Welcome to Elena Surovyatkina's Monsoon Page!

October 16, 2024

First Ever Forecast of Monsoon Withdrawal Dates Across India: Verification 

I am thrilled to inform my readers that my forecasts of the withdrawal  dates of the Indian Summer Monsoon for 10 regions across India from Rajasthan to Kerala have been proven correct. Each of 10 regions corresponds to the isochrone of the normal date of monsoon retreat, and my prediction of monsoon withdrawal dates concurs with other regions belonging to the entier isochrone (see figure of forecast with isochrones below). By using isochrone prediction, I am able to predict the monsoon withdrawal dates for the whole of India.

On September 14, 2024, I published a forecast of the monsoon's withdrawal providing precise withdrawal periods for each region:

1. North Gujarat, Most Western Rajasthan
2. West Rajasthan
3. Haryana, Chandigarh, Delhi
4. West Uttar Pradesh
5. West Maharashtra, Mumbai, Pune
6. East Madhya Pradesh
7. Central India, Vidarbha
8. Chhattisgarh North
9. Telangana
10. Kerala

You can find concrete evidence on the map provided by the Indian Meteorological Department, where the red isochrones represent the normal monsoon withdrawal dates and the blue isochrones represent the actual monsoon onset. The IMD's observation is in perfect agreement with my forecasts across India from Rajasthan to Kerala.

September 14, 2024

Monsoon Withdrawal Forecasts for the first time in 10 regions in India 

This forecast marks a groundbreaking step in India's meteorological and agricultural landscape. By issuing monsoon withdrawal forecasts for the first time in 10 regions, I am addressing a critical gap that has long affected farmers' ability to plan for the end of the monsoon season. This scientific endeavour is inspired by the successful monsoon onset forecasts for 22 regions, and the goal is to offer Indian farmers a reliable schedule of monsoon withdrawal over the whole of India from West Rajasthan to Kerala, with an uncertainty of +/- 5 days.

Accurate predictions of the withdrawal period will allow farmers to make well-informed decisions regarding harvesting and post-monsoon agricultural activities. This is particularly important for regions where the narrow time window between the summer monsoon withdrawal and the onset of the winter monsoon poses a challenge for agricultural planning, especially in northern and central-eastern India.

In conclusion, this effort to forecast monsoon withdrawal is a significant investment in India’s future. It directly supports the national priorities of enhancing agricultural productivity, protecting rural livelihoods, and improving the accuracy of forecasting systems. The success of this forecast has the potential to revolutionize Indian agriculture, strengthen food security, and enhance the country’s preparedness for extreme weather events like droughts and floods, contributing to a more sustainable and resilient future for the nation.

August 6, 2024

Breakthrough in Monsoon Onset Forecasting: 12 States in India in 2024

Executive Summary:

Elena Surovyatkina, from the Potsdam Institute for Climate Impact Research (PIK), has made an unprecedented breakthrough in the 200-year history of monsoon research. She has successfully developed and implemented a methodology for accurate long-term monsoon onset predictions, covering half of the Indian subcontinent.

According to Prof. Elena Surovyatkina, predicting the monsoon onset date 40 days in advance and the withdrawal date 70 days in advance can solve problems. This solution offers various benefits. Firstly, it can help farmers plan their crop cycles and irrigation schedules more efficiently, increasing productivity and crop yields, positively impacting food security and reducing food prices. Secondly, early warnings about monsoon onset and withdrawal can mitigate risks associated with extreme weather events such as floods and droughts, which can damage crops and livelihoods. Finally, accurate forecasts of monsoon onset and withdrawal can play a crucial role in the sustainable transformation of agriculture and food systems. It can help the government prepare for changing weather patterns and build resilience in agricultural practices, contributing to the overall transformation of agriculture.

Verification of the forecast of the onset of the monsoon for the first time for seven regions in India

Delayed Onset Predictions: This year, my predictions have proven remarkably precise with the delayed monsoon onset. On May 10, I predicted a delay in the monsoon onset for Telangana, Central India, and Delhi. This prediction accurately forecasted the delayed monsoon, which began on June 26 in Western Telangana and June 27 in Eastern and Northern Telangana, perfectly aligning with the forecasted dates of 20-28 June, 21-29, and 22-30. Furthermore, the monsoon onset in the Eastern Ghats on June 27 fell precisely within the forecasted period of 22-30 June.
In Eastern Madhya Pradesh (EMP), the monsoon established itself on June 26, three days earlier than the predicted window of June 29 to July 6 and three weeks later on July 22 in Delhi, one day later than predicted. The evidence from observational data unambiguously supports the forecasts of delayed monsoon onset made on May 8, 2024, more than 40 days in advance (and, in fact, 50 days) and remained unchanged since the initial forecast with acceptable minor deviations.

Expansion of forecast area to 12 states in India

I announced the monsoon forecast for seven regions on May 10, and then I received many requests from India, especially regarding agriculture. In response, I adapted my research, originally planned for seven regions, to include the central and northern parts of India. This adaptability allowed me to make forecasts for twelve states of India. Below, you will find the verification of the monsoon onset forecast in these states, a testament to the flexibility of my method.

1. Telangana (West, East, North)                                                  (June 26-27)
2. Maharashtra - East (Vidarbha), Central (Marathwada)   (June 27)
3. Chhattisgarh                                                                                    (June 27)
4. Uttarakhand                                                                                    (June 21)
5. Odisha                                                                                                (June 27)
6. Gangetic West Bengal                                                                  (June 26)
7. Jharkhand                                                                                        (June 26)
8. Bihar                                                                                                   (June 24)
9. Uttar Pradesh - East, West                                                          (July 22)
10. Rajasthan – East                                                                          (July 16)
11. Gujarat region                                                                              (July 28)
12. Madhya Pradesh - East, West                                                  (June 26)

The monsoon onset dates in 12 states in India, confirmed by temperature and relative humidity data, indicate a delayed monsoon in the central part of India. And this delay was resolved by well-organized and sustained rainfall from June 21 to June 27, marking the beginning of the rainy season. Furthermore, there was a considerable delay in the monsoon onset in Eastern Rajasthan and Delhi, where the monsoon commenced only on July 16 and July 22 respectively. These observations are consistent with the forecasts I made 50 days in advance.

Furthermore, the monsoon onset date is consistent in regions along the isochron in Central India (June 27) and Eastern Madhya Pradesh (June 26). This signifies organised and sustained rainfall, which is the true definition of monsoon onset. The figure below shows subdivisions highlighted by transparent blue circles along the isochrones where the correct dates of monsoon onset forecast were observed based on daily temperature and relative humidity observation data between 21 to 30 June in Central India and 22 July in Delhi in 2024. Hence, this correct 12-states monsoon forecast covers half of the Indian subcontinent, benefiting millions of Indian farmers for agriculture planning in 2024.

Mutually beneficial collaboration:

 Since 2016, I have visited India annually to share my knowledge and accomplishments with the meteorological community and academic researchers. I have had the opportunity to collaborate with esteemed institutions such as the India Meteorological Department in Delhi, the National Centre for Medium-Range Weather Forecasting (NCMRWF), the Indian Institute of Tropical Meteorology in Pune, the India Meteorological Department in Mumbai, The University of Delhi, Banaras Hindu University, the Institute of Environment & Sustainable Development, and DST – Mahamana Centre of Excellence in Climate Change, among others. These interactions have allowed us to exchange insights about monsoon features, and for me to share my recent advances in forecasting monsoon onset and withdrawal. I am awaiting your response to my 12 states forecast for 2024, as our collaboration will not only benefit our respective institutions but also contribute significantly to the field of meteorology.

Prof. Elena Surovyatkina is visiting the University of Delhi, the NCMRWF, the IITM, Pune, the IMD Mumbai, Banaras Hindu University. Contact: Elena Surovyatkina (elena.surovyatkina@pik-potsdam.de)

Agricultural Implications:

The ability to predict monsoon timing with such accuracy is a game-changer for India's agricultural sector. Farmers, who constitute 70% of India's workforce, depend heavily on timely monsoon rains for their crop cycles. My forecasts enable them to plan their planting and irrigation schedules more effectively, reducing the risk of crop failure and enhancing food security.

A Case Study in Telangana:

In Telangana, my forecasts have demonstrated their practical value. The Indian Meteorological Department (IMD) announced the monsoon's onset over the state between June 6 and June 10,
despite no rainfall in those regions. The IMD's declaration led farmers to expect good rainfall and start seeding. However, a dry spell from June 13th to June 20th appeared, which could have tragic consequences for farmers who followed the IMD's monsoon onset declaration. However, my forecast of a delayed onset on June 26-27 helped farmers avoid premature planting. Those who followed my forecast were able to plan their agricultural activities better, mitigating the risk posed by the subsequent dry spell.

International Collaboration with IRRI:

I made a precise forecast for the monsoon onset as part of an experiment with the International Rice Research Institute (IRRI). This project aims to advance the rice-wheat cropping systems in Bihar. The monsoon arrived in Bihar on June 24, which falls within the forecasted period of June 21-29, and was beneficial for farmers. Accurate forecasting is crucial for timely agricultural operations, climate resilience, resource management, and enhanced productivity. The experiment focuses on predicting the onset of continuous rainfall to ensure the latest scientific advancements are utilised effectively.

Definition of Monsoon Onset by Elena Surovyatkina

My definition of monsoon onset is universal for every location and is based on the critical values of two atmospheric variables: temperature (Tc) and relative humidity (RHc). When the temperature drops below the critical temperature and the relative humidity stays above 80%, it implies the arrival of the monsoon season. It's also important to note that spatially organized and sustained rainfall over time is crucial in defining monsoon onset.
Outgoing longwave radiation (OLR) is the third indicator for the upcoming monsoon, characterised by convective activity, implying scarcity or deep convective clouds. However, the OLR data has yet to be available. Nevertheless, it can be used in historical analysis.

My definition differs from IMD, which applies only to Kerala and cannot be extended to other regions. This is why the IMD never forecasted monsoon onset in any other region in India. With my universal approach, I can forecast monsoons across the tropics.

The observational evidence in support of each forecast by temperature (T) and relative humidity (RH)

Daily mean near-surface air temperature till July 20, 2024, for Western Maharashtra (red). For reference, the purple line is past the 5-year average for the same regions. The dashed line shows the boundary between pre-monsoon and monsoon. Once the temperature drops below the dashed line, monsoons start inevitably. I forecasted monsoon onset to arrive in Western Maharashtra between June 22 and 30. Additionally, I forecasted the appearance of a dry spell after June 3, before monsoon onset. The southwest monsoon arrived in Western Maharashtra on June 25, 2024, precisely according to a schedule I had forecasted.

Daily mean near-surface relative humidity (Rh) till July 20, 2024, for Western Maharashtra (red). For reference, the purple line is past the 5-year average for the same regions. The vertical dashed line shows the boundary between pre-monsoon and monsoon. Once the relative humidity crosses 80% on June 25 and stays above 80%, it indicates the arrival of the monsoon season, which continues until October. For reference, the vertical grey line from the right shows the 5-year average date of monsoon withdrawal from the region.

Daily mean near-surface air temperature till July 20, 2024, for Western Telangana (red). For reference, the purple line is past the 5-year average for the same regions. The dashed line shows the boundary between pre-monsoon and monsoon. Once the temperature drops below the dashed line, monsoons start inevitably. I forecasted monsoon onset to arrive in Western Telangana between June 20 and 28. Additionally, I forecasted the appearance of a dry spell after June 3, before monsoon onset. The southwest monsoon arrived in Western Telangana on June 26, 2024, precisely according to a schedule that I had forecasted.

Daily mean near-surface relative humidity (Rh) till July 20, 2024, for Western Telangana (red). For reference, the purple line is past the 5-year average for the same regions. The vertical dashed line shows the boundary between pre-monsoon and monsoon. Once the relative humidity crosses 80% on June 27 and stays above 80%, it indicates the arrival of the monsoon season, which continues until October. For reference, the vertical grey line from the right shows the 5-year average date of monsoon withdrawal from the region.

