Pollution diaries

I was recently back in Delhi in December 2024. Air Quality Index (AQI) during certain periods was well over 10x healthy levels driven by particle pollution (PM 2.5, PM 10) primarily from unburnt carbon [CPCB India AQI, Google AQI]. It is an alarming health hazard considering 20-30% adolescent kids tested in Delhi were asthmatic courtesy air pollution [NIH paper]; with a national survey showing 6% [NIH paper]. According to research, this level of pollution erases 7.8 years from the average life expectancy [AQLI by EPIC research, M1, M2].

While it’s a topic covered well by media [M1, M2, M3] – it is an issue that hits personally, and I was still unable to comprehend how we are unable to solve this problem as a society for decade(s) now. And, why, this isn't a national emergency to enforce any and all measures to stop this tragedy. I pestered everyone to learn about the issue and about what could be done.

And, while I am optimistic from what I learnt, this problem is far from fully solved. Change can't happen fast enough. I share some what I learnt for others who are equally confounded or looking to support measures that can be taken to speed up the resolution via policy, governance, business model or technical innovations. As a follow-up to this monologue, I intend to collaborate with Wheels Global Foundation on this cause.

Root cause of the pollution issue

There are several contributors to the poor AQI, the sources are regionally spread out, vary by time of the year, time of the day – primary ones for Delhi-NCR are biomass burning for cooking, heat (indoor), agriculture residue burning (stubble burning), industry, transport and power (ambient). [iFOREST study, IITK study CPCB, CSE study - mobility]

For context, over the last two decades, there have been initiatives in transport such as switch to CNG, Metro, in cooking fuel such as PM Ujjwala yojana to expand LPG as clean cooking fuel, cheaper electricity with government subsidies for electric heating, and dismantling of thermal power plants and pushing out of polluting industry from the capital. [History of Delhi NCR Pollution]

And so, while all are important to address to improve AQI year round – and it remains a topic of debate, the leading cause of winter season smog spike is stubble burning. [DPCB]. Intense burning days contribution is 30-40% for PMI in Delhi [article cites IITM study]. Stubble burning is an activity by farmers to quickly clear their field of any residue after harvesting the crop (in this case, rice or paddy) by way of burning the stubble to get field ready for seeding next crop. It is typically tracked via NASA satellites. Snapshots Image 1, Image 2 from last year, and increasing trend of fires are recorded in northwest India at time of year since 2003 via source NASA worldview].

The stubble burning maps are generated by using algorithms to detect brightness levels for fire and thermal anomalies – typically have a resolution of 4kmx4km for a pixel and happen at certain scheduled times in a day (10.30am, 1.30pm).

Official data from Government of India is reported by the commission for air quality management (CAQM) in the National Capital Region (NCR) and adjoining areas has directed the Punjab government to follow the protocol laid by the Indian Space Research Organization (ISRO) to record stubble burning cases in the state. According to these statistics, the number of incidents of paddy stubble burning, have witnessed a significant decline in 2024 during paddy harvest season. Punjab & Haryana have recorded 87.7% & 80.2% reduction in number of burning incidents in year 2024 as compared to year 2021 [CAQM 2023 , Government; Environment Min 2024 Filing; Stubble updates].

However, the trend is disputed as Korean satellites detected fires later at 4.30pm in areas that were not burnt during fly past of NASA satellites; additionally, data didn’t cover the full season, and it didn’t match with the AQI and PM2.5 sensor statistics. [ToI, the print, NASA Scientist] Other measurements such as using Aerosol optical depth, which uses federation of ground-based remote sensing aerosol networks using photometers?alongside satellite systems can potentially give a lasting inference of stubble fires via smoke plumes “where there’s smoke, there’s fire”. Another way is to monitor burnt area and char on open fields [Counting fires].

Smoke plumes can be tracked in this way by NASA satellites traverse from northwest India via the Indo Gangetic plains towards the Bay of Bengal. And then, similar smoke plumes arising from central India are carried to Arabian Sea. Anecdotally, I have experienced wildfire smoke plumes traveling 200-400 miles in US Northwest, and it resembles my Delhi NCR smog experience in November.

Root cause of stubble burning

Simply put, burning the residue is perceived by the farmer to be the most fast, efficient and cost-effective way to switch to the next crop after harvesting. Tragedy of the commons – as the farmer optimizes individual economics; the result is the worst case for everyone.

There is a lot of context behind the farmers burning stubble and crop residue and no doubt this is also an extremely hot socio-political issue [remember Farmer protests of 2021]. Before we get into the reasons, let’s not forget the context – that is Indian agriculture.

