2021 South Asia monsoon outlook (Technical edition)
By
Jason Nicholls, AccuWeather senior meteorologist
Updated Apr 21, 2021 2:19 PM EDT
The Southwest monsoon in India runs from June through September and rainfall during the Southwest monsoon accounts for about 70% of India’s total annual rainfall. The monsoon is also an integral part of India’s economy since farming makes up 15% of India’s GDP and employs more than half of the population. In addition to rainfall determining yields for crops such as rice, wheat, sugarcane and oil seeds, it also helps replenish reservoirs and ground water for irrigation, as well as boosts hydropower production.
So what does the 2021 southwest monsoon season hold in store for India? First we will take a quick look at the revised onset and withdrawal dates from the India Meteorological Department. The normal onset date in Kerala is still around 1 June, but the monsoon is now set to clear all of India nearly a week earlier. The biggest change is the withdrawal starting over two weeks later under the revised dates.
Now that we briefly looked at climatology, we will dive into some main teleconnections to look at regarding the monsoon. The first one is ENSO. As a general rule, El Nino events tend to favor weak monsoon years, while La Nina tends to favor strong monsoon seasons. The recent La Nina peaked over the past winter and has since trended to neutral, but there is a lingering cool pool near the central equatorial Pacific leading to a La Nina Modoki or weak CP-based La Nina signature with water temperature anomalies. This weak central Pacific La Nina is expected to prevail into the start of the monsoon with the possibility to carry through much of the season.
A classic eastern-based La Nina and central-based La Nina do result in somewhat different patterns globally due to the anomalous double Walker circulation resulting from a La Nina Modoki. Below are precipitation and temperature anomalies for the JJA period associated with a classic eastern La Nina and a La Nina Modoki or central Pacific-based La Nina. The biggest difference of note across India is that a La Nina Modoki is not nearly as wet as a classic La Nina.
In addition to ENSO during the monsoon season, the phase of ENSO during the preceding winter can influence rainfall across India during the summer. Some excerpts from a paper by Arindam Chakraborty (below) point out how the winter phase of ENSO can influence summer rainfall in India. Namely, a winter La Nina to a neutral ENSO summer can reduce rainfall by 4% across India with the impact most pronounced in southern and northwest India.
Taking a look at teleconnections closer to India, the big one is the Indian Ocean Dipole (IOD) and the atmospheric counterpart called EQUINOO. Here is just a brief refresher before looking at the current and forecast states of both the IOD and EQUINOO.
Currently the IOD and EQUINOO are both neutral, but several climate models indicate that both can trend toward negative during summer before typically peaking in autumn. A moderate to strong negative IOD and EQUINOO could mean a reduction in rainfall across India, especially later in the season. However, at this point we are leaning toward the possibility of a weak to perhaps moderate negative IOD.
Some other Indian Ocean-based sea-surface temperature-based teleconnections are shown below with the regions of importance highlighted with black squares. The warm anomalies off northwest Australia this past DJF should favor good late-season rainfall in India based on other papers. The warm anomalies in the central tropical Indian Ocean this past SON are another favorable sign for rainfall in India. However, the cool anomalies in the Arabian Sea this past winter are worrisome, especially over northwest India. These teleconnections seem to work best when all three are in the same phase.
Farther east, sea-surface temperature anomalies in the northern Pacific and Atlantic have also been shown to have some influence on India’s summer rainfall. The two main teleconnections we are looking at here are the Pacific Decadal Oscillation (PDO) and the Atlantic multi-decadal Oscillation (AMO). Sometimes we can look to the Atlantic tripole, but there is no evidence of a tripole this year. The following image is just a brief refresher on the PDO and AMO.
Currently there is a negative PDO which is forecast to trend toward neutral over the next few months, but it may stay weakly negative. A negative PDO tends to favor above-normal rainfall during the monsoon with the effect amplified when in phase with ENSO or when there is a La Nina event. Both the PDO and ENSO signals may be weak this summer but a weak PDO and weak La Nina Modoki could work together to support a good monsoon. Meanwhile, the AMO is weak, warm and shown to become warmer over the coming month. A warm AMO is favorable for monsoonal rains across India, so it is another positive signal for the coming summer.