Daily mean near-surface air temperature till July 20, 2024, for Central India(red). For reference, the purple line is past the 5-year average for the same regions. The dashed line shows the boundary between pre-monsoon and monsoon. Once the temperature drops below the dashed line, monsoons start inevitably. I forecasted monsoon onset to arrive in Central India between June 22 and 30. Additionally, I forecasted the appearance of a dry spell after June 12, before monsoon onset. The southwest monsoon arrived in Central India on June 27, 2024, precisely according to a schedule that I had forecasted.

Daily mean near-surface relative humidity (Rh) till July 20, 2024, for Central India (red). For reference, the purple line is past the 5-year average for the same regions. The vertical dashed line shows the boundary between pre-monsoon and monsoon. Once the relative humidity crosses 80% on June 27 and stays above 80%, it indicates the arrival of the monsoon season, which continues until October. For reference, the vertical grey line from the right shows the 5-year average date of monsoon withdrawal from the region.

Daily mean near-surface air temperature till July 20, 2024, for Eastern Madhya Pradesh (red). For reference, the purple line is past the 5-year average for the same regions. The dashed line shows the boundary between pre-monsoon and monsoon. Once the temperature drops below the dashed line, monsoons start inevitably. I forecasted monsoon onset to arrive in Eastern Madhya Pradesh between June 29 and July 6. Additionally, I forecasted the appearance of a dry spell after June 14, before monsoon onset. The southwest monsoon arrived in Eastern Madhya Pradesh on June 26, 2024; three days earlier to the schedule that I had forecasted.

Daily mean near-surface relative humidity (Rh) till July 20, 2024, for Eastern Madhya Pradesh (red). For reference, the purple line is past the 5-year average for the same regions. The vertical dashed line shows the boundary between pre-monsoon and monsoon. Once the relative humidity crosses 80% on June 26 and stays above 80%, it indicates the arrival of the monsoon season, which continues until October. For reference, the vertical grey line from the right shows the 5-year average date of monsoon withdrawal from the region.

Daily mean near-surface air temperature till July 20, 2024, for Delhi (red). For reference, the purple line is past the 5-year average for the same regions. The dashed line shows the boundary between pre-monsoon and monsoon. Once the temperature drops below the dashed line, monsoons start inevitably. I forecasted monsoon onset to arrive in Delhi between June 14-2. Additionally, I forecasted the appearance of a dry spell after June 29, before monsoon onset. The southwest monsoon arrived in Delhi on July 22, 2024; one days earlier to the schedule that I had forecasted.

Daily mean near-surface relative humidity (Rh) till July 20, 2024, for Delhi (red). For reference, the purple line is past the 5-year average for the same regions. The vertical dashed line shows the boundary between pre-monsoon and monsoon. An extreme event appeared as pre-monsoon rainfall in the region because, after June 28, Delhi did not record any "heavy" rainfall day until July 22. It has been particularly hot and humid, with calm winds and high humidity above 80%, but the population in Delhi did not see any significant rain until July 21. On July 22, the monsoon season began in Delhi. This agrees with my forecast of a dry spell from June 29 until the monsoon onset, which I predicted from July 14-21 and that appeared one day later, June 22.

Daily mean near-surface air temperature till July 20, 2024, for Uttarakhand (red). For reference, the purple line is past the 5-year average for the same regions. The dashed line shows the boundary between pre-monsoon and monsoon. Once the temperature drops below the dashed line, monsoons start inevitably. I forecasted monsoon onset to arrive in Bihar between June 20 and 28. Additionally, I forecasted the appearance of a dry spell after June 3, before monsoon onset. The southwest monsoon arrived in Uttarakhand on June 21, 2024, precisely according to a schedule had forecasted.

Daily mean near-surface relative humidity (Rh) till July 20, 2024, for Uttarakhand (red). For reference, the purple line is past the 5-year average for the same regions. The vertical dashed line shows the boundary between pre-monsoon and monsoon. Once the relative humidity crosses 80% on June 21 and stays above 80%, it indicates the arrival of the monsoon season, which continues until October. For reference, the vertical grey line from the right shows the 5-year average date of monsoon withdrawal from the region.

Daily mean near-surface air temperature till July 20, 2024, for Bihar (red). For reference, the purple line is past the 5-year average for the same regions. The dashed line shows the boundary between pre-monsoon and monsoon. Once the temperature drops below the dashed line, monsoons start inevitably. I forecasted monsoon onset to arrive in Bihar between June 21 and 29. Additionally, I forecasted the appearance of a dry spell after June 3, before monsoon onset. The southwest monsoon arrived in Central India on June 24, 2024, precisely according to a schedule had forecasted.

Daily mean near-surface relative humidity (Rh) till July 20, 2024, for Bihar (red). For reference, the purple line is past the 5-year average for the same regions. The vertical dashed line shows the boundary between pre-monsoon and monsoon. Once the relative humidity crosses 80% on June 24 and stays above 80%, it indicates the arrival of the monsoon season, which continues until October. For reference, the vertical grey line from the right shows the 5-year average date of monsoon withdrawal from the region.

June 18, 2024

Understanding the Monsoon Delays in Telangana and Maharashtra

We must admit that in the first half of June, most of the monsoon rainfall went to Mainland Southeast Asia, leaving India with light to moderate isolated rainfall and only a few strong short-term rainfalls in isolated places on the coast of the Arabian Sea.

As Ananthakrishnan and M.K. Soman noted in their 1990 study, "The onset of monsoon... is not a transition from a regime of no rain to rain; it is a transition from a regime of sporadic rainfall to spatially organized and temporally sustained rainfall." This essential characteristic of the monsoon has not yet been observed in Telangana and Maharashtra. Instead, these regions have experienced no spatially organized or temporally sustained rainfall.

The declaration by the IMD of monsoon onset over Telangana and western Maharashtra did not align with the expected weather patterns; the true monsoons did not show up in these regions.
So, what is happening? The monsoon is delayed, and since June 10, we have not observed a true onset of the monsoon. Instead, we are witnessing a transition from pre-monsoon to monsoon in a series of steps, with rains alternating with dry spells.

It is important to consider that the forecast onset dates I provided for seven regions in India signify the beginning of continuous rainfall - the true monsoon season. This continuous and organized rainfall is critical for agriculture, water management, and disaster preparedness. Understanding these nuances is essential for effectively utilizing monsoon forecasts to plan and manage resources.

May 10, 2024

First-time Forecast of the Onset dates of the Indian Summer Monsoon for Seven Regions over India: Schedule the Advancement of the Monsoon

Every year, the onset of the monsoon poses a mystery by coming at different times and creating challenges for Indian society. This is particularly true for farmers who need to plan their agriculture cycle in advance. A few days ago, the pre-monsoon rains began in India, and it is crucial to know when the actual monsoon season will begin.

For the first time, I am issuing a forecast for the monsoon onset in seven regions to schedule the advancement of the monsoon. Over eight years of experience reliably forecasting monsoons in Central India (see results at the end of the news) encourage me to take this step forward. This is a big scientific challenge that has never been done before, but we need to start doing it.

Based on current conditions, I anticipate a delay in the onset of the monsoon this year in Telangana, Central India, and especially in Delhi. The reason for the delayed monsoon is the negative anomaly of temperature over central and especially Northern parts of India from March to April, as shown on the map below. Thus, it will push the onset of the monsoon season to the end of June. Additionally, the dry spell in these regions during June will also contribute to the delay.

It is crucial for farmers to have information about the timing of dry spells and the onset of the monsoon in their regions, as they occur at different times in different parts of India. Such forecasts help them choose the right time for planting their crops. The unique monsoon onset forecast can provide valuable insights that the government can leverage for strategic planning and disaster response.

Here are my monsoon onset forecasts for regions between 15°N-30°N latitudes in India. I will update the forecast and notify you of any significant changes in June.

The forecasts for each region are presented below.

1. Daily mean near-surface air temperature till May 5, 2024, for Western Telangana (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. Dry spell will likely occur after June 3rd, and the monsoon onset is expected between June 20-28.

2. Daily mean near-surface air temperature till May 5, 2024, for Eastern Telangana (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. Dry spell will likely occur after June 9th, and the onset of monsoon is expected between June 21-29.

3. Daily mean near-surface air temperature till May 5, 2024, for Northern Telangana (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. Dry spell will likely occur after June 12th, and the onset of monsoon is expected between June 22-30.

4. Daily mean near-surface air temperature till May 5, 2024, for Western Maharashtra, Mumbai, Pune (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. Dry spell will likely occur after June 3rd, and the monsoon onset is expected between June 22-30.

5. Daily mean near-surface air temperature till May 5, 2024, for Central India (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. Dry spell will likely occur after June 12th, and the monsoon onset is expected between June 22-30.

6. Daily mean near-surface air temperature till May 5, 2024, for Eastern Madhya Pradesh (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. Dry spell will likely occur after June 14th, and the monsoon onset is expected between June 29 - July 6.

7. Daily mean near-surface air temperature till May 5, 2024, for Delhi (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. Dry spell will likely occur after June 29th, and the monsoon onset is expected between July 14 - July 21.

Surface air temperature anomaly for April 2024 relative to the April average for 1991-2020. Data source: ERA5. Credit: Copernicus Climate Change Service/ECMWF. 
https://climate.copernicus.eu/surface-air-temperature-april-2024

One of the reasons for the delayed monsoon season is the negative anomaly of temperature over central and especially northern parts of India from March to April, as shown on the map. Thus, it will push the onset of the monsoon season to the end of June.

Monsoon Onset Forecast Experiment in Bihar for Enhancing Rice-Wheat Cropping Systems

I am pleased to inform my readers about an ongoing experiment aimed at providing accurate monsoon onset forecasts to support agricultural practices in Bihar. This initiative is particularly focused on benefiting the coupled rice-wheat cropping systems in eastern India, an area where precise timing is crucial for managing resources, building climate resilience, and enhancing productivity.

The Importance of Accurate Forecasting:

1. Forecasting the onset date of the monsoon is vital for the following reasons:
Timely Agricultural Operations: Accurate forecasts allow farmers to plan their planting schedules more effectively, ensuring that rice is sown at the optimal time for maximum yield.
2. Climate Resilience: By anticipating the monsoon, farmers can implement necessary measures to protect their crops from extreme weather conditions, thereby reducing the risk of crop failure.
3. Resource Management: Understanding the timing of the monsoon helps in the efficient use of water resources, fertilizers, and labour, leading to overall better farm management.
4. Enhanced Productivity: With precise forecasts, the transition between rice and wheat cultivation can be managed more smoothly, resulting in higher productivity and profitability. 

The Experiment:

I am conducting this experiment in Bihar to provide accurate monsoon onset forecasts specifically tailored for the region. The forecast aims to predict the onset of continuous rainfall, which marks the true beginning of the monsoon season.

Collaboration with IRRI:

Given the significant role of the International Rice Research Institute (IRRI), South Asia Regional Centre (ISARC), Varanasi, UP, India, in advancing rice research and improving agricultural practices, I believe my collaboration with Dr. Amit Srivastava from the ISARC will be highly beneficial. By working together, we can ensure that the latest scientific advancements in monsoon forecasting are effectively utilized to support the rice-wheat cropping systems in Eastern India.

 Daily mean near-surface air temperature till May 5, 2024, for Bihar (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. A dry spell will likely occur after June 3rd, and the monsoon onset is expected between June 21 - 29.

October 1, 2023

Proof of monsoon withdrawal forecast for Central India, 2023

Proof of monsoon withdrawal forecast for Delhi for 2023

I am sharing observational evidence that proves the success of the long-term forecast 40 days in advance of the monsoon withdrawal in Delhi in 2023. The monsoon withdrawal from Delhi on 30 September falls within the forecasted dates of 30 Sep-9 Oct. The map of the Withdrawal of Southwest Monsoon 2023 was produced by the Indian Meteorological Department by observation shows the evidence. The red lines on the map show the average date, while the blue lines show the withdrawal of the monsoon from Delhi on 30 September, observed by fact.