Agriculture is the backbone of Indian economy contributing 16% of the GDP with over 40% of workforce. States like Punjab in green belt have 80% of economy and employment from agriculture. And, to preserve food security, farmers in India are supported with heavy subsidies from agricultural equipment, fertilizer, seeds, power, irrigation, crop insurance, fuel-diesel, and residue management. India produces 132MMT rice and exports 16 MMT (milled rice) with Punjab nearly 10% of production [USDA 1; USDA 2]. This generates 141 MMT of paddy crop residue of which 92 MMT is burned; 10MMT just in Punjab. This burning just in Punjab generates 50 MMT of carbon dioxide (CO2), millions of tons of carbon monoxide (CO), substantial quantities of particulate matter (PM), sulfur oxides (SOx), nitrogen oxides (NOx), and black carbon. [IITK study]

Reasons for paddy stubble burning [ex-IAS explanation]:

• Overall, paddy crop also has a high government MSP (minimum support price) that has also distorted the economics in favor of paddy crop. Unfortunately, also 80% of the paddy grown in northwest, especially Punjab takes an extended 6-month period to mature, consumes most water and generates most stubble. Also, as put simply by a farmer I spoke with - paddy straw and stubble are not considered a good livelihood feed for usage in-situ by grazing.
• One of the factors is that Punjab Preservation of Subsoil Water Act limited sowing of paddy before mid-May to conserve ground water; however, this led to a short turnaround of just two weeks between paddy harvest in November and the next crop – while overlapping with a season with low to no winds keeping the burnt residue hanging in the lower atmosphere.
• There is also unintended result from mechanization and combine harvesters. When the farmers used to harvest by labor by hand, they left no stubble and cut from the root. However, modern harvesters optimize the speed and efficiency of cutting the crop to ensure maximum return to the farmer. These harvesters, however, don’t take care of the straw or 6-inch stubble in paddy.

To get some empathy with a farmer’s unit economics:

• Land: Average land holding is 2.5acres per farmer (70% are smaller than 5 acres)
• Rice yield per acre: 25 quintals (1 quintal = 100 kgs)
• MSP for rice: Rs 2300 per quintal (market rate is close to Rs 4k India)
• Minimum revenue for farmer per acre: Rs 57k (better than most alternatives in season)
• Cost per acre [Gov]: Rs 21k
• Crop residue per acre: 2.5 tons of stubble as residue per acre
• Additional operational cost for current crop residue management: Rs 10k per acre if on fully rental operation basis; Rs 2.5k per acre if operating owned machine and Capex separate

So, there is significant impact (between 5-20% revenue, reducing cash flow to pay back loans and run livelihood by up to 33%) to the farmer’s economics to manage the crop residue – which, to their perception doesn’t give any real return. Paddy straw and stubble don’t support livestock well; and rotting and fear of pests are added concerns for the farmer – on top of quickly preparing for seeding. Farmers agree that stubble burning is harmful overall; but they see it as a trade-off against a risk to their livelihood and the next crop seeding.

Meanwhile, the Supreme court has ruled that burning stubble is a punishable offense, and the Government of India has recently doubled these fines in 2024 for farmers to Rs 5k per acre for these violations. It is also reported that the government has registered FIRs against offenders and made recoveries against these reports [PPCB]. However, stubble burning is not a criminal act; and any ability to prosecute farmers due to non-compliance went away with the farmer protest of 2021. Local officials are in a tough position as enforcement at this scale will also risks social unrest – even if one farmer accuses of coercion – could lead to disaster for the government.

So, what can be done?

Alternatives to stubble burning

• First and foremost, there is acceptance that crop selection should be on market basis and that the MSP artificially incentivizes rice’s selection as a crop for regions such as Punjab with ground water concerns. Recommendations are to diversify potentially toward millets with much lower water requirement, and much lower waste generation. Additionally, crop cycle distortion due to ground water conservation is potentially to be adapted.
• Second, farmers need to be able to remove the stubble from the field; and then, either manage the removed residue in-situ or the export this residue ex-situ and transport and prepare it for end use - without significant additional expense. They need to be able to do from a farmer’s perspective “(i) clear the crop waste by end Nov, (ii) don’t ask me for money”.

Crop residue removal from the field through mechanization

A combine harvester leaves about 6 inch stubble and leaves straw lying on the field behind it because its primary function is to harvest the grain from the crop, meaning it cuts and collects the grain heads while leaving the lower stem portion, called stubble, behind in the field; this is a deliberate design to efficiently separate the edible part from the plant and minimize unnecessary material being processed by the machine at harvest time. [1, 2, 3, 4, 5] [Simulation, Wiki]. Both, straw and stubble on the field create a risk of pests and perceived challenges in seeding.

