Page 3 of 5

A Citizen Weather Network

by Saurabh Chandra

We often wonder about how it sometimes rains in one part of Bangalore but not in another or how the temperature drops a few degrees once you cross the Hebbal flyover. Thus, we all intuitively talk about the microclimates that exist in the city but the weather data that we get to see online or on TV is from just a couple of weather stations that the Indian Meteorological Department (IMD) runs in the city.

The Citizen Weather Network initiative aims for this simple never-done-before objective – capture real-time weather data from 25+ locations in the city and make it available for free over a web API. We believe it will give us fresh insights into our climate and allow us to trace microclimate changes over time.

We started out on this goal simply thinking that we should be able to buy some simple weather stations from abroad, convince fellow citizens to buy them and pool in the data online. However, this simple objective led us into a not-so-trivial chain of events. A subsequent post will detail the factors that we considered, and how in conclusion we ended up designing our own Automated Weather Station (AWS) in collaboration with a local startup.

The weather station will collect and publish real-time weather data onto a cloud server, where we can aggregate and do some useful analysis with it. Or we can simply expose it for someone to pull it and do something creative instead. We are almost ready to deploy the initial few units of the weather stations in Bangalore and will be publishing results of the field trial. Hopefully, the results will be encouraging enough for more volunteers to chip in and install this network of stations to create the data density we wish to achieve.

We also hope that this is a useful community project that will involve and promote citizen science. Weather is an amazing topic that brings together science, engineering, commerce together with the most mundane question – do I need an umbrella today or a most existential one – is it getting warmer by the day. A station at our home, apartment complex, institution or school will create a rare opportunity to associate around a community initiative centred around science.

We are excited and nervous. Whatever be the outcome, we will be wiser about our weather (amongst other things) in this endeavour. Stay tuned, and join us to form India’s first Citizen Weather Network in the city of Bangalore.

1. Frankly, even if IMD had more weather stations the data being recorded would have had the same fate as those in the existing stations around India – it would locked up in myriad formats and not be available through an API for citizens to use. The Karnataka Government has made a great effort in this front by setting up the Karnataka State Disaster Management Centre, which has deployed about a 100 weather stations in Bangalore. Unfortunately, the data produced from them is still not available for the public.

2. Pavan ran an interesting social experiment last April on trying to map rain in the city by crowdsourcing information.

Saurabh Chandra is a tech entrepreneur based in Bangalore and a weather enthusiast.

Internships at Know Your Climate

A few of us at Know Your Climate are in the early stages of working on a pilot project to build an urban Automated Weather Station (AWS) and we are offering internships to interested undergraduate and graduate students.

Urban India needs a high density and high quality weather station network that is currently missing in India. As an early post on this blog shows, weather can vary significantly across a city like Bangalore, and perhaps even climate. The current weather infrastructure is unable to capture this complexity.

We invite applications from those would like to apply their programming and analytical skills on real-life challenges. Coding for AWS is a great way to practice your programming skills and, at the same time, help make the blue prints of a weather station for your community. This is a first of its kind project in India.

We are looking for people familiar with javascript and plotting toolkits like D3 and rickshaw 3D. You will create the web front end for real-time streaming and visualization of weather data.

What we expect of you:
* Become familiar with the aws project
* Learn about real-time streaming of data
* Learn git and how to work with github
* Learn D3 and rickshaw toolkit.
* Contribute to discussions and documents

What you should expect
* You will have fun!
* working on a real life problem
* your work will really have an impact
* You will have a clearly defined mentor for technical discussions

The internships are based in Bangalore, India and applications are being accepted on a rolling basis. Send an email to know (dot) your (dot) climate (at) gmail (dot) com with a résumé and a covering letter if you wish to work with us.

Not every disaster is man-made

By Pavan Srinath

Uttarakhand has been a scene of unfolding horror for the past four days, and is a human tragedy occuring at a scale that is staggering. For many people in India, it is also a disaster that hits home – as millions have visited Uttarakhand on pilgrimage and have seen the places that we now see on the television with dread.

The scale of damage due to floods is not yet known, but is certainly immense. The loss of human lives above all, and the destruction of public and private property will likely haunt the residents for many years. The loss of lives, currently estimated in the hundreds – can go up to the thousands or even more, given the large number of people currently reported as missing. A disaster such as this requires rapid, thorough rescue and relief operations, of which by all counts the army and the state officials are doing an admirable job. Thereafter comes time for rebuilding and sombre reflection, as well as thorough investigations into the causes for the disaster, the amplifiers, and the role of human error, malfeasance and failures.

What do we have instead? Loud war cries that the disaster in Uttarakhand was man-made, and that political parties gave in to various mafias and increased the scale of destruction unleashed upon much of Uttarakhand.