August 20, 2023

Forecast of the Withdrawal Date of Indian Summer Monsoon - 2023 from Delhi, the Central part of India, and Northern Telangana

In 2023 the Withdrawal of the Indian Summer Monsoon (Southwest Monsoon) over Northwest India and Delhi is expected to occur within the timeframe of September 30 to October 9.

The withdrawal of the Southwest Monsoon from the Eastern Ghats will likely not be abrupt. Rainfall reduction might begin from the beginning of October, but the last rainfall outbreak might happen between the 7-17 October. In the middle of October, it is also possible that the Winter Monsoon (Northeast Monsoon) may arrive and bring rainfall to the Eastern Ghats until November.

The Southwest Monsoon will withdraw from Northern Telangana in the period from October 10 to 20. Northern Telangana also may also experience rainfall from the Northeast Monsoon.

What caused the Monsoon Break in August 2023?

From April to July 2023, in northwestern India and north Pakistan, colder-than-average surface air temperature anomalies of around 4 °C persist (See Surface air temperature anomaly below). In July, the daily average temperatures in this area were significantly colder, by 4-6 °C less than usual, which is a rare occurrence. Due to this, there were minimal differences in temperatures between the northwestern subtropics near north Pakistan and the tropics in the central part of India. Consequently, the monsoon trough appeared in unstable conditions and was forced to shift, in particular, north of its normal position to the foothills of the Himalayas in the first half of August, resulting in good rainfall in Uttarakhand and the Northeast states of India, but leading to "Break Monsoon" conditions where rainfall activity over the rest of India reduced or ceased.
However, during the latter half of August, the temperature anomalies mentioned earlier decreased slightly. With two months remaining, it is possible that these anomalies may disappear. Therefore, they should not greatly impact the withdrawal of the monsoon in October.

I have postponed my monsoon withdrawal forecast this year due to the unclear trend that persisted two weeks ago. As a result, I issue the forecast from 40 to 50 days in advance of the withdrawal.

Daily mean near-surface air temperature and relative humidity till August 17, 2023, for the Delhi region (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. The Indian Summer Monsoon (Southwest Monsoon) will likely withdraw from Northwest India and Delhi between September 30 and October 9, 2023.

Daily mean near-surface air temperature and relative humidity till August 17, 2023, for the Eastern Ghats (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. The Southwest Monsoon will likely retreat from the Eastern Ghats region around October 7-17. Moreover, the Winter Monsoon (Northeast Monsoon) may make its way in the middle of October and cause precipitation in the area until November.

Daily mean near-surface air temperature till August 17, 2023, for Telangana (red) and Afghanistan (blue). Violet and grey lines are past the 5-year average for the same regions. Between October 10 and 20, the Southwest Monsoon is expected to recede from Northern Telangana. During this time, Northern Telangana may also receive rainfall from the Northeast Monsoon.

Surface air temperature anomaly for July 2023 relative to the July average for the period 1991-2020. Data source: ERA5. Credit: Copernicus Climate Change Service/ECMWF. https://climate.copernicus.eu/surface-air-temperature-july-2023
Between April and July 2023, there were persistently colder-than-average surface air temperature anomalies of approximately 4°C in northwestern India and north Pakistan, as indicated by the Surface air temperature anomaly map. Please find the map for April 2023 below for comparison. In August, this resulted in unstable conditions of the monsoon trough, which shifted north, causing heavy rain in Uttarakhand and Northeastern India, while the rest of India experienced decreased or no rainfall due to "Break Monsoon" conditions.

On a final note

As we look at the issue of monsoon withdrawal in India, it becomes clear that accurate forecasting is essential for farmers in the region. My unique monsoon withdrawal forecast (no any another agency provides it due to a lack of methodology) offers valuable insights into this phenomenon, which can be leveraged by the government for strategic planning and disaster response.

With the increasing frequency of severe storms and floods during monsoon withdrawal, it is critical that preventive measures are taken to avoid overfilling and flooding in October. The government should consider integrating the withdrawal forecast into their dam system management to ensure effective resource consolidation and capacity building.

I believe that the use of advanced forecasting methods like this is essential for sustainable development in the face of a changing climate. By providing reliable and timely information, we can empower communities to make informed decisions and build resilience against future challenges.

Forecasts of monsoon onset 2023 verification

I am sharing observational evidence that verifies the long-term forecasts of the monsoon onset in Telangana, Central India, and Delhi in 2023. The monsoon onset occurred on 23 June in Northern Telangana, which falls within the forecasted dates of 19-24 June. In the Eastern Ghats, the onset occurred on 24 June, which is also within the forecasted dates of 15-26 June. This matches the monsoon onset forecast I made on 6 May. You can view the map of the Advance Southwest Monsoon 2023 produced by the Indian Meteorological Department to see the evidence. The red lines on the map show the average date, while the blue lines show the onset of the monsoon observed by fact.

The IMD declared the onset of the monsoon in Delhi on 30 June, but it turned out to be pre-monsoon rainfall. The temperatures in Delhi increased after that, and the real onset of monsoon appeared on 9-10 July 2023. Delhi received an excess of rainfalls, which were devastating due to their intense and extreme nature. I had forecasted the monsoon onset in Delhi on 12 July +/- 4 days, and it appears that my forecast was correct.

Progress of SW Monsoon 2023 by the IMD - India Meteorological Department. The map displays red lines indicating the average date of monsoon onset and blue lines representing the actual onset of the monsoon as observed.
https://mausam.imd.gov.in/imd_latest/contents/monsoon.php

May 8, 2023

Forecast of the Onset dates of the Indian Summer Monsoon - 2023 over the central part of India, Telangana, and Delhi

Every year, the onset of the monsoon poses a puzzle by coming in different ways. However, I expect some good news - pre-monsoon temperatures in the Central, North, and Northwest parts of India might be a bit cooler than in the previous year. However, the advance of the monsoon may not be as organized as desired.
With the purpose of assisting farmers with planning their agriculture cycle, I am conducting a second-year experiment by providing detailed information about the monsoon onset forecast. This includes estimations for intermittent rain periods, dry spells, and the beginning of persistent rains. By knowing where and when the monsoon starts, as well as if there will be a dry spell after the initial rain, farmers can make informed decisions about planting at the appropriate time.

Delayed Monsoon Onset Forecasts over the Central part of India, North Telangana, and Delhi

According to my analysis, in Central India, pre-monsoon intermittent rains are likely to occur before June 15. Please be aware that a dry spell may be experienced until June 26. Eventually, continuous monsoon rainfalls are expected after June 26. My forecasted dates are subject to an uncertainty range of +/- 4 days. I performed this forecast 40 days in advance.

In North Telangana, the pre-monsoon weather conditions are relatively milder compared to Central India. It's possible to experience occasional pre-monsoon showers before 19th June, but there's a high likelihood of a dry spell between 19th to 24th June. From 24th June onwards, I can anticipate persistent monsoon rains. Please note that the forecasted dates have a margin of error of +/- 4 days. I made this prediction 40 days ahead of time.

In the Delhi region, it is possible to experience pre-monsoon rains towards the end of June. However, the onset of Monsoon rains is expected to be after the 12th of July 2022. There may be some dry spells between the 19th to 27th of July. However, from the 27th of July onwards, the region might experience more persistent monsoon rainfall. The predicted dates may have a margin of error of +/- 4 days. I made this forecast 60 days in advance.

What are the special conditions for the upcoming monsoon 2023?

1. During March, a cold anomaly zone developed over Telangana, the Central, and South regions of India, which later shifted to the North in April. This resulted in temperatures being 2° - 8° below average in Central India and 3°-6° in North Pakistan (See figure 'Surface air temperature anomaly for April 2023' below). These below-average temperatures are expected to persist until May and early June, prior to the onset of the monsoon. The Central part, Telangana and Delhi, are experiencing favorable conditions that may delay the onset of the monsoon.

2. During the months of March and April, the Southern Indian Ocean had higher air temperatures than usual. Meanwhile, the equatorial eastern Pacific Ocean was gradually warming up after experiencing below-average La Niña temperatures for almost three years (See figure 'Surface air temperature anomaly for April 2023' below). This warming indicates that there may be a transition to El Niño conditions, which is happening slowly. As a result, in June, when the Indian summer monsoon arrives, the equatorial Pacific Ocean will be in a neutral state, which is unstable. This condition will cause alternating intermittent rains and dry spells before the monsoon settles.

Daily mean near-surface air temperature and relative humidity till May 3, 2023, for the Eastern Ghats (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. Central India can expect intermittent pre-monsoon rains before June 15. After that, be aware that a dry spell might appear until June 26. Persistent monsoon rainfalls are expected after the 26th of June. Please keep in mind that my forecasted dates have a narrow uncertainty range of +/- 4 days.

Daily mean near-surface air temperature till May 3, 2023, for Telangana (red) and Afghanistan (blue). Violet and grey lines are past the 5-year average for the same regions. In Northern Telangana, the pre-monsoon weather is more gentle than in Central India. There may be intermittent pre-monsoon showers before June 19th, but a dry period is likely to occur between June 19th and 24th. Starting from June 24th, I expect persistent monsoon rains. It's important to keep in mind that the predicted dates may have a deviation of up to four days.

Daily mean near-surface air temperature and relative humidity till May 3, 2023, for the Delhi region (red) and North Pakistan (blue). Violet and grey lines are past the 5-year average for the same regions. In the Delhi area, pre-monsoon rains can be expected towards the end of June, while I predict the onset of monsoon rains to occur after July 12th, 2022. There may be occasional dry periods between July 19th and 27th, but from August 27th onwards, the region is likely to experience more consistent monsoon rainfall. Notably, the predicted dates may have a margin of error of +/- 4 days.

Surface air temperature anomaly for April 2023 relative to the April average for the period 1991-2020. Data source: ERA5. Credit: Copernicus Climate Change Service/ECMWF. https://climate.copernicus.eu/surface-air-temperature-april-2023
Please take note of the cold anomaly zone (showing below-average temperatures) present over India and the warm anomaly (featuring above-average temperatures) observed in the equatorial Pacific Ocean. These factors, along with others, may influence the monsoon season's delay.

August 9, 2022

Forecast of the Withdrawal dates of the Indian Summer Monsoon - 2022 over the central part of India, Telangana, and Delhi

The challenging summer monsoon season-2022 in Delhi is not going to improve. The alternation of a dry spell and intermittent rainfalls in Delhi will likely stand until October 6, and then the monsoon season in Delhi will be over.

In 2022 the Withdrawal of the Indian Summer Monsoon (Southwest Monsoon) over the Eastern Ghats will likely not be sharp. Rainfall reduction begins on October 4, but isolated rains are very likely until October 13. However, in the period 14-18 October, Winter Monsoon (Northeast Monsoon) will reach the Eastern Ghats and brings rainfall again that might continue until October 23 and even longer until November 1.

The Southwest Monsoon will withdraw from Northern Telangana 2-3 days later than the Eastern Ghats. Northern Telangana also will similarly receive the Northeast Monsoon rainfall.

Daily mean near-surface air temperature till August 6, 2022, for Delhi (red) and North Pakistan (blue). Violet and grey lines are past 5 years average for the same regions. The forecasted withdrawal dates are from the 1st to the 6 of October 2022.

Daily mean near-surface air temperature till August 6, 2022, for Central India (red) and North Pakistan (blue). Violet and grey lines are past 5-years average for the same regions. The forecasted withdrawal dates are from the 4th to the 12th of October 2022.

Daily mean near-surface air temperature till August 6, 2022, for Telangana (red) and North Pakistan (blue). Violet and grey lines are past 5 years average for the same regions. The forecasted withdrawal dates are from the 6th to the 14th of October 2022.