The last couple of teleconnections we will look at are Eurasian Snowcover and the Quasi-biennial Oscillation (QBO). The level of the QBO to look at is 30 mb, which is currently weakly westerly or positive and is forecast to become easterly or negative during the summer. A strong negative QBO at 30 mb tends to result in a weak monsoon but if the easterlies reach 30 mb this summer, they may be fairly weak so this teleconnection may have little impact on the monsoon pattern. Meanwhile, Eurasian snow cover during this past winter into spring has been largely below normal, which papers have shown tends to favor above-normal rainfall in India during the southwest monsoon.
All these teleconnections in mind and looking at several other factors we compiled several analog years for the monsoon season. The following image shows the composite rainfall for the top analog years (top) and all analog years (bottom). The overall pattern for both is not too dissimilar. However, there is a more pronounced dry anomaly in northwest India and more of a wet anomaly in far southern India for our top analog years.
A quick look at several climate models are below. Most show a wet monsoon across much of India, but they may be a little overdone in some areas. The biggest issue with some of the modeling is that they show above-normal rainfall across nearly all of India which is rare even in flood years. Climatology would suggest that parts of India will likely not be as wet as the modeling suggests.
A mix of climate modeling, teleconnections and analogs has led us to our outlook for the 2021 monsoon season. Overall, for India and Sri Lanka we are going with rainfall 100-105% of normal with an error of + or - 5%. The rainfall forecast for Pakistan is a bit more pessimistic with rainfall expected to be 95-100% of normal for the nation as a whole with the same margin of +/- 5%. The driest parts of Pakistan should tend to be across central districts.
Some highlights of the forecast include…
1. Onset in Kerala is around the normal onset date of 1 June with a margin of +/- 4 days with onset in Sri Lanka by the end of May.
2. Monsoon advance should be relatively steady with a few usual surges and stalls, but it should clear most of India and enter Pakistan by mid-July.
3. Rainfall during June should be near to above normal in many areas, but it can be drier than normal in northwest India.
4. Areas of dryness can develop in Kerala, Karnataka and parts of northeast India in July or early August. We are also concerned for dryness becoming an issue in portions of northwest India, Tamil Nadu, Bangladesh and Sri Lanka late in the season.
Possible risks for the monsoon forecast are as follows…
1. A negative IOD developing late in the season as expected should result in dryness in northwest India as forecast. However, if the IOD remains neutral, then some teleconnections suggest a wetter pattern in northwest India, especially late in the season. This could result in rainfall closer to normal or perhaps even above normal in northwest India. The ECMWF model even hints at a weakly positive IOD which could also mean wetter than forecast conditions in northwest India.
2. If ENSO trends back toward a classic La Nina in late summer/autumn then rainfall across India could be greater than forecast. La Nina to neutral usually results in rainfall reduced by 4% across India, but back to back La Ninas tend to favor above normal rainfall in many areas. The exception is toward the southern tip of peninsular India.
3. More lows forming in the Bay of Bengal this year versus 2020 could result in better rainfall across northeast India and Bangladesh than currently forecast.
4. Climate modeling suggests a wetter outlook than forecast, which is a risk.
Lastly, a quick look at the tropical northern Indian Ocean which can be a wildcard for the monsoon. We are currently seeing signals that suggest that the Bay of Bengal should have an increase in the number of lows and depressions pre-monsoon and during the monsoon as compared to 2020 when the basin was fairly quiet until late summer and autumn. We are expecting two to three depressions from May through September versus only one tropical system in May 2020. There is also the risk for a couple of cyclonic storms to develop in the northern Indian Ocean during May and June, with a leaning toward the Bay of Bengal for such activity. However, the Arabian Sea can still have tropical activity pre-monsoon, but it is expected to be less active than 2020 with zero or one tropical feature in May and June compared to two systems in 2020.
Some highlights of the forecast include…
1. Onset in Kerala around the normal onset date of 1 June with a margin of +/- 4 days with onset in Sri Lanka by the end of May.