One opportunity is to either redesign or retrofit the combine harvester itself with the stubble remover and the straw and stubble cutter for in-situ methods or straw baler for the ex-situ methods. Few attempts (in public domain) have been made to tackle both the stubble and straw at the time of harvesting – primarily with the view of optimizing for harvesting time and yield and not worrying about the full crop rotation process. In my view the challenge in combining harvesting and stubble removal would be maintaining high reliability of the harvester while also cutting stubble and optimizing both processes within the available horsepower of a single tractor. I am actively looking to collaborate with Engineering teams to research this topic.

In-situ mechanisms

The Government of India along with state governments have extended approaches that can allow the farmer to not have to remove the stubble from the field but rather use it like organic manure.

Delhi Government for example is providing a bio-digester enzyme that if distributed over the stubble and straw turns it into mush and organic manure within a few weeks [1].

Punjab Government has promoted subsidy schemes for machines (also called CRM for crop residue management machines) to cut the stubble and incentivize it’s use within the field as organic manure – thereby possibly supporting higher yield [2].

This operation happens in multiple steps:

• First, the farmer needs to use an attachment to the combine harvester called the Super Straw Management (SMS) system that chops the straw into fine pieces during harvesting and equally distributes it in the field. The 6-inch stubble, however, remains post harvesting.
• Second, the farmer when ready for seeding the next crop uses a machine called ‘Super Seeder’ or ‘Happy Seeder’ that allow for seeding without ploughing and these machines also cut and distribute the remaining stubble into the ground as part of the seeding process.

Both these machines are heavily subsidized (50% for individual farmers, 90% for community use); only operating cost to the farmer is labor and diesel to run this pass with the seeder – which is not necessarily extra expense.

Ex-situ mechanisms

Paddy stubble is proven useful across following uses, and each has its unique valuation of the stubble residue and constraints – whether economic, supply chain, or operational:

• Thermal Power Plants [3] – individual fired or co-fired in coal plants (Govt mandates 5%)
• Compressed Biogas and bio-manure plants [4, 5, 6] – using enzyme anaerobic digestion
• Bioethanol/Bio-diesel Plants [7, 8] – Indian government has a mandate for fuel blending
• Alternate uses [9, 10] – such as briquettes, packaging material, tile, fiber, paper etc.
• Bio-char - partial combustion near the site to create a more potent fuel richer in Carbon
• Local/neighborhood steam generation for heating (industrial / home heating)
• Local small-scale bio-digester
• Additional unlisted uses

Typically, for most ex-situ use cases, the straw and stubble need to be uprooted and collected and baled into rolls or heaps for ease of transportation. This machine is typically available with a significant subsidy from government - especially if used in a community setting. Transport is using tractor trailer. However, one limitation to keep in mind is that transporting long distance is not viable since density is low (110 kg/m3), just 15% of that of crop.

Innovations

A business model innovation in ex-situ is the extension of services provided by the baler. The baler, as an entrepreneur [Ref 1, Ref 2; Ref 3]. The baler purchases baling machine fleet using government subsidy for community use (80-90% subsidy). Baler then employs labor and takes contracts for farmers looking to get rid of stubble and straw. Deal is struck with minimal payment to/from the farmer, the baler creates the bales, picks up the stubble, sells and transports to end use plants for a profit. There are many calls for the government to appoint private contractors for baling, logistics to directly alleviate the farmer from the burden of residue management [Ref]. ?

Another innovative approach hugely successful is from Deloitte crop residue management [programs]. Deloitte deployed a model in Haryana on 666 villages resulting in a 54% drop in stubble burning. As part of this initiative, they registered 150k farmers, educated 68k farmers and provided them an app ‘Krishi Yantra Sathi’, which allowed the farmers to share agricultural equipment such as the baler, super seeder, happy seeder – an Uber – increasing utilization rate and reducing cost per acre for farmers. Deloitte is next assisting Punjab with similar pilots post SC directive.

Another innovative model proposed is to allow farmers to gain from carbon abatement programs or other pollutant reduction incentive programs. As an example, Google bought carbon removal credits from Indian farmers to mitigate it’s emissions for data centers [Google] – though this mechanism hadn’t worked yet [carbon credit]. Government of India has initiated efforts to create cap and trade programs for carbon [National cap and trade, 1, 2]. Another emissions trading scheme is currently tied to industrial PM2.5 pollution in Gujarat [EPIC 1, 2]. In 2022, Government of India initiated efforts to create cap and trade programs for carbon [National cap and trade; Terano]. Challenge in relying on such markets is the uncertainty about who will be the buyer and how will they be motivated to adhere to fulfilling their needs. With US withdrawing from climate accords, voluntary carbon market prices remain negligible.