Forecasts of monsoon onset verification

Here, I present the evidence from observations for verifying long-term forecasts of monsoon onset in central India and Telangana in 2022. According to the IMD, the monsoon onset occurred over Northern Telangana and the Eastern Ghats between June 13th and 16th; see map below. That entirely agrees with my forecast for June 14-15. Moreover, a dry spell appeared before beginning continuous rainfall, which started on June 26. Hence, my predictions made 35 days in advance appear to be accurate.

I have predicted the long-delayed monsoon in Delhi on July 10 with an extended dry period before onset. According to the IMD, the monsoon onset occurred over Delhi on June 30. However, the temperature plot above shows that it was bogus onset because, after June 30, the temperature rose again, but after July 10 temperature gradually decreased to monsoon temperature. Moreover, I predicted a dry spell around August 14 that is currently developing in Delhi. Hence, all three predictions regarding Delhi made 60 days in advance appear to be correct.

Why is the Southwest Monsoon 2022 in India so challenging?

While during a heatwave from mid-month onward, numerous July records for maximum temperature were broken in Europe, with temperatures 40°C, the regions bordering the western Indian Ocean, from the Horn of Africa in the west, to southern India in the east, showed below-average temperatures, which saw much above average precipitation. The map shows a cooler than average temperature in blue and warmer in red.
High-temperature anomalies dramatically impact the Indian summer monsoon. While Southern Peninsula, Central, and West India kept the monsoon rain in the vigorous stage, North West and North East India struggled to record good rains without success. The temperature anomaly map in India explains India's tilted separation into wet (blue color) and dry (red color) parts. Two extremes are seen clearly on the map.

Surface air temperature anomaly for July 2022 relative to the July average for the period 1991-2020. Data source: ERA5. Credit: Copernicus Climate Change Service/ECMWF. https://climate.copernicus.eu/surface-air-temperature-july-2022

May 10, 2022

Forecast of the Onset dates of the Indian Summer Monsoon - 2022 over the central part of India, Telangana, and Delhi

August 24, 2021

Forecast of the Withdrawal Date of Indian Summer Monsoon - 2021 from Delhi, the Central part of India, and Northern Telangana

The Indian Summer Monsoon (Southwest Monsoon) is likely to withdraw from Delhi between the 13th and 23rd of September and then from the Central part of India (20N, 80E) between the 30th of September and 10th of October 2021 and 2 days later from Northern Telangana. That is around five days earlier than the normal dates. However, it is highly likely that after the withdrawal of the Southwest monsoon and a few dry days, the Northeast Monsoon begins in the middle of October.

In the last decade, Northeast Monsoon starts closer to the end of October and pours over southeast peninsular India and parts of Sri Lanka. In 2020 earlier start of the Northeast Monsoon in the middle of October happened with unexpected penetration to the central part of India, leading to strong precipitation over the region until the end of October 2020.

The cause for such a predicted scenario is low temperatures (about 3- 4°C lower than average) in northern Pakistan, northern and central India. A similar situation occurred last year, in 2020, and led to rapid cooling in October from Northwest to Southeast. Thus, when the Indian subcontinent gets colder faster, it attracts Northeast wind earlier.

In India, October is the harvesting season due to the end of the monsoon season. The most important message for farmers is that October 2021 might be rainy, with a short dry spell between summer and winter monsoons. In such a circumstance, it would be better to be prepared for a scenario similar to October 2020.

Successful prediction of monsoon onset 2021

The Indian Summer Monsoon began in Northern Telangana on the 24th of June and then neared the Eastern Ghats on the 29 of June. After that, a dry spell appeared and lasted until 10 July.
My onset dates are distinct from the IMD dates due to different definitions of onset. While the IMD declares onset after 48-hours of continuous rainfall, I define a monsoon onset by critical temperature (T_c) and relative humidity (RH_c).
According to the IMD, the monsoon onset occurred over Northern Telangana and the Eastern Ghats between June 11th and 17th. However, according to my prediction, the premature isolated rainfall appeared within this period on June 12 -19. I explicitly specified this period as premature isolated rainfall because my monsoon onset conditions have not been fulfilled. It turned out to be correct because, in the period 19 June -11 July, stagnation in monsoon advance appeared. Hence, my predictions for Northern Telangana and Central India were accurate.

For the first time, I issued the long-term forecast for the Delhi region, where the monsoon forecast has never been made. According to my prediction, the monsoon will begin in Delhi with a significant delay, namely after July 11. Monsoon has reached Delhi on July 13. Thus, my long-term prediction made 63 days in advance was correct.

The evidence from observational data sees in the figures above.

Monsoon onset 2021 forecast was published in Indian newspapers DownToEarth, HindustanTimes, Indian portal Weather&Radar India, and PIK-News.

See evidence of summer monsoon onset factual appearance in 2021 in Indian newspapers: DownToEarth, DownToEarth, DownToEarth.

May 8, 2021

Forecast of the Onset dates of the Indian Summer Monsoon - 2021 over the central part of India, Telangana, and Delhi

April 28, 2021

Msimu rains to retreat from southern Tanzania by the second week of May

The rainfall season in southern Tanzania is coming to an end. Continues rainfall expected until the 5th of May, a sharp decline in rain activity can be expected on the second week of May. The region of the forecast is South Tanzania (10°S, 35°E), namely towards the South of Mbeya, Lake Malawi, and Songea regions.

in the Swahili language by Emanuel, Jacob (ICRISAT-Tanzania):

Mwua za Msimu Kusini mwa Tanzania kufikia mwisho wiki ya pili ya Mwezi Mei

Msimu wa mvua kusini mwa Tanzania unaelekea mwisho. Mvua zinazoendelea zinatarajiwa kufikia mwisho tarehe 5 Mei, huku kukoma kabisa kwa mvua kunatarajiwa wiki ya pili ya mwezi Mei. Utabiri huu unajumuisha maeneo ya kusini mwa Tanzania ( Nyuzi 10 Kusini, 35 Mashariki) yaani maeneo ya kusini mwa mkoa wa Mbeya, ziwa Malawi, na mkoa wa Songea.

The PIK- monsoon/rainfall season monitor shows how far away we are from the beginning of the rainy season. Daily mean near-surface air temperature till April 25, 2021, for South Tanzania (green) and South Africa (blue). Grey lines are past 5-years average for the same regions. The wet season to retreat between 5-10 May, 2021. The data at the PIK-monitor will be updated weekly. Follow our updates.

Daily mean Relative Humidity till April 25, 2021, for South Tanzania (green) and South Africa (blue). Grey lines are past 5-years average for the same regions. Relative Humidity increase in Tanzania (green) shows the wet season's beginning.

March 26, 2021

The last five years: summary of Indian summer monsoon forecasts (2016-2020)

Here, I present the evidence from observations on long-term forecasts of monsoon onset and withdrawal dates in central India for five years, 2016-2020 (Fig. 1). The forecast is based on teleconnection between the two Tipping elements of the Indian summer monsoon - the Eastern Ghats (EG) and North Pakistan (NP).

The region of the forecasts locates in the central part of India in the area of the Eastern Ghats (EG): namely in the south-eastern part of Maharashtra state and the western part of Chhattisgarh state and the northern part of Telangana state. It is an area around the geographical point (20°N, 80°E) surrounded by a square of 2.5°x2.5° (with an area of approximately 77000 km²). North Pakistan area, namely in North Punjab, where monsoon season ceases to exist, is used for performing forecasts for the EG.

The long-term forecast means 40 days in advance for the onset date, and 70 days in advance for the withdrawal date thus offering the earliest prediction of monsoon timing and the only available withdrawal date in India.

The forecast relays on temperature and relative humidity (from NCEP/NCAR re-analyses data set) in two regions (EG and NP). Figure 1 shows the daily mean near-surface air temperature and relative humidity for each year for EG and NP (presented by red and blue respectively; the past five years' average values for the same regions are presented by violet and grey). The grey vertical dashed lines show real onset and withdrawal dates, which bound the monsoon duration indicated by the light blue transparent rectangle.

As it is clear from Figure 1, the onset day appears when a falling temperature in EG intersects with a rising NP temperature. That happens simultaneously with the exponential growth of relative humidity in EG. Subsequently, the withdrawal day occurs when the temperature in NP crosses the monsoon temperature in EG. It concurs with the drop of relative humidity in EG. It is important to clarify that during monsoon season in EG, relative humidity holds above 80%. In 2019 and 2020, it was an anomaly when the tail of relative humidity after the monsoon appeared due to the winter monsoon's short time penetration to central India. Hence, the relative humidity data presented in the figures verify that onset and withdrawal dates match the rainy season's duration. Thus, the features described above completely confirm our conditions of onset and withdrawal date in Stolbova V. et.al, 2016.

Fig. 1. Evidence from observations on forecasts of monsoon onset and withdrawal dates in central India for five years 2016-2020. The forecast region is the Eastern Ghats (EG), North Pakistan (NP) is used for performing forecast for the EG, shown by the red and blue rectangle in map B). The annual cycles of air temperature and relative humidity for each year for EG (red) and NP (blue) are shown in A, C-F. The light blue transparent rectangle indicates the actual duration of monsoon season in EG.

Table 1 shows the comparison of forecasted dates of onset and withdrawal of the upcoming monsoon with observational dates of real monsoon timing presented in Fig.1. It is clearly seen that from 2016 through 2020, the actual dates of monsoon onset and withdrawal lie in the frame of the forecasting period. Such, the presented observations for all five years follow the forecasts.

It is essential to highlight a considerable interannual variability: for onset date - from 9 to 26 of July, and for withdrawal date - from 3 to 26 of October. Such variability is impossible to capture by numerical simulation. It poses significant challenges for forecasting monsoon timing for meteorological services. However, the Tipping elements approach allows us to perform the long-term forecasts: 40 days in advance for the onset date and 70 days in advance for the withdrawal date.

Table 1. The verification of forecasted dates of onset and withdrawal of monsoon with observational dates from NCEP/NCAR re-analyses data set.

Every forecast was published on the PIK Monsoon web page (see below), PIK-News, in Indian newspapers and other: PIK News (20.04.16), PIK news (27.07.16), PIK news (14.11.16), PIK news (05.08.17), PIK news (18.10.17), PIK news (07.05.18), Sakal Times (15.05.18), Hindustan Times (15.05.2018), The Hindu (20.06.18), Telangana Today (10.06.2018), odishatv.in (31.08.19), DownToEarth (20.09.19), The Wire (30.10.2019), The Hindu Business Line (19.09.19), HindustanTimes (07.06.2019), Der Tagesspiegel (21.07.19), Deutsches Klimaportal (26.06.19), TheWeatherChannel (20.09.2019), pressreader (19.09.19), Telangana Today (07.05.19), PIK-News (12.05.20), DownToEarth (01.09.20), DownToEarth(20.09.20), DownToEarth(28.09.20), TheHindu (14.05.2020), Research in Germany (13.05.20), PIK-News (21.09.20).

Recently, I have been informed by the National Centre for Medium-Range Weather Forecasting (NCMRWF) in Delhi that they performed the examination of onset and withdrawal dates of our forecasts for 20 years (2008-2018) l on Indian reanalysis data sets, and found that our forecasting scheme works well. In 2021, I will use Indian reanalysis data sets. I am very grateful to my colleagues from the NCMRWF for providing me with OLR data for the next forecast.

I am happy to collaborate with my colleagues from the Indian Institute of Tropical Meteorology, the India Meteorological Department, and the National Centre for Medium-Range Weather Forecasting. We are working together to improve the monsoon forecast, deliver an accurate forecast of onset and withdrawal more than one month in advance, and make India more climate-resilient.

December 21, 2020

Successful prediction of rainfall season 2020 in southern Tanzania

The rainfall season in southern Tanzania was late and began after the 10th of December 2020; continuous rainfalls are expected after the 31st of December. On the initial phase, isolated rains alternated with periods of dry spells, as I predicted on the 20th of October, 2020.

