Select an episode
Not playing

Roof of the World: Melting Water Tower

On the Tibetan Plateau, glaciers retreat and permafrost thaws. Yak herders, railway engineers, and downstream neighbors on the Mekong and Brahmaputra confront a future of floods, droughts, and data diplomacy.

Episode Narrative

In the vast expanse of the Earth, where the mountains touch the skies, lies a land known as the Roof of the World. This is a region defined by its towering peaks, flowing rivers, and a rich tapestry of cultures. The landscape is both beautiful and treacherous, teeming with life yet profoundly affected by the forces of nature. Over the last few decades, this area has become a battleground against climate change, grappling with the realities of a warming planet. From 1991 to 2020, fluctuations in precipitation during critical growing seasons have had a profound impact on grain production in China, revealing both the vulnerabilities of its agricultural systems and the resilience of its people.

The year is 2020. Across China, the rivers swell with rainwater; the sky unleashes storms that seem to come straight from a nightmare. Flooding, a recurring theme in this story, has claimed the lives of more than 282,737 people since 1950. The 2020 deluge alone affected 27 provinces, laying waste to communities and causing economic losses estimated at a staggering US$25 billion. The Yangtze River, long a symbol of life and prosperity, has turned into a harbinger of devastation, its surging waters reflecting the chaos unfolding across the land. As the rain falls heavily overhead, we are reminded how fragile life can be against the relentless power of nature.

Yet, the plight of China extends beyond the immediate horrors of flooding. A deeper examination reveals that the rhythm of life amid these disasters is intricately intertwined with agricultural practices, particularly the growing reliance on nitrogen fertilizers. These substances, while meant to boost production, have, over time, negatively impacted grain stability. Agricultural landscapes that were once vibrant with growth now mirror the ecological degradation that threatens to unravel the very foundations of society. Climate variability has thus transformed the relationship between man and nature into one of fragile dependence.

As droughts interlace with floods in an unpredictable dance, regions like the Jiaojiang River Basin present cautionary tales. This southeastern territory bears witness to a distinctive interplay with the El Niño-Southern Oscillation, where shifts in climatic patterns yield droughts that display significantly different characteristics depending on the phase of this global phenomenon. In stark contrast to floodwaters that ravage, these droughts create a silent despair, one that can linger in the soil, robbing farmers of their livelihoods.

Compounding these climatic challenges is the demographic growth