2. Monsoon advance should be relatively steady with a few usual surges and stalls, but it should clear most of India and enter Pakistan by mid-July.
3. Rainfall during June should be near to above normal in many areas, but it can be drier than normal in northwest India.
4. Areas of dryness can develop in Kerala, Karnataka and parts of northeast India in July or early August. We are also concerned for dryness becoming an issue in portions of northwest India, Tamil Nadu, Bangladesh and Sri Lanka late in the season.
Possible risks for the monsoon forecast are as follows…
1. A negative IOD developing late in the season as expected should result in dryness in northwest India as forecast. However, if the IOD remains neutral, then some teleconnections suggest a wetter pattern in northwest India, especially late in the season. This could result in rainfall closer to normal or perhaps even above normal in northwest India. The ECMWF model even hints at a weakly positive IOD which could also mean wetter than forecast conditions in northwest India.
2. If ENSO trends back toward a classic La Nina in late summer/autumn then rainfall across India could be greater than forecast. La Nina to neutral usually results in rainfall reduced by 4% across India, but back to back La Ninas tend to favor above-normal rainfall in many areas. The exception is toward the southern tip of peninsular India.
3. More lows forming in the Bay of Bengal this year versus 2020 could result in better rainfall across northeast India and Bangladesh than currently forecast.
4. Climate modeling suggests a wetter outlook than forecast which is a risk.
Lastly, a quick look at the tropical northern Indian Ocean which can be a wildcard for the monsoon. We are currently seeing signals that suggest that the Bay of Bengal should have an increase in the number of lows and depressions pre-monsoon and during the monsoon as compared to 2020 when the basin was fairly quiet until late summer and autumn. We are expecting two to three depressions from May through September versus only one tropical system in May 2020. There is also the risk for a couple of cyclonic storms to develop in the northern Indian Ocean during May and June, with a leaning toward the Bay of Bengal for such activity. However, the Arabian Sea can still have tropical activity pre-monsoon, but is expected to be less active than 2020 with zero or one tropical feature in May and June compared to two system in 2020.
Report a Typo
Weather Blogs / Global weather
2021 South Asia monsoon outlook (Technical edition)
By Jason Nicholls, AccuWeather senior meteorologist
Updated Apr 21, 2021 2:19 PM EDT
The Southwest monsoon in India runs from June through September and rainfall during the Southwest monsoon accounts for about 70% of India’s total annual rainfall. The monsoon is also an integral part of India’s economy since farming makes up 15% of India’s GDP and employs more than half of the population. In addition to rainfall determining yields for crops such as rice, wheat, sugarcane and oil seeds, it also helps replenish reservoirs and ground water for irrigation, as well as boosts hydropower production.
So what does the 2021 southwest monsoon season hold in store for India? First we will take a quick look at the revised onset and withdrawal dates from the India Meteorological Department. The normal onset date in Kerala is still around 1 June, but the monsoon is now set to clear all of India nearly a week earlier. The biggest change is the withdrawal starting over two weeks later under the revised dates.
Now that we briefly looked at climatology, we will dive into some main teleconnections to look at regarding the monsoon. The first one is ENSO. As a general rule, El Nino events tend to favor weak monsoon years, while La Nina tends to favor strong monsoon seasons. The recent La Nina peaked over the past winter and has since trended to neutral, but there is a lingering cool pool near the central equatorial Pacific leading to a La Nina Modoki or weak CP-based La Nina signature with water temperature anomalies. This weak central Pacific La Nina is expected to prevail into the start of the monsoon with the possibility to carry through much of the season.
A classic eastern-based La Nina and central-based La Nina do result in somewhat different patterns globally due to the anomalous double Walker circulation resulting from a La Nina Modoki. Below are precipitation and temperature anomalies for the JJA period associated with a classic eastern La Nina and a La Nina Modoki or central Pacific-based La Nina. The biggest difference of note across India is that a La Nina Modoki is not nearly as wet as a classic La Nina.