Msimu ('seasonal' in the Swahili language) rains in Southern Tanzania start in November and end between April and May of the following year. The words' Msimu' and 'Monsoon' are derived from the Arabic word (/mɒnˈsuːn/).

Because it started late than usual, it wasn't well-organized onset. Slowly but inevitably, Msimu rainfalls are advancing further to the south over South Africa, becoming more intensive and sustainable.

The delayed onset of the Msimu rains shortens the agricultural season and diminishes cereal production prospects. The delay in agricultural activities affects income-earning opportunities for farmers. Thus, the long-term prediction of the timing of the rainy season is desperately needed.

The evidence from observational data

The PIK- monsoon/rainfall season monitor shows how far away we are from the beginning of the rainy season. Daily mean near-surface air temperature till December 18, 2020, for South Tanzania (green) and South Africa (blue). Grey lines are past 5-years average for the same regions. The wet season began after 10 of December 2020 with sporadic rainfall and dry spell in between. The data at the PIK-monitor will be updated weekly. Follow our updates.

Daily mean Relative Humidity till December 18, 2020, for South Tanzania (green) and South Africa (blue). Grey lines are past 5-years average for the same regions. Relative Humidity increase in Tanzania (green) shows the wet season's beginning.

Delayed Msimu rainfalls in Southern Tanzania. Daily Relative Humidity and Wind maps using the NCEP/NCAR data at 111 m/1000hPa. Rain area with high humidity is shown by blue color, dry area with low humidity – by brawn color.

October 21, 2020

A rainfall season is expected to start late than usual in southern Tanzania in Africa

The first long-term forecast for Africa. The rainfall season in southern Tanzania will likely begin after the 10 of December 2020. Isolated rainfall and dry spells are expected between the 10th to the 31st of December. Continuous rainfall is expected after the 31st of December. The region of the forecast is South Tanzania (10°S, 35°E), namely towards the South of Mbeya, Lake Malawi, and Songea regions. Elena Surovyatkina and Nitin Babu performed the first long-term forecast for Africa on the 20th of October, 40 days in advance of the rainfall season.

A study on rainfall season in the Southern Highlands of Tanzania - the grain basket of Tanzania, is part of the EPICC project. The rainy season in this region starts at the end of November and lasts until April. Tanzania's economy depends heavily on agriculture, and the livelihood of the Tanzanian farmer largely depends on the rains. Our monsoon research group offers a unique methodology to identify tipping elements in climate variables, which allows a long-term prediction of the rainy season's timing. A long-term forecast means 40 days in advance for the start of the rainy season and 70 days in advance for its end. Such long-term forecasts aid agriculture planning in consolidating resources and strengthening capacity to respond effectively to disasters such as floods and droughts.

The PIK- monsoon/rainfall season monitor shows how far away we are from the beginning of the rainy season. Daily mean near-surface air temperature till October 19, 2020, for South Tanzania (green) and South Africa (blue). Grey lines are past 5-years average for the same regions. The wet season begins after 10 of December 2020 with sporadic rainfall and dry spell in between until 31st of December, afterwards, continuous rainfalls are expected. The data at the PIK-monitor will be updated weekly. Follow our updates.

November 1, 2020

Successful prediction of monsoon withdrawal 2020

The Southwest monsoon withdrew from Central India on the 13th of October. After that, as I predicted on the 14th of August, 2020, other post-monsoon isolated rainfalls appeared in some parts of central & northwest India.

The evidence from observational data

Daily mean near-surface air temperature in two tipping Elements: Eastern Ghats (EG) and North Pakistan (NP), began to converge at the 28th of September when the IMD declared the beginning of monsoon withdrawal from Northwest India. Climatologically, it takes 23 days for the withdrawal of the monsoon to move from NP to the EG. In 2020, within 15 days, almost the Northern part of India (excluding Bangladesh and the neighborhood) showed an abrupt decrease in precipitation. It happens quickly because the difference in the temperatures in the EG and NP was 4°C lower than average. As I predicted, it took less time when the temperature in North Pakistan reaches the temperature level in the EG. On the 14th of October temperature in North Pakistan departed from Monsoon temperature. Since then shows a persistently cooling trend. Since the temperature gradient has disappeared, there are no conditions for monsoon any longer.

Daily mean near-surface air temperature till October 29, 2020, for the Eastern Ghats (red) and North Pakistan (blue). Violet and grey lines are past 5-years average for the same regions. The forecasted withdrawal dates were from 3^rd to 13^th of October 2020. The real onset date was on the 26ht of June, the real withdrawal date was on the 13th of October.

The data of daily accumulated rainfall provided by the IMD (see Figure below) shows how the area with zero mm precipitation on the 4th of October, spread from Northern to Central India and reached the area of forecast on the 13th of October. After that, mostly in the Southern part of India, the precipitation increased, but in the Northern part, only some isolated rains appeared.

The daily accumulated rainfall provided by the IMD, https://mausam.imd.gov.in/imd_latest/contents/all_india_forcast_bulletin.php . The black number shows only 0 mm precipitation, the red number - more than 0 mm. An oval shows the area with zero mm precipitation during withdrawal of Southwest Monsoon; a rectangle shows the area of the forecast. First row: on the 4th and the 6th of October, the area from Northwest to South India and Sri Lanka exhibits zero-mm precipitation, then on the 13th and the 14th of October, a dry area in Northern part extended significantly. Second row: post-monsoon rains fall mainly south of the forecast area.

After the 14th of October, the smooth withdrawal of monsoon was interrupted by large scale wind pattern which appeared as a result of the seasonal reverse of winds over East Asia and the Pacific Ocean (See Figure below). The Figure shows the wind direction and the speed for the period between the 7th and the 20th of October. On the 7th of October, it shows the Northeast wind pattern along the coast of East Asia, which cross South Asia and penetrates to the Bay of Bengal. It caused rainfall over South part of India and Bangladesh with the neighbourhood (See Figure with Accumulated Rainfall from 16 to 22, October). Notably, that during this period the Southwest monsoon wind, being below 10°N, could not contribute to the precipitation. The invasion from the Northeast was continued for almost three weeks, with 3 days break, and it is still continuing on the 1st of November.

The invasion from the Northeast. Wind direction and the speed for the period between the 7th and the 20th of October. "Warm" colours orange and like red indicate higher wind speeds. Northeast wind pattern originates from East Asia due to the seasonal reverse of winds. The season in East Asia changes, the wind changes direction from the Pacific Ocean to land. Northeast wind going along the coast and then cross South Asia and penetrates to the Bay of Bengal. It produces circulation over the Bay of Bengal and brings precipitation to South India (below 20 N). Southwest Monsoon is below Indian Subcontinent.

Meanwhile, on the 28th of October 2020, the IMD declared the Withdrawal of Southwest Monsoon from the entire country and simultaneous commencement of the Northeast Monsoon rains over extreme south peninsular India. According to the IMD, the Southwest monsoon withdrew from the Northwest (around 23 N) to the South (from the entire country) in 3 days (i.e. from 25 to the 28th of October). That is hard to explain from a physics point of view. Surprisingly, exactly the same days 25-28, the invasion from the Northeast took a break.

The misleading appeared due to fuzzy definition of monsoon withdrawal, mostly in part such as "keeping the spatial continuity, reduction in moisture as seen in the water vapour imageries and dry weather for 5 days". Such a Linear approach to characterise a complex monsoon system is not productive. The Figure with daily accumulated rainfall provided by the IMD shows that it is almost impossible to fulfil this condition. It is why the IMD artificially delayed monsoon withdrawal until 21st of October in Northwest India.

From my point of view, the seasonal reverse of winds over East Asia and the Pacific Ocean marks a withdrawal of Southwest monsoon and beginning of the Northeast Monsoon. Further research needs to define this event. Meanwhile, we have the data, which could help in such intricate case to identify wind pattern.

In my forecast, I rely on thermodynamical variables such as a temperature, relative humidity and its spatial relation. It no needs other interpretation. Hence, it is more reliable.

Finally, I would like to acknowledge the IMD for collecting the weather data, which is the origin for Reanalysis and others, and which I use for my forecast.

August 14, 2020

Forecast of the Withdrawal Date of Indian Summer Monsoon - 2020 from the Central part of India

The Indian Summer Monsoon (Southwest Monsoon) is likely to withdraw from the Central part of India (20N, 80E) between 3rd and 13th of October 2020. Temperature anomaly in North Pakistan shrinks the duration of the monsoon season. Nevertheless, after the monsoon withdrawal date in October, others post-monsoon rainfalls might appear over the central & northwest India.

Such scenarios happened in 1974 and 1975, but that time such a temperature in North Pakistan was not abnormal. The post-monsoon rainfalls were also observed in recent years, in particular, in 2019, 2017 and 2014.

The forecast is based on teleconnection between the Eastern Ghats (EG) and North Pakistan (NP) - Tipping Elements of Indian Summer Monsoon.

The region of this forecast locates in the central part of India in the area of the Eastern Ghats (EG): namely in the south-eastern part of Maharashtra state, and the western part of Chhattisgarh state and the northern part of Telangana state. It is an area around the geographical point (20°N, 80°E) surrounded by a square of 2.5°x2.5° (with an area of approximately 77000 km²). North Pakistan area means namely in North Punjab where monsoon season ceases to exist.

Temperature anomaly in North Pakistan

During the monsoon season, the temperature in North Pakistan is higher than in the Eastern Ghats because of significantly less rainfall and shorter rainy period. While at the end of August, the temperature in the NP decreases, in the EG temperature remains around the same level. When the temperature in North Pakistan reaches the monsoon temperature in the EG, monsoon leaves Indian subcontinent.

In 2020, since April, the daily average temperature in North Pakistan remains less about 4°C less compared to the average five last years. However, in the EG there is no such anomaly. Hence, it takes less time when the temperature in North Pakistan reaches the temperature level in the EG. Thus, the anomaly in North Pakistan might notably shift forward the time of monsoon withdrawal date and shrink the duration of the monsoon season. Nevertheless, after the monsoon withdrawal date, others post-monsoon rainfalls might appear over the central & northwest India.

The PIK- monsoon onset monitor shows how far away we are from the Monsoon withdrawal day. Daily mean near-surface air temperature till August 14, 2020, for the Eastern Ghats (red) and North Pakistan (blue). Violet and gray lines are past 5-years average for the same regions. The forecasted withdrawal dates are from 3^rd to 13^th of October 2020. The data at the PIK-monitor will be updated weekly. Follow our updates.

Successful prediction of monsoon onset

The Northern Limit of Monsoon declared by IMD appeared over EG on 12th of June. In the preceding days, a low-pressure system formed in the east-central Bay of Bengal and moved towards south Odisha. This event resulted in rainfall in many regions in central India, excluding EG region where the relative humidity (RH) had been below 50% at a temperature of 34°C – that means no rain in this area (See Figure below on 14th of June). Although there was a short-lived increase in RH over EG on 16 -17 of June, the RH fell down to 50% in the subsequent 2 days. Then from 17-26 June, a dry spell appeared unexpectedly (See plot on June 22), (https://timesofindia.indiatimes.com/city/pune/dry-spell-for-state-may-end-with-monsoon-revival-by-sat/articleshow/76539081.cms)

In fact, the real monsoon rainfall began over the EG from 26 of June, when relative humidity overcame 80 %- threshold - which is holding around 90% level until present. See the evidence on the plot above, where the period of isolated rain is highlighted by light blue colour, and the period of continuous rainfall - by a darker blue colour. Thus, my prediction of delayed monsoon onset dates between 18th and 26th appears to be correct.

What happened with monsoon advance this year and Why the dry spell surprisingly appeared?