Current situation and future projections

If we just focus on Punjab (at 10% of the country), total paddy is 13 MMT rice (milled) and approximately 14-15 MMT straw and stubble in 2023. Currently 4-5 MMT (or 30%) paddy residue is utilized in-situ or ex-situ. Utilization of residue needs to go up to three times to solve this issue. This will need crop residue removal, supply chain logistics to end use and increased end use capacity. In terms of economics, the extra cost of removing this residue of 10MMT will cost farmers Rs 1000 crore (at operational cost of Rs 2500/acre). This is the current funding gap for the farmer to not burn. The Punjab government is looking to the center to fund this opex beyond machine capex subsidy [CM]; this is where innovation and new end uses need to creatively bridge this economic gap.

This section based on [CEEW study]

Focusing on Punjab, crop residue management target currently is 50% is in-situ and 25% ex-situ and crop diversification each [CEEW study]. However, nearly half of the in-situ practicing farmers followed partial burning of paddy residue before using the in-situ CRM machines in Kharif 2022.

In-situ approach hasn’t been as effective as hoped. Despite the promotions, 64 % of the farmers in the study used combine harvesters without a Super SMS attached in 2022. 38 % of them also felt that using a Super SMS– attached combine harvester would be expensive, adding a Super SMS will cost Rs 500–800 extra per acre when compared to partial burning.

In-situ?CRM machines such as?Happy and Super Seeders?should theoretically have enabled farmers to?shun residue burning.?Backed by the central government’s subsidy scheme, the stock of in-situ CRM machines has swelled in the region. Based on the Government reports, 110k total CRM machines are presently in market (MoAFW 2022b) [43,452 Super Seeders and 13,560 Happy Seeders]. A single Super Seeder, in contrast, can cover 120 acres of land in one season (20 days). Together, these machines can cover 100% of paddy fields in Punjab if deployed at maximum capacity- by way of sharing or having rotating contractors.

However, nearly 88% of surveyed farmers who practiced partial burning in 2022 said that they do so to ensure the smooth operation of their in-situ CRM machines (Rajpurohit 2022). Farmers are afraid of pests and risk of maintenance of CRM machines during the short crop rotation time window.

Ex-situ: In contrast to in-situ, majority of the 25% of farmers who adopted ex-situ methods responded that they were driven to ex-situ due to high costs of in-situ methods (Rs 2.5k operational expenses). Nearly 86% agreed that they would continue with ex-situ methods in Kharif 2023. Consequently, the 2023 Kharif season involved more balers in action.

However, there are several constraints in scaling ex-situ crop residue management:

• a shortage of baling machines and residue end-users continues to remain a significant barrier in the sector
• due to low density transportation is expensive and infeasible across long distances without additional processing of the bales (compaction, palletization, briquetting)
• the supply is seasonal, but the usage is annual – so beyond the transportation, there is the cost of storage and maintenance
• gaining security of supply contracts is difficult for setting up new end use plants – typically, for financing larger end-usages, long term contracts need to be in place – with agreed upon economics (for example, coal plants have a long-term fuel supply agreement for years)

Still, currently, power, biogas and bio-ethanol initiatives appear the most scalable opportunities to grow the crop residue processing capacity for Punjab to 20MMT and as an extension for the country. As I discover more, will share my learnings on the end uses, their unit economics and respective challenges.

A better future

After going through the journey and understanding the factors at play as well as the wheels in motion, I do feel problems are surmountable. Highest levels of Governments both state and center are aligned, people want this change, farmers and laborers suffer as much from the air pollution and they are increasingly aligned. We still do need to help make the economics work, investment to scale happen, and do better – across:

• Sensing (remote, satellite), data, AQI, fire measurement, verification, transparency & statistics
• Technological innovation in mechanization of stubble removal and end use of stubble
• Business innovation in reducing effort and raising incentives for farmers via new business models
• Supply and value chain optimization for residue to generate energy
• Funding and financing mechanisms to support deployment of capex for stubble end use
• Influencing policy, regulation, and public finance sources

As shared, I intend to continue to investigate opportunities in this space, sharing the latest hypothesis and gaining feedback on refining the direction. And, we still have 9 months till this coming season tries to kill us!

Note: No, this article is not written by Chat GPT - I operate with limited knowledge, learning from others, pareto prioritization, human reasoning, sleepless nights and real hallucinations!

Thank you to my family, school and IIT friends, pan-IIT collaborators, and ex Tata Power colleagues who helped me to learn about the topic and to all the reference reports and articles cited (especially iFOREST, CEEW, EPIC India, Deloitte CRM - sustainability initiative); any errors or omissions are mine alone but not intentional. Amit Gupta Saurav Raaj Sreekumar Kuriyedath Abhishek Dabas Mayank Garg Deepak Trivedi Amit Gupta Abhimanyu Singh Rana Nishant Sood Sanjay Kaler Varun Aggarwal Nitesh Kumar Ratan Agarwal Mudit Narain

I would love to collaborate around initiatives around each of these opportunities to make this an issue of the past. Feel free to also share ideas and opportunities you have investigated in Comments for others that are interested.