over the northwest Pacific since the late 1970s. Nat Geosci 9:753–757. https://doi.org/10.1038/ngeo2792
Murakami H, Li T, Peng M (2013) Changes to environmental parameters that control tropical cyclone genesis under global warming. Geophys Res Lett 40:2265–2270. https://doi.org/10.1002/grl.50393
Norquist DC, Recker EE, Reed RJ (1977) The energetics of African wave disturbances as observed during Phase III of GATE. Mon Weather Rev 105:334–342
Peng MS, Fu B, Li T, Stevens DE (2012) Developing versus nondeveloping disturbances for tropical cyclone formation. Part I: North Atlantic. Mon Weather Rev 140:1047–1066. https://doi.org/10.1175/2011MWR3617.1
Raavi PH, Walsh KJE (2020) Basinwise statistical analysis of factors limiting tropical storm formation from an initial tropical circulation. J Geophys Res Atmos 125:1–20. https://doi.org/10.1029/2019JD032006
Ross RS, Krishnamurti TN, Pattnaik S, Simon A (2009) Energy transformation and diabatic processes in developing and nondeveloping African easterly waves observed during the NAMMA project of 2006. Weather Forecast 24:1524–1548. https://doi.org/10.1175/2009WAF2222235.1
Ross RS, Krishnamurti TN, Pattnaik S (2012) Interactions of diabatic heating in convective superbursts with energy conversion processes in the genesis of Cape Verde hurricanes from African easterly waves. Mon Weather Rev 140:748–773. https://doi.org/10.1175/2011MWR3621.1
Schreck CJ, Molinari J, Aiyyer A (2012) A global view of equatorial waves and tropical cyclogenesis. Mon Weather Rev 140:774–788. https://doi.org/10.1175/MWR-D-11-00110.1
Sharmila S, Walsh KJE (2017) Impact of large-scale dynamic versus thermodynamic climate conditions on contrasting tropical cyclone genesis frequency. J Clim 30:8865–8883. https://doi.org/10.1175/JCLI-D-16-0900.1
Thorncroft C, Hodges K (2001) African easterly wave variability and its relationship to Atlantic tropical cyclone activity. J Clim 14:1166–1179. https://doi.org/10.1175/1520-0442(2001)014%3c1166:AEWVAI%3e2.0.CO;2
Thorncroft CD, Rowell DP (1998) Interannual variability of African wave activity in a general circulation model. Int J Climatol 18:1305–1323. https://doi.org/10.1002/(SICI)1097-0088(1998100)18:12%3c1305::AID-JOC281%3e3.0.CO;2-N
Wang Z, Hankes I (2014) Characteristics of tropical easterly wave pouches during tropical cyclone formation. Mon Weather Rev 142:626–633. https://doi.org/10.1175/MWR-D-13-00267.1
Wang Z, Montgomery MT, Fritz C (2012) A first look at the structure of the wave pouch during the 2009 PREDICT-GRIP dry runs over the Atlantic. Mon Weather Rev 140:1144–1163. https://doi.org/10.1175/MWR-D-10-05063.1
Webster PJ, Holland GJ, Curry JA, Chang HR (2005) Changes in tropical cyclone number, duration, and intensity in a warming environment. Science 309:1844–1846
Wu L, Takahashi M (2018) Contributions of tropical waves to tropical cyclone genesis over the western North Pacific. Clim Dyn 50:4635–4649. https://doi.org/10.1007/s00382-017-3895-3
Wu L, Wen Z, Li T, Huang R (2014) ENSO-phase dependent TD and MRG wave activity in the western North Pacific. Clim Dyn 42:1217–1227. https://doi.org/10.1007/s00382-013-1754-4
Wu L, Wen Z, Wu R (2015a) Influence of the monsoon trough on westward-propagating tropical waves over the western North Pacific. Part I: Observations. J Clim 28:7108–7127. https://doi.org/10.1175/JCLI-D-14-00806.1
Wu L, Wen Z, Wu R (2015b) Influence of the monsoon trough on westward-propagating tropical waves over the western North Pacific. Part II: Energetics and numerical experiments. J Clim 28:9332–9349. https://doi.org/10.1175/JCLI-D-14-00807.1
Wu L, Zhang H, Chen J-M, Feng T (2020) Characteristics of tropical cyclone activity over the South China Sea: Local and nonlocal tropical cyclones. Terr Atmos Ocean Sci 31:261–271. https://doi.org/10.3319/TAO.2019.07.01.02
Yong L, Chen D (2019) An interdecadal change in the interannual variability of boreal summer tropical cyclone genesis frequency over the western North Pacific around the early 1990s. Theor Appl Climatol 137:1843–1853. https://doi.org/10.1007/s00704-018-2710-3
Zawislak J (2020) Global survey of precipitation properties observed during tropical cyclogenesis and their differences compared to nondeveloping disturbances. Mon Weather Rev 148:1585–1606. https://doi.org/10.1175/MWR-D-18-0407.1
Zehr RM (1992) Tropical cyclogenesis in the western North Pacific. NOAA Tech Rep 61:181
Zhan R, Wang Y, Liu Q (2017) Salient differences in tropical cyclone activity over the western North Pacific between 1998 and 2016. J Clim 30:9979–9997. https://doi.org/10.1175/JCLI-D-17-0263.1
Zhao J, Zhan R, Wang Y, Xu H (2018) Contribution of the Interdecadal Pacific Oscillation to the recent abrupt decrease in tropical cyclone genesis frequency over the western North Pacific since 1998. J Clim 31:8211–8224. https://doi.org/10.1175/JCLI-D-18-0202.1
Acknowledgements
This work is jointly supported by the National Natural Science Foundation of China Grants 41875117, 42175050 and 42075015, the Shanghai Typhoon Research Foundation Grant TFJJ202002, and the Youth Innovation Promotion Association CAS Y2021030.
Author information
Authors and Affiliations
Contributions
L.W. and Z.W. designed the research. Y.G. performed analysis and wrote the drafts. L.W. conceived the ideas. L.W., R. Z., and Z.W. discussed the results and refined and improved the manuscript. All authors read and approved the final manuscript.
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This manuscript is submitted to “Terrestrial, Atmospheric and Oceanic Sciences”.
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, visithttp://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Gu, Y., Wu, L., Zhan, R. et al. Climatology of developing and nondeveloping disturbances for tropical cyclone genesis over the western North Pacific. TAO 33, 13 (2022). https://doi.org/10.1007/s44195-022-00012-4
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s44195-022-00012-4