To answer these questions, we analysed the data of daily temperature, relative humidity and wind components at different levels of the atmosphere. We found substantial intrusion of northwesterly mid‐tropospheric dry air from Afghanistan and Pakistan to Northwest and Central part of India between 14th and 26th of June (See Figure below). The mid-level north-westerlies formed a wedge of dry air that was deep in the far northwest of India and over-ran the monsoon flow.
The wind pattern on June 14^th and 16^th (166 DOY) at 600 hPa shows that the northwesterly flow was disrupted towards the southeast and in fact, turned towards the southwest. A low-pressure synoptic event appeared over Bay of Bengal, which moved west-northwestwards, and cyclonic circulation occurred over central India. It happened due to the intensive westerly flow passing over the Arabian Sea towards northeast India (Figure 16 June). However, once the synoptic event passed, the dry northwesterly flow returned to its normal position towards the southeast and dry spell appeared for 10 days (June 16th).

This is evidence that the retreat of the dry winds towards the north had not occurred, and the moist southwesterly/westerly winds had not eroded the dry winds to the extent needed. On June 24th, the dry winds did not extend till southeast India. The near-surface RH at EG was still below 80%. On June 26th, we observe the true retreat of the dry winds away from southeast India and RH increased above 80% at EG. The time series of RH after June 26th shows that the high humidity is sustained.

Based on the above findings, we conclude that the true onset of monsoon at EG occurred after June 26th. According to Parker et al. 2016, synoptic events could influence the moist westerly/southwesterly stream or dry northwesterly stream. The rainfall over central India during the period June 12-14 was a synoptic and temporary event, followed immediately by a dry spell in the region. This low-pressure event temporarily eroded the northwesterly dry stream. However, the conditions for the actual onset where sustained rainfall is possible did not occur until 26th of June.

Figure: The dry spell appearance at monsoon advance 2020 year made monsoon onset in central India considerably delayed. The figure shows the interaction between the low-level monsoon flow at 850 hPa (lower row) and an overrunning intrusion of mid-tropospheric dry air at 600 hPa (upper row). The contour plot represents the relative humidity which is shown by color. In the contour plot, blue indicates moist conditions while red indicates dry conditions. The dashed silver contours represent RH levels above 80 % that correspond to rainy conditions. We have acquired near-surface (1000 hPa) relative humidity and horizontal wind components at 850 hPa and 600 hPa from the NCEP/NCAR reanalysis dataset (Kalnay et al. 1996). The black vectors indicate wind patterns at 850 hPa while the gray vectors represent wind at 600 hPa.

May 8, 2020

Forecast of the Onset date of the Indian Summer Monsoon - 2020 over the central part of India

The Indian Summer Monsoon (the Southwest Monsoon) is likely to set over the central part of India, the Eastern Ghats region (20°N,80°E) between the 18th and 26th of June 2020. The heavy rains are expected to start from the 26th of June.

The region of our forecast is located in the central part of India in the area of the Eastern Ghats (EG), namely in the southeastern part of Maharashtra state, and the western part of Chhattisgarh state and the northern part of Telangana state (the geographical point (20°N, 80°E) surrounded by the square of 2.5°x2.5° with the area of approximately 77000 km²).

Delayed monsoon onset in Central India

Will this year's monsoon bring similar conditions as the last year? Our Monsoon Research Group has already been hard at work, answering this question.

By my observations, the pre-monsoon rainfalls usually disturb the well-organization of monsoon, making it delayed and weaker. Moreover, there is an essential condition for monsoon onset: before monsoon onset, the daily temperature in the Eastern Ghats is decreasing, but in North Pakistan, it is increasing, and at monsoon onset, the temperatures must be equal. Compared to the average five last years, the daily average temperature in the Eastern Ghats in April is near 1°less, but in North Pakistan is about 4°C less. Hence, it takes several days when the temperature in the EG drop-down until the NP-level that notably delays the time of monsoon onset.

The One-Handed Monsoon

India is surrounded by the Bay of Bengal on the east and the Arabian Sea on the west. Summer monsoon arrives in two branches: from the Arabian Sea and from the Bay of Bengal. When both branches are strong, the whole Indian subcontinent receives a good amount of rainfall. However, it is a trend in the last years when the Arabian Sea branch is becoming stronger. In Monsoon 2020 this tendency remains, and the One-Handed Monsoon expected to be again. That means that the western coast of India could receive a lot of rainfall from the Arabian Sea branch, but on the eastern coast might be a deficit of rain. And that, of course, raises the question of how much of a rainfall deficit.

The Deficit of the Monsoon Rainfall

In this year, our Monsoon Research Group has extended the scope of expertise, and we developed a new methodology for forecasting the regional amount of monsoon rainfall for the period from June to September. Here, we present our first test of the upcoming monsoon. According to the estimation of Dr Jingfang Fan, the forecast for the monsoon 2020 is the following:

for All India 842.85 mm (-5% below average),

for Central India - 854.07 mm (-12%),

for East & Northeast India - 1245.54 mm (-13%).

These results indicate, compared to the normal rate, a deficit of the monsoon rainfall. The specific of the upcoming season might be caused in particular, according to Dr Josef Ludescher, by the conditions of the oncoming El Nino at the end of 2020.

While the western coast of India might face the floods, the eastern coast needs to save water. Thus, sharing water is a challenge of upcoming Monsoon 2020.

How far away are we from the Monsoon onset day?

The PIK- monsoon onset monitor shows how far away we are from the Monsoon onset day. Daily mean near-surface air temperature and relative humidity till June 22, 2020, for the Eastern Ghats (red) and North Pakistan (blue). Violet and grey lines are past 5-years average for the same regions. The tipping point (red) indicates the critical temperature and the forecasted onset date. The forecasted onset dates are from 18^th to 26^th of June, 2020. The data at the PIK-monitor will be updated once a week. Follow our updates.

Forecast of onset of Monsoon for the Telangana state

Summer Monsoon in Northern Telangana will likely begin between 16th and 24th of June 2020. Isolated rainfall and dry spells expected between 24th of June to 15th July. Continues rainfall expected between 15th July to 13 October.

Daily mean near-surface air temperature and relative humidity till June 22, 2020, for the Northern part of Telangana state (17.5°N, 80°E). The data at the PIK-monitor for Telangana will be updated twice a week. Follow our updates.

October 17, 2019

Successful forecast of the Withdrawal Date of Indian Summer Monsoon - 2019 from the Central part of India

According to the India Meteorological Department, the Indian Summer Monsoon has withdrawn from the Central part of India (20N, 80E) 14th October 2018 and from the entire country on 16th of October (see the IMD's map below). I issued the forecast in August that it happens between 14th and 24th October 2019. Such my forecast performed 66 days in advance is correct.

The southwest monsoon began its retreat from the boundary with Pakistan 39 days later than the normal date, it was the most delayed monsoon withdrawal recorded before. Nevertheless, then it made quickly retreat from the north to center of India, and it has withdrawn from the Central part of India on the 14th of October as it was predicted. Despite unprecedented monsoon, which broken canonical rules as in advance and at withdrawal as well, both of my forecast of onset and withdrawal dates are correct.

The Map of Withdrawal of Southwest Monsoon across India on October 16th, 2019 provided by the Indian Meteorological Department.
The red lines show an average date of a long history of monsoon timing considered by the IMD as a normal. The green curves indicate the northernmost limits of monsoon (NLM) from which it withdrew on any given day. See the green line - the position of NLM on 14th October, which crossed our region of the forecast. That means the monsoon has withdrawn from the Central part of India on 14 of October as it was forecasted 66 days in advance.

August 9, 2019

Forecast of the Withdrawal Date of Indian Summer Monsoon - 2019 from the Central part of India

The Indian Summer Monsoon (Southwest Monsoon) is likely to withdraw from the Central part of India (20N, 80E) around 19th October (+/- 5 days), namely between 14th and 24th October 2019.

The region of this forecast locates in the central part of India in the area of the Eastern Ghats (EG): namely in the south-eastern part of Maharashtra state, and the western part of Chhattisgarh state and the northern part of Telangana state. This is an area around the geographical point (20°N, 80°E) surrounded by a square of 2.5°x2.5° with an area of approximately 77000 km².

This forecast made for 70 days in advance is the only one available in India. Last years during withdrawal of monsoon in the middle of October, the severe cyclonic storm led to evacuation hundred thousand people, devastating rainfall triggered dams spilling and flood. Such a long-term forecast could help the Government to do strategic planning, consolidate resources, and strengthen capacity to respond effectively to disasters.

How far away are we from the Monsoon withdrawal day?

The PIK- monsoon onset monitor shows how far away we are from the Monsoon withdrawal day. Daily mean near-surface air temperature till October 14, 2019, for the Eastern Ghats (red) and North Pakistan (blue). Violet and gray lines are past 5-years average for the same regions. The forecasted withdrawal dates are from 14^th to 24^th of October 2019. The data at the PIK-monitor will be updated weekly. Follow our updates.

June 26, 2019

Successful forecast of the Onset Date of Indian Summer Monsoon - 2019 for the Central part of India

The Indian Summer Monsoon (the Southwest Monsoon) neared the Eastern Ghats (20N, 80 E) on 18th of June, and then on 19th of June rainfall started. On the 20th of June, the intensity of rainfall significantly increased indicating the established monsoon in this area. My prediction made 40 days in advance for the monsoon onset date was between 10 -18 of June.

According to the map of Advance of Southwest Monsoon provided by the Indian Meteorological Department, the boundary identified by fact after 48 hours-rainfall along the country from south-west to north-east, so-called the Northern Limit of Monsoon (NLM), is located in the Eastern Ghats region between 21th -22nd of June (See the map below).

I want to clarify that I am not forecasting when the first intensive rain will start. I predict the date when the daily mean temperature crosses a certain threshold. For the Eastern Ghats, under Climate Change conditions, it is currently 36 degree Celsius, but for example, in the 90s it was 3 degrees less. Beyond that limit a transition to monsoon becomes inevitable. While in some years the transition from pre-monsoon to monsoon was fast, this year it was slow. My prediction for the monsoon onset date was between 10 -18 of June. The daily mean temperature crosses a 36-degree Celsius-threshold on 17th of June. After the disappearance of cyclone 'Vayu', it took some time for the monsoon system to accumulate sufficient moisture again. Finally, the monsoon came after crossing the temperature threshold. Please find the evidence on the maps provided by the Indian Meteorological Department (the IMD) below.

The daily forecast over India on June 16-20 provided by the Indian Meteorological Department

The Map of daily rainfall over India on June 16-21 provided by the Indian Meteorological Department. The color bar indicates the amount of daily rainfall in mm. The region of the forecast is marked by a box. On the 19th of June, rainfall is observed in parts of Odisha and Chhattisgarh. Rainfall is observed within the region of forecast on 19th and 21st of June indicating that monsoon has arrived here.

Monsoon under Climate Change

In the pre-monsoon period, on May 22, the temperature in Vidarbha in Central India has reached 47.8 degree Celsius. This is the region of my forecast - the hottest place in India. Please note, the Vidarbha is not a desert; it is a highly populated area in the Eastern Ghats.

Then on 12th June 2019, Kuwait and Saudi Arabia record highest temperature on earth reaching 52.2 degrees Celsius in the shadows and 63 degrees Celsius under direct sunlight ( https://gulfnews.com/world/gulf/kuwait-and-saudi-arabia-record-highest-temperature-on-earth-1.1560325581417).

Monsoon started under such overheating conditions resulting in severe cyclone 'Vayu'. It was the completely disorganized beginning of the monsoon. Instead of covering whole Indian subcontinent gradually by intensive showers, monsoon this year resembles a leaky blanket. With the severe cyclone 'Vayu' on the West coast, causing the evacuation of three million people, the rains on the remaining part of the country were very local, sudden and in some places very intense, but went away after one downpour. Even though large parts of the country received rainfall, distribution was non-uniform and the amount of rainfall was less than certain criterion based on the average amount of rainfall in the past 30 years. This is why, within ten days after the beginning of Monsoon on the South, the rainfall over almost the entire country was officially declared as pre-monsoon rainfall. Usually, pre-monsoon rainfall is very local, but the pre-monsoon rainfall on the scale of the country is unprecedented. Moreover, intensive rainfall in North India started even before the monsoon in the central part of India. That is breaking the well-established rules of monsoon.