One human factor that can be brough into this discussion as a causative agent is climate change, but only with great care. While anthropogenic climate change has been established as a very likely cause for the increase in the frequency and intensity of extreme weather events in India and elsewhere in the world, there are two strong caveats to this link. First, it is impossible to say whether an individual event has a greenhouse gas or a warm climate footprint. This is the case for everything from Hurricane Sandy to the cloudburst over Uttarakhand. Second, empirical evidence for the relationship between the monsoon and climate change is still very limited. There are many theories on what climate change is likely to do to the Indian summer monsoon, but much of it is still unknown. While the summer monsoon hit the coast of Kerala around the usual date this year, its march over the long leagues from Kanya Kumari to the Himalayas was exceptionally quick. The most honest, if uncomfortable, statement is that we don’t know if climate change caused the cloudburst over Uttarakhand, nor do we know that climate change could make such events more frequent or intense.

The reasons for declaring the disaster as man-made were given in a Down to Earth home page feature as the increase in hydel projects in the state, roads and infrastructure destabilising the mountains, and development increasing the frequency and intensity of landslides.

Is any of this true? On the first count of hydroelectric power projects and excessive dam-building in Uttarakhand, the reality is far from the rhetoric. While it is true that there are ambitious plans for dam construction in the state, especially on the Ganga and its tributaries, very few projects have actually been implemented and are operational. The map below from SANDRP shows that on the Ganga, only 16 hydel projects had been commissioned, 13 were under construction, and 54 were proposed as of a year or two ago. The picture has not changed rapidly since then. We can do better than blaming widespread floods on paper dams.



On all other counts of “development” causing or worsening the disaster, the litmus test is the impact at Kedarnath. The holy pilgrimage site of Kedarnath is a valley on the banks of the river Mandakini that lies high above much of the upper Gangetic basin at 3600 metres above sea-level [See Kedarnath on Google Maps]. Above it is wildnerness and  inhabitable mountains, and motorable roads are yet to reach the place. Pilgrims drive up to Gauri Kund, and trek up the last 14 kilometres, climbing some six thousand feet in the process. There are no roads, bridges or extensive artificial interventions around Kedarnath, except for the temple and surrounding hotels and housing that has sprung up.

In spite of this, Kedarnath has been among the worst hit areas in this disaster. Floodwaters swept into the settlement, bringing with them vast amounts of debris and cutting off access for about 8,000 people from the rest of the region.

We have to live in an evidence-free world to say that the horrific natural disaster that struck Kedarnath was man-made. Kedarnath, as the map below shows, lies high above even proposed dams and has only the most minimal amounts of development. It is the benchmark by which one can say that the flooding in Uttarakhand has been more prolific than any other in living memory, above and beyond any “man-made” effects.



All this has been said in full recognition of the fact that Uttarakhand has always been profoundly vulnerable to flooding, and that there has always been a high risk of natural disasters. The notion that such floods could happen some day was far from unknown. The hope that it may not happen to us or in our lifetimes as free of evidence as some of the claims I mentioned above. Places between Rudraprayag and Rishikesh on the Ganga have evidently not built any resilience against an event such as this.

Unfortunately, the value for human life in India still remains disturbingly low. It is specious to singularly blame governments for this, without also pointing fingers to all of us as a society. But it is certainly better to reflect on how we can build resilience to natural disasters than to think in terms of false choices such as “Is it just another flash flood or is it a man made disaster?“.

Originally written by Pavan Srinath in The Transition State blog on the Indian National Interest.

The Changing Local Climate of Bangalore: Part One

Bangalore is a rapidly changing city, both for the better and for worse. If somebody from the Bangalore of the ’80s or the ’90s were to visit the city today, it would be very difficult for them to recognize the city – and so it is to those who return from elsewhere after decades. The same is said of Bangalore’s weather, of how ‘it isn’t what it used to be‘. Cold winters, warm but never-too-hot summers and an extended season of rains populate the Bangalore weather of our dreams – dreams that appear to be increasingly disconnected from reality.

Many blame global climate change as at least part of the reason for a hotter Bangalore, along with local urbanization and related heat island effects, while others focus only on the latter. What does data from weather stations tell us? This post seeks to find out.

The question of whether Bangalore is heating up can be broken down into four parts. First, is the climate of Bangalore getting hotter in the long run? Second, is the change in temperatures consistent with global climate change? Third, is this warming related to the changes that we might be feeling? And fourth, could there be changes in temperature in the city above and beyond what gets recorded at a weather station?

The long term temperature trends of Bangalore tell an astonishing story.