In addition to ENSO during the monsoon season, the phase of ENSO during the preceding winter can influence rainfall across India during the summer. Some excerpts from a paper by Arindam Chakraborty (below) point out how the winter phase of ENSO can influence summer rainfall in India. Namely, a winter La Nina to a neutral ENSO summer can reduce rainfall by 4% across India with the impact most pronounced in southern and northwest India.
Taking a look at teleconnections closer to India, the big one is the Indian Ocean Dipole (IOD) and the atmospheric counterpart called EQUINOO. Here is just a brief refresher before looking at the current and forecast states of both the IOD and EQUINOO.
Currently the IOD and EQUINOO are both neutral, but several climate models indicate that both can trend toward negative during summer before typically peaking in autumn. A moderate to strong negative IOD and EQUINOO could mean a reduction in rainfall across India, especially later in the season. However, at this point we are leaning toward the possibility of a weak to perhaps moderate negative IOD.
Some other Indian Ocean-based sea-surface temperature-based teleconnections are shown below with the regions of importance highlighted with black squares. The warm anomalies off northwest Australia this past DJF should favor good late-season rainfall in India based on other papers. The warm anomalies in the central tropical Indian Ocean this past SON are another favorable sign for rainfall in India. However, the cool anomalies in the Arabian Sea this past winter are worrisome, especially over northwest India. These teleconnections seem to work best when all three are in the same phase.
Farther east, sea-surface temperature anomalies in the northern Pacific and Atlantic have also been shown to have some influence on India’s summer rainfall. The two main teleconnections we are looking at here are the Pacific Decadal Oscillation (PDO) and the Atlantic multi-decadal Oscillation (AMO). Sometimes we can look to the Atlantic tripole, but there is no evidence of a tripole this year. The following image is just a brief refresher on the PDO and AMO.
Currently there is a negative PDO which is forecast to trend toward neutral over the next few months, but it may stay weakly negative. A negative PDO tends to favor above-normal rainfall during the monsoon with the effect amplified when in phase with ENSO or when there is a La Nina event. Both the PDO and ENSO signals may be weak this summer but a weak PDO and weak La Nina Modoki could work together to support a good monsoon. Meanwhile, the AMO is weak, warm and shown to become warmer over the coming month. A warm AMO is favorable for monsoonal rains across India, so it is another positive signal for the coming summer.
The last couple of teleconnections we will look at are Eurasian Snowcover and the Quasi-biennial Oscillation (QBO). The level of the QBO to look at is 30 mb, which is currently weakly westerly or positive and is forecast to become easterly or negative during the summer. A strong negative QBO at 30 mb tends to result in a weak monsoon but if the easterlies reach 30 mb this summer, they may be fairly weak so this teleconnection may have little impact on the monsoon pattern. Meanwhile, Eurasian snow cover during this past winter into spring has been largely below normal, which papers have shown tends to favor above-normal rainfall in India during the southwest monsoon.
All these teleconnections in mind and looking at several other factors we compiled several analog years for the monsoon season. The following image shows the composite rainfall for the top analog years (top) and all analog years (bottom). The overall pattern for both is not too dissimilar. However, there is a more pronounced dry anomaly in northwest India and more of a wet anomaly in far southern India for our top analog years.
A quick look at several climate models are below. Most show a wet monsoon across much of India, but they may be a little overdone in some areas. The biggest issue with some of the modeling is that they show above-normal rainfall across nearly all of India which is rare even in flood years. Climatology would suggest that parts of India will likely not be as wet as the modeling suggests.
A mix of climate modeling, teleconnections and analogs has led us to our outlook for the 2021 monsoon season. Overall, for India and Sri Lanka we are going with rainfall 100-105% of normal with an error of + or - 5%. The rainfall forecast for Pakistan is a bit more pessimistic with rainfall expected to be 95-100% of normal for the nation as a whole with the same margin of +/- 5%. The driest parts of Pakistan should tend to be across central districts.
Some highlights of the forecast include…
1. Onset in Kerala is around the normal onset date of 1 June with a margin of +/- 4 days with onset in Sri Lanka by the end of May.
2. Monsoon advance should be relatively steady with a few usual surges and stalls, but it should clear most of India and enter Pakistan by mid-July.
3. Rainfall during June should be near to above normal in many areas, but it can be drier than normal in northwest India.
4. Areas of dryness can develop in Kerala, Karnataka and parts of northeast India in July or early August. We are also concerned for dryness becoming an issue in portions of northwest India, Tamil Nadu, Bangladesh and Sri Lanka late in the season.
Possible risks for the monsoon forecast are as follows…
1. A negative IOD developing late in the season as expected should result in dryness in northwest India as forecast. However, if the IOD remains neutral, then some teleconnections suggest a wetter pattern in northwest India, especially late in the season. This could result in rainfall closer to normal or perhaps even above normal in northwest India. The ECMWF model even hints at a weakly positive IOD which could also mean wetter than forecast conditions in northwest India.
2. If ENSO trends back toward a classic La Nina in late summer/autumn then rainfall across India could be greater than forecast. La Nina to neutral usually results in rainfall reduced by 4% across India, but back to back La Ninas tend to favor above normal rainfall in many areas. The exception is toward the southern tip of peninsular India.
3. More lows forming in the Bay of Bengal this year versus 2020 could result in better rainfall across northeast India and Bangladesh than currently forecast.
4. Climate modeling suggests a wetter outlook than forecast, which is a risk.
Lastly, a quick look at the tropical northern Indian Ocean which can be a wildcard for the monsoon. We are currently seeing signals that suggest that the Bay of Bengal should have an increase in the number of lows and depressions pre-monsoon and during the monsoon as compared to 2020 when the basin was fairly quiet until late summer and autumn. We are expecting two to three depressions from May through September versus only one tropical system in May 2020. There is also the risk for a couple of cyclonic storms to develop in the northern Indian Ocean during May and June, with a leaning toward the Bay of Bengal for such activity. However, the Arabian Sea can still have tropical activity pre-monsoon, but it is expected to be less active than 2020 with zero or one tropical feature in May and June compared to two systems in 2020.
Some highlights of the forecast include…
1. Onset in Kerala around the normal onset date of 1 June with a margin of +/- 4 days with onset in Sri Lanka by the end of May.
2. Monsoon advance should be relatively steady with a few usual surges and stalls, but it should clear most of India and enter Pakistan by mid-July.
3. Rainfall during June should be near to above normal in many areas, but it can be drier than normal in northwest India.
4. Areas of dryness can develop in Kerala, Karnataka and parts of northeast India in July or early August. We are also concerned for dryness becoming an issue in portions of northwest India, Tamil Nadu, Bangladesh and Sri Lanka late in the season.
Possible risks for the monsoon forecast are as follows…
1. A negative IOD developing late in the season as expected should result in dryness in northwest India as forecast. However, if the IOD remains neutral, then some teleconnections suggest a wetter pattern in northwest India, especially late in the season. This could result in rainfall closer to normal or perhaps even above normal in northwest India. The ECMWF model even hints at a weakly positive IOD which could also mean wetter than forecast conditions in northwest India.
2. If ENSO trends back toward a classic La Nina in late summer/autumn then rainfall across India could be greater than forecast. La Nina to neutral usually results in rainfall reduced by 4% across India, but back to back La Ninas tend to favor above-normal rainfall in many areas. The exception is toward the southern tip of peninsular India.
3. More lows forming in the Bay of Bengal this year versus 2020 could result in better rainfall across northeast India and Bangladesh than currently forecast.
4. Climate modeling suggests a wetter outlook than forecast which is a risk.
Lastly, a quick look at the tropical northern Indian Ocean which can be a wildcard for the monsoon. We are currently seeing signals that suggest that the Bay of Bengal should have an increase in the number of lows and depressions pre-monsoon and during the monsoon as compared to 2020 when the basin was fairly quiet until late summer and autumn. We are expecting two to three depressions from May through September versus only one tropical system in May 2020. There is also the risk for a couple of cyclonic storms to develop in the northern Indian Ocean during May and June, with a leaning toward the Bay of Bengal for such activity. However, the Arabian Sea can still have tropical activity pre-monsoon, but is expected to be less active than 2020 with zero or one tropical feature in May and June compared to two system in 2020.
Report a Typo