Monsoon is changing with climate change. It requires to refine the criteria of monsoon onset definition accordingly. Climatological norms, which are a 30-year average of a weather variable, must be reconsidered in the context of climate change.

The Map of Advance of Southwest Monsoon across India on June 18-th provided by the Indian Meteorological Department
The red lines show an average date of a long history of monsoon timing considered by the IMD as a normal. The green curves indicate the northernmost limits of monsoon (NLM) up to which it has advanced on any given day. See the last green line - the position of NLM on 21th-22th June, which crossed my region of the forecast, that means the monsoon rainfall in this area continued 48 hours already and monsoon advanced up to the central part of India.

May 7, 2019

Forecast of the Onset date of the Indian Summer Monsoon - 2019 over the central part of India

The Indian Summer Monsoon (the Southwest Monsoon) is likely to set over the central part of India, the Eastern Ghats region (20°N,80°E) between the 10th and 18th of June 2019.

The region of our forecast is located in the central part of India in the area of the Eastern Ghats (EG), namely in the southeastern part of Maharashtra state, and the western part of Chhattisgarh state and the northern part of Telangana state (the geographical point (20°N, 80°E) surrounded by the square of 2.5°x2.5° with the area of approximately 77000 km²).

How far away are we from the Monsoon onset day?

The PIK- monsoon onset monitor shows how far away we are from the Monsoon onset day. Daily mean near-surface air temperature till June 24, 2019, for the Eastern Ghats (red) and North Pakistan (blue). Violet and gray lines are past 5-years average for the same regions. The tipping point (red) indicates the critical temperature and the forecasted onset date. The forecasted onset dates are from 10^th to 18^th of June, 2019. The data at the PIK-monitor will be updated twice a week. Follow our updates.

Forecast of onset of Monsoon for the Telangana state

Since 2016, I issue such a forecast for the central part of India in the area of the Eastern Ghats (EG). In 2019, I start to test the forecast for the Telangana state (17.5°N, 80°E), namely towards the East of Hyderabad, south-east of Warangal, west of the Godavari river and north of Khammam. Telangana is the 'Seed Bowl' of India; monsoon forecast is vital for the state. Export of seed can create economic opportunities for small and marginal farmers.

I estimate the date of monsoon onset for Telangana as between the 8th and 16th of June 2019.

The data at the PIK-monitor for Telangana will be updated twice a week. Follow our updates.

The Map of daily rainfall over India on June 16-18 provided by the Indian Meteorological Department. The color bar indicates the amount of daily rainfall in mm. The region of the forecast is marked by a box. On 16th of June, rainfall is observed over Odisha and parts of Chhattisgarh, which are to the right of the box. The appearance of green patches within the square on 17th and 18th of June indicates that the monsoon has started in the region of Telangana.

Even so, those two regions, Eastern Ghats and Telangana, are neighbouring; there is a considerable difference in the interplay of the two branches of Monsoon (one from the Arabian Sea and another from the Bay of Bengal) in these regions. In some years, monsoon onset appears at the same date in these two areas, while in other years it appears - with a difference of 6-10 days. Such a difference might be caused by the simultaneous appearance of two cyclones in the Arabian Sea and in the Bay of Bengal, which prevent the advance of monsoon for several days. Another reason might be an anticyclone on the north of the Arabian Sea, which decelerates the turning of the Monsoon winds in the Bay of Bengal toward Indian subcontinent. These synoptic events are impossible to predict well in advance and it could be forecasted near the date of its appearance only. This is why the forecast in Telangana region is challenging. In such a case the Indian Meteorological Department track cyclones perfectly that helps me to update a monsoon onset date in June, on the eve of monsoon. We are collaborating with the Indian Meteorological Department and our forecasts complement each other. We are working together to improve climate resilience by building new capacities via the early forecast of monsoon.

October 21, 2018

Successful earliest forecast of the Withdrawal Date of Indian Summer Monsoon - 2018 from the Central part of India

The Indian Summer Monsoon (the Southwest Monsoon) has withdrawn from the Central part of India (20N, 80E) 18th - 21st October 2018. I issued our forecast on July 30, 2018, that the monsoon will withdraw from the Central part of India around 18th of October (+/- 5 days), namely between 13rd and 23rd of October 2018. Hence, my earliest forecast of the withdrawal date made 80 days in advance was correct.

What is important to note is that my forecast of monsoon withdrawal date is the only one available in India. In 2016 and 2017, I issued successful predictions around 70 days in advance. In 2018, for the first time, I performed a successful forecast 80 days in advance.

How did the monsoon withdrawal in 2018?

In the first decade of October, namely between 2nd-6th of October, it seemed that the rainy season is over for whole Northern part of India, and the Monsoon quickly left in the English way without saying goodbye. This is why the withdrawal line appeared on the Map of Withdrawal Southwest of Monsoon at 16°N on October 6 (http://www.imd.gov.in/pages/monsoon_main.php). Does it mean, predicting withdrawal between 13 and 23 of October, I have missed?

In fact, it was the calm before a storm or dry spell in Monsoon how it is called in India. Starting from October 7, Monsoon came back to Northeast part of India with Very Severe Cyclonic Storm Titli; widespread rains hit Odisha and Andhra Pradesh, districts, killing nine people, on October 11. About 300000 people were evacuated from five coastal districts. The Heavy Rainfall and Wind Warnings were continued until 14th of October and then isolated rains in Central part of India until 18th of October.

According to the IMD, "Southwest Monsoon has withdrawn from the remaining parts of the country and thus from the entire country today, the 21st October 2018."

Thus, my prediction for very late withdrawal of Monsoon between 13 and 23 of October was correct.

Please find the evidence on the maps provided by the Indian Meteorological Department (the IMD) below.

Summarising results of my tests of forecasting Monsoon onset and withdrawal for the period from 2016 to 2018, I emphasize for the three years in a row; all six forecasts were successful.

July 30, 2018
Earliest Forecast of the Withdrawal Date of Indian Summer Monsoon - 2018 from the Central part of India.

The Indian Summer Monsoon (Southwest Monsoon) is likely (with an 84% probability) to withdraw from the Central part of India (20N, 80E) around 18th October (+/- 5 days), namely between 13th and 23rd October 2018.

The region of our forecast locates in the central part of India in the area of the Easter Ghats (EG), namely in the southeastern part of Maharashtra state, and the western part of Chhattisgarh state and the northern part of Telangana state (the geographical point (20°N, 80°E) surrounded by the square of 2.5°x2.5° with the area of approximately 77000 km²).

Our earliest forecast (70 days in advance) is the only one available the Central part of India. Last years severe and devastating rainfall poured down on the eve of withdrawal of monsoon, triggering dams spilling and floods. Our long-range forecast of the withdrawal date of monsoon rainfall is crucial for taking appropriate decisions for as for local and central Government as well as farmers, sugar companies, managing water and energy resources, and other stakeholders.

The PIK- monsoon withdrawal monitor shows how far away we are from the Monsoon withdrawal day. Daily mean near-surface air temperature till December 30, 2018, for the Eastern Ghats (red) and North Pakistan (blue). Violet and gray lines are past 5-years average for the same regions. The tipping point (red) indicates the critical temperature and the forecasted withdrawal date. The forecasted withdrawal dates are from 13^th to 23rd October. The data at the PIK-monitor will be updated twice a week. Follow our updates.

June 19, 2018

Successful earliest forecast of the Onset date of the Indian Summer Monsoon - 2018 over the central part of India

The Indian Summer Monsoon (the Southwest Monsoon) neared the Eastern Ghats (20N, 80 E) on 9th of June, and then it embraced the region of forecast on 12th of June. We made our prediction on May 7, 2018, that the monsoon will set over central India on 15th June (+/- 4 days). Hence, our prediction made 40 days in advance was correct.

After 12th of June, there is no further advance in internal part of the Indian subcontinent during the next week until today, 19th June, due to (i) a weakening of the monsoon circulation over the Bay of Bengal, (ii) local overheating of the Eastern Ghats (20N,80 E) region, where temperature reached 44°C that is linked to global warming, and (ii) premonsoon rainfall 7- 8 June appeared due to overheating.

According to the map of Advance of Southwest Monsoon on 19 June 2018 provided by the Indian Meteorological Department, the Northern Limit of Monsoon (NLM) is located in the Eastern Ghats region during 9th -19th of June (See the map below).

Our next forecast for the monsoon withdrawal date from the central part of India will be issued on July 27th, 2018. The PIK-monitor will be updated twice a week to show how far away we are from the Monsoon withdrawal day. Follow our updates.

The Map of Advance of Southwest Monsoon across India on June 19th provided by the Indian Meteorological Department.
The red lines show an average date of a long history of monsoon timing considered by the IMD as a normal. The green curves indicate the northernmost limits of monsoon (NLM) from which it withdrew on any given day. See the last green line - the position of NLM on 9th-19th October, which crossed our region of the forecast. That means the monsoon has withdrawn from the Central part of India.

May 07, 2018
Forecast of the Onset date of the Indian Summer Monsoon - 2018 over the central part of India

The Indian Summer Monsoon (the Southwest Monsoon) is likely to set over the central part of India, the Eastern Ghats region (20°N,80°E) around 15th June (+/- 4 days) namely between 11th to 19th June 2018.

The region of our forecast locates in the central part of India in the area of the Eastern Ghats (EG), namely in the southeastern part of Maharashtra state, and the western part of Chhattisgarh state and the northern part of Telangana state (the geographical point (20°N, 80°E) surrounded by the square of 2.5°x2.5° with the area of approximately 77000 km²).

October 16, 2017

Successful earliest forecast of the Withdrawal Date of Indian Summer Monsoon from the Central part of India.

The Indian Summer Monsoon (the Southwest Monsoon) has withdrawn from the Central part of India (20N, 80E) 15th-16th October 2017. We issued our forecast on July 30, 2017, that the monsoon will withdraw from the Central part of India around 12th October (+/- 5 days), namely between 7th and 17th October 2017. Hence, our earliest forecast of the withdrawal date made 70 days in advance was correct.

The withdrawal of monsoon from northwest India normally starts on September 1. However, the 2017 monsoon withdrawal was delayed for nearly 3 weeks and commenced on 27th September. Nevertheless, the monsoon withdrawal reached the region of our forecast on 15th October, and then it passed the Eastern Ghats region on 16th October, i.e. within our prediction interval. It was confirmed today by the information provided by the Indian Meteorological Department (the IMD), it declared that the monsoon withdrawal line passes through the region of our forecast (the Eastern Ghats, 20N, 80E) 15th -16th October 2017. Please find the evidence on the map below.

The Map of Withdrawal of Southwest Monsoon across India on October 16th provided by the Indian Meteorological Department.
The red lines show an average date of a long history of monsoon timing considered by the IMD as a normal. The green curves indicate the northernmost limits of monsoon (NLM) from which it withdrew on any given day. See the last green line - the position of NLM on 15th-16th October, which crossed our region of the forecast. That means the monsoon has withdrawn from the Central part of India.

July 30, 2017
Earliest Forecast of the Withdrawal Date of Indian Summer Monsoon - 2017 from the Central part of India.

The Indian Summer Monsoon (Southwest Monsoon) is likely (with an 84% probability) to withdraw from the Central part of India (20N, 80E) around 12th October (+/- 5 days), namely between 7th and 17th October 2017.

The region of our forecast locates in the central part of India in the area of the Eastern Ghats (EG), namely in the southeastern part of Maharashtra state, and the western part of Chhattisgarh state and the northern part of Telangana state (the geographical point (20°N, 80°E) surrounded by the square of 2.5°x2.5° with the area of approximately 77000 km²).

Our earliest forecast (70 days in advance) is the only one available the Central part of India. Our goal is to provide farmers, agriculture companies, managing water, and energy resources and other stakeholders from national to local levels with the information on the monsoon withdrawal dates with the aim of minimizing regional risks of the Indian population vulnerability from climate-related phenomena.