Changing Local Climate of Bangalore 1901-2000

It turns out that over the last century, between 1901 and 2000, minimum temperatures have been consistently rising in Bangalore at a rapid rate. However, maximum temperatures don’t quite follow the same trend. Instead, it appears that maximum temperatures peaked in the 1920s and 1930s! Since then, they fell sharply by 1940 by about 1.5°C and have been rising steadily ever since.

Why was Bangalore facing such high maximum temperatures in the 1920s and ’30s? The answer is that we simply don’t know. It is important to note while there are prominent global trends at play like anthropogenic climate change, there are often distinct but important local trends in climate that need to be understood and dealt with locally. These trends may be anthropogenic or natural, but their impacts are often quite real. The wind systems around Bangalore could have been different then, or cloud cover was unusually low for a couple of decades, or it could have been entirely something else.

Apart from the mysterious hump in maximum temperatures, it is also pertinent to note that temperatures are rising faster in Bangalore than the global average! Minimum temperatures in the city have increased by about 1.32 degrees Celsius over the last hundred years (± 0.11°C). This is about double the global average. It is equally interesting to note that maximum temperatures are rising even faster since 1940.

Trends in minimum temperatures of individual months also indicate that December, February and March are warming faster than the other months. All months display the same peculiarities in maximum temperatures as seen in the graph above. It is also unfortunate that comparable data is not currently available to examine how temperatures in Bangalore have been in the last decade.

Do these changes confirm our perceptions of the changing climate of Bangalore? The answer to that is both yes and no. Climate data strongly supports the notion that winter nights are getting significantly warmer in the city – with summer nights joining in. Day time temperatures are not increasing (despite what people think!), at least at the century timescale. The difficulty when it comes to understanding perceptions of weather and climate is that people are experiencing weather in the present, and employing memory to remember weather from the past. Both of these are quite flawed, and that is not a slur on anyone’s judgement. Also, it is quite difficult to perceive a 1° change over a century (or even 2 decades!) when temperatures change by about 10°C every day, and an even larger number over the year.

Finally, the faster rise in minimum temperatures in Bangalore could indeed be because of urbanisation, land use change, and heat island effects. But one cannot conclude for certain by looking at this data alone. Also, the built up area near somebody’s home could have increased, with a corresponding loss of trees. Data from a weather station in the central business district of Bangalore can never be able to tell us the effect of that.

How has urbanisation affected ambient climate in Bangalore? Are there other factors that determine whether you feel that a summer is particularly hot? More on these questions in the second part of this article.

Written by Pavan Srinath

The data used in this article is available here from IMD. You can also download the CSV file here. This article is the first in a series being written in the run-up to the Open Data Camp Bangalore 2013 that is happening over the weekend of March 2-3.

The Plans of Mice and Men

Sitting through what are perhaps the last few days of the rains this season in South India, Priya Ravichandran writes about Hurricane Sandy and Cyclone Nilam, and how we reacted to the both of them. 

Starting a few days before late October from when Hurricane Sandy was to hit the eastern coast of the United States, the state governments, specifically in New York and New Jersey had started preparing for the storm that was forecast to be the biggest ever to hit the north eastern seaboard. With a sharp forecast, the focus was on who needed to evacuate, who needed to stand by to evacuate, the shoring up of supplies, emergency drills, emergency routes in case of flooding, and preparations for all other possible worst case scenarios. Information on whom to approach, where to go and when to go were all detailed out over radio, news, state and local websites, social networks. Police, fire, medical personnel, and other emergency workers were kept on stand by and information kept flowing, updating people on the nature, the breadth and the impact of the storm. Weather channels went into overdrive with the mapping and explanation of how strong and serious Sandy was shaping up to be. Mayors and government officials held press conferences informing, warning and assuring people. The hurricane made a landfall on October 29th.

A day before Sandy hit land; an atmospheric depression started gaining strength in the Bay of Bengal, off the south eastern seaboard of India. The depression was officially declared as Cyclone Nilam on the 30th of October was to hit the eastern coast of India specifically the southern coastal states of Tamil Nadu and Andhra Pradesh. Low-key warnings on the Indian Meteorological Department’s website, which then reached people through the announcements in the weather sections of regional TV news programmes. Much of the national media however, including most English news channels had heavy programming focusing on evacuation, emergency preparedness, the potential aftereffects and worst case scenarios… not for Cyclone Nilam, but for Hurricane Sandy. Yes. Hurricane Sandy.

Hurricane Sandy hit produced high winds up to 175 kilometres per hour (110 miles/hr) while Nilam hit the coast with winds just under 80 km/hr (50 miles/hr). The death toll for Sandy in the US alone is close to 130 and might increase. Nilam has so far taken close to 56 lives. Both numbers remain a conservative estimate, given that the delay of recovery is resulting in more fatalities.  The economic damage from Sandy is predicted to approximately $50 billion USD to $56 million USD for Nilam. Most generic comparisons of the impact of the two storms remains specious because of the differing scale of both weather events. Hurricane Sandy was dubbed as a “Frankenstorm”, and became the largest ever Atlantic hurricane in size in recorded history. Nilam, while no pushover, is the strongest cyclone to have hit the Indian coast since Cyclone Jal in 2010. 

The purpose here is to use this coincidence of large storms hitting the United States and India simultaneously – and learning what public officials and governments could have done in India and what we can learn from the preparations for Hurricane Sandy.

A lot of Sandy’s preparation learnt from what went wrong when Hurricane Katrina hit New Orleans. Several critical analyses of the havoc caused by Katrina were available and all the measures carried out where announced through every media outlet possible to ensure enough people got the message and got out or took precautions. As the NPR reported, the federal Disaster Management Agency had its paperwork in order and close to $3.6 billion ready for temporary housing even before Sandy hit.

In contrast, preparations for Nilam began quite late with a weak warning and very little information available to the public. Worse, between the IMD and state governments, the cyclone was not adequately tracked after it made landfall, where an it is possible that an assumption was made that the worst was over. Though the low pressure front weakened, heavy rains fell on Andhra Pradesh well after the cyclone made landfall – resulting in much greater loss of lives, crop damage and necessitated far wider evacuations of about 68,000 people, than what was expected earlier.

The eastern coast of India is no stranger to cyclones and some of the worst natural disasters in India have occurred near coastal AP and TN. A national workshop for developing strategy for cyclone mitigation met in 2003 and came up with a list of proposals to ensure minimizing the damage from cyclones. A budget of more than Rs.1490 Crores (~$300 million) was allotted then for cyclone damage mitigation,with very little to show for the money so far.

Basic measures like radio transistors, cyclone alarms, escape routes, flood drainage, reliable, cyclone resistant, cyclone shelters, warning systems at sea, sea walls and development of mangroves were all suggested but never carried through. The lack of planning showed.

In spite of cyclone structures being available and food being arranged, there was very little instruction on how to get to those places, or information on facilities within the structures. Given the estimated strength of the cyclone, very little precautionary measures were announced for residents. While the fisher folk were evacuated and other evacuation orders for people living in low lying settlements were given, no information on emergency procedures were issued for urban dwellers. There were also no additional measures taken for the differentially abled people.

There were very little, sometimes no updates from higher officials like the mayor on storm preparedness, absolutely no information on government websites state and local, very sparse coverage by English media and over the top, sensationalist coverage by regional media that took to comparing the cyclone to the 2004 tsunami. This was in direct contrast to the regular updates by NY mayor Bloomberg who was always accompanied by a sign language translator, giving to the point updates and helping reassure the people.

A lot of damage could have been prevented had there been fairly basic preparation both on the eve of the cyclone, and in general. Unclogging storm drainage systems, covering of potholes, and warning on the hazards of driving or being out on the streets could have been handled well before Nilam started. Part of the suggestions by the workshop for cyclone mitigation called for disaster preparedness including the provision of cyclone proof infrastructure. Natural sea side vegetation like mangrove forests were strongly recommended to act as natural barriers. Very little has been done to mitigate flooding in the affected areas in spite of recommendations. Every report on handling natural disasters has usually had a committee submitting a list of recommendations, extending a budget for the states to use and the states failing to put the list into action and misusing the budget with shoddy or no implementation.

Additional measures like declaring state wide holidays, ensuring additional power for hospitals and basic resources in schools and shelters where people were evacuated to could have been carried out. The most critical part of any disaster preparation has to be for the people who are going to be affected directly to have the information well before the disaster happens and to have it from a trustworthy public source. A lower minister who has had no experience in communication being sent out to warn people in fits and measures does not inspire confidence. Indian politicians often underestimate the importance of what really good PR skills are – they are not just about taking down the opposition, but providing and projecting leadership in times of duress. Calm, prompt and precise communication goes a long way in reassuring people. With the advent of social media and 24 hour news channels, the state and other governments ought to have taken an initiative to make its presence felt.

The simple fact of the matter is that by and large, the residents of coastal Tamil Nadu as Cyclone Nilam lost much of its intensity over the sea and the impact on land was lesser than expected. Good fortune does not excuse the complacency of the officials in handling natural disasters. Naturally occurring phenomena like cyclones and hurricanes become disasters only on reaching human settlements. While we need more robust forecasts and predictions, mitigatory and adaptive measures to protect ourselves can ensure that disasters need not happen as frequently as they do.

Priya Ravichandran is a Programme Officer with the Takshashila Institution and a resident of Chennai, India. Views are personal.

« Older posts Newer posts »

© 2017 Know Your Climate

Theme by Anders NorenUp ↑