June 18, 2017
Successful earliest forecast of the onset of Southwest Monsoon 2017 over the central part of India

The Indian Summer Monsoon (the Southwest Monsoon) has set in over the central part of India, the Eastern Ghats region (20°N,80°E) 16-th June 2017. We made our prediction on May 8-th 2017 that the monsoon will set over the central India on 18-th June (+/- 4 days). Hence, our prediction made 40 days in advance was correct.

In fact, the monsoon neared the region of forecast on 14-th of June, and then it approached the Eastern Ghats region on 16-th of June. The Indian Meteorological Department (the IMD) declared that the Northern Limit of Monsoon (NLM) is passing through the Eastern Ghats region during 16-th -18-th of June (See the map below).

Our next forecast for the monsoon withdrawal date from the central part of India will be issued on July 27-th. The PIK-monitor will be updated twice a week to show how far away we are from the Monsoon withdrawal day. Follow our updates.

The Map of Advance of Southwest Monsoon across India on June 18-th provided by the Indian Meteorological Department and updated on the progress of monsoon.
The red lines show an average date of a long history of monsoon timing considered by the IMD as a normal. The green curves indicate the northern most limits of monsoon (NLM) up to which it has advanced on any given day. See the last green line - the position of NLM on 16th-18th June, which crossed our region of the forecast. That means the monsoon advanced up to the central part of India.

May 08, 2017
Forecast of the Onset date of the Indian Summer Monsoon - 2017 over the central part of India

When do we issue our forecasts?

This test forecasting has started in May 2016; the forecast will be updated yearly before and during the Monsoon season. Normally, a forecast for the onset of Monsoon will be made available on May 8-th. A forecast for the monsoon withdrawal will be issued on July 27-th.

We’ve created this website to provide stakeholders on national, regional and local levels with the information on the monsoon timing with the aim to minimizing the regional risk of the Indian population vulnerability from climate-related phenomena.

This page is also intended for researchers who interested in a monsoon forecast, who use forecasts in their research, or who wish to learn more about current efforts to improve the forecasting of the monsoon timing. Here we also provide the results of forecasts of the previous years (see also press releases and news from 2016).

Why is a forecast of monsoon rainfall so crucial for India?

The economy of India is able to maintain its GDP in the wake of a good monsoon. However, if monsoon gets delayed by even two weeks, it can spell disaster because of the high population depending on agriculture - 70% of its people directly related to farming. Agriculture, in turn, is dependent on the monsoon.

The long-range forecast of the onset and withdrawal dates of monsoon rainfall is critical for taking appropriate decisions at various levels from farmer’s field (e.g. a choice of plowing and seeding days) to the central government (e.g. managing water and energy resources, food procurement policies and trade etc.).

Although the rainy season happens annually between June and September, the time of monsoon season’s onset and withdrawal varies within a month from year to year. The important feature of the monsoon is that it starts and ends suddenly. Hence, despite enormous progress having been made in predicting monsoon since 1886, it remains a significant scientific challenge.

How do we make predictions?

To make predictions of monsoon timing, we apply our recently developed method [Stolbova V. et.al, 2016] which focuses on Tipping elements of the Indian monsoon - the Eastern Ghats (EG) and North Pakistan (NP).

Our prediction relies on observations of near-surface air temperature and relative humidity from both the ERA-40, the European Center for Medium-range Weather Forecasts (ECMWF) and NCEP Reanalysis data provided by the NOAA/OAR/ESRL PSD, Boulder, Colorado, USA.

Important to note that the Indian Meteorological Department (the IMD) forecasts monsoon two weeks in advance and only for Kerala state on the southern tip of India and not for the other 28 states of the country, and the IMD does not forecast withdrawal date. Our study concerns the central part of India. We performed both of our forecasts for the onset and withdrawal of monsoon for the region of the Eastern Ghats (20N,80E) where the prediction has never been made.

We were inspired by the paper of Ananthakrishanan and Soman [R. Ananthakrishnan and M.K. Soman, 1990], where the authors defined the onset of monsoon as follows: "it is not a transition from a regime of no rain to rain; it is a transition from a sporadic rainfall to a spatially organized and temporally sustained rainfall…».
In our study, we have found the evidence in observational data that we can consider the onset of monsoon as a critical transition - a sudden transition to the monsoon when critical thresholds (in particular, in near-surface air temperature, relative humidity) are reached.

If the onset of monsoon is a critical transition, then critical phenomena can be detected. In our study, we used the phenomenon of growth of fluctuations [E. Surovyatkina et al.,2015] on the eve of the transition as a precursor of a sudden transition to the monsoon.

In contrast to traditional approaches which use precursors for prediction of the time of the critical transition (that work only retrospectively, [Kefi et al.,2014]), we discovered how to use precursors in a new way – to find regions where critical conditions for an occurrence of the Indian monsoon originate.

We revealed two geographic regions with highest fluctuations in the near-surface air temperature on the eve of the monsoon - the Eastern Ghats (EG) and North Pakistan (NP). The highest fluctuations in EG and NP show upcoming instabilities (a cyclone and anticyclone). Highly developed instability occurring in these regions create necessary conditions for the spatially organized and temporally sustained monsoon rainfall. These factors allow us to identify EG and NP as tipping elements of the monsoon.

Moreover, we found that on the eve of the onset and the withdrawal of monsoon in the central part of India the temperature and relative humidity in two tipping elements equalize. We use these findings making predictions of monsoon timing.

Results of our forecasts of 2016

The prior knowledge of dates of onset and withdrawal of monsoon is of vital importance for the population of the Indian subcontinent. In May 2016 before monsoon season, India recorded its highest-ever temperature of 51C. Hot waves have decimated crops, killed livestock and left 330 million people without enough water. At the end of monsoon season the floods in Indian this year have also broken previous records. Severe and devastating rainfall poured down, triggering dams spilling and floods. Such extreme conditions pose the vital questions such as: When will the monsoon come? When will the monsoon withdraw?

Daily Maps provided by the Indian Meteorological Department,

The center of the region of our forecast (20°N,80°E) is indicated by the red cross.

We predicted the monsoon arrival to the Eastern Ghats (20N,80E) on the 13th of June with a deviation of +/-4 days. The prediction was made on May 6-th, 2016, that is 40 days in advance of the date of the forecast. The actual monsoon arrival was June 17-th. In this day near-surface air temperature and relative humidity overcame the critical values and the monsoon season started, that was confirmed by observations of meteorological stations located around the EG-region.

We forecasted the monsoon withdrawal from the Eastern Ghats on the 5th of October with a deviation of +/-5 days. We delivered this prediction on July 27-th, 2016, namely 70 days in advance. The actual monsoon withdrawal started on October 10-th when the relative humidity in the region started to decrease, then it passed the 80 percent threshold, and a transition back to a monsoon became impossible, meteorological stations registered it also, and on October 12-th meteorological stations reported 'No rain' in the EG and also in areas located across the subcontinent in the direction from the North Pakistan to the Bay of Bengal. Hence, the date of monsoon withdrawal - October 10-th, predicted 70 days in advance, lies within our prediction interval.

We emphasize that our forecasts of the monsoon onset and withdrawal were delivered for 40 and 70 days in advance respectively, and both of our forecasts lie within our prediction interval.

Our results show that our method allows predicting the monsoon not only retrospectively but in the future. In 2016 we predicted of the onset and withdrawal dates of the Southwest monsoon over the Eastern Ghats region in Central India for 40 and 70 days in advance respectively. Hence, in 2016 we proved that such early prediction of the monsoon timing is possible.

Article: Stolbova, V., E. Surovyatkina, B. Bookhagen, and J. Kurths (2016): Tipping elements of the Indian monsoon: Prediction of onset and withdrawal. Geophys. Res. Lett., 43, 1–9 [doi:10.1002/2016GL068392]

Weblink to the article: http://onlinelibrary.wiley.com/doi/10.1002/2016GL068392/full

Data sources

Prof. Dr. Dr.Sc Elena Surovyatkina is the group leader of the monsoon research within the EPICC project. This project is part of the International Climate Initiative (IKI). The Federal Ministry for the Environment, Nature Conservation and Nuclear Safety (BMU) supports this initiative on the basis of a decision adopted by the German Bundestag.

Inquiries:

Elena Surovyatkina,

Potsdam Institute for Climate Impact Research (PIK), Complexity Science: Machine Learning, Nonlinear Methods and Decision Strategies

elena.surovyatkina[at]pik-potsdam.de

The Indian Summer Monsoon (the Southwest Monsoon) is likely (with a 73% probability) to set over the central part of India, the Eastern Ghats region (20°N,80°E) on or around 18th June (+/- 4 days).

The region of our forecast locates in the central part of India in the area of the Eastern Ghats (EG), namely in the southeastern part of Maharashtra state, and the western part of Chhattisgarh state and the northern part of Telangana state (the geographical point (20°N,80°E) surrounded by the square of 2.5°x2.5° with the area of approximately 77000 km²).

How far away are we from the Monsoon withdrawal day?

The PIK- monsoon onset monitor shows how far away we are from the Monsoon withdrawal day. Daily mean near-surface air temperature till October 22, 2017, for the Eastern Ghats (red) and North Pakistan (blue).Violet and gray lines- past 5-years average for same regions.The tipping point (red) indicates the critical temperature and the forecasted onset date. The forecasted withdrawal dates are from 7th to 17th October. The data at the PIK-monitor will be updated twice a week. Follow our updates.

In our definition, the onset of monsoon in a particular region of the Indian subcontinent is a date when daily mean values of near-surface air temperature and relative humidity overcome the tipping point. After this date, the temperature falls abruptly, humidity growth rapidly and then in at least two days monsoon rainfall inevitably come. The tipping point means a critical threshold that we estimated for daily mean near-surface air temperature to be 35.4C and for daily mean relative humidity - 36.5% for the region of our forecast in the current global warming conditions.

Described above conditions cannot appear at a bogus monsoon, therefore our forecast excludes a bogus onset. Nevertheless, if a bogus onset appears prior the forecasted date then the monsoon onset date will shift for the duration of a bogus monsoon. In this case, we will update our forecast.

The data at the PIK- monsoon onset monitor will be updated twice a week. Follow our updates.

The Indian Summer Monsoon (the Southwest Monsoon) is likely (with a 73% probability) to set over the central part of India, the Eastern Ghats region (20°N,80°E) on or around 18th June (+/- 4 days).

The region of our forecast locates in the central part of India in the area of the Eastern Ghats (EG), namely in the southeastern part of Maharashtra state, and the western part of Chhattisgarh state and the northern part of Telangana state (the geographical point (20°N,80°E) surrounded by the square of 2.5°x2.5° with the area of approximately 77000 km²).

How far away are we from the Monsoon withdrawal day?

The PIK- monsoon onset monitor shows how far away we are from the Monsoon withdrawal day. Daily mean near-surface air temperature till October 22, 2017, for the Eastern Ghats (red) and North Pakistan (blue).Violet and gray lines- past 5-years average for same regions.The tipping point (red) indicates the critical temperature and the forecasted onset date. The forecasted withdrawal dates are from 7th to 17th October. The data at the PIK-monitor will be updated twice a week. Follow our updates.

In our definition, the onset of monsoon in a particular region of the Indian subcontinent is a date when daily mean values of near-surface air temperature and relative humidity overcome the tipping point. After this date, the temperature falls abruptly, humidity growth rapidly and then in at least two days monsoon rainfall inevitably come. The tipping point means a critical threshold that we estimated for daily mean near-surface air temperature to be 35.4C and for daily mean relative humidity - 36.5% for the region of our forecast in the current global warming conditions.

Described above conditions cannot appear at a bogus monsoon, therefore our forecast excludes a bogus onset. Nevertheless, if a bogus onset appears prior the forecasted date then the monsoon onset date will shift for the duration of a bogus monsoon. In this case, we will update our forecast.

The data at the PIK- monsoon onset monitor will be updated twice a week. Follow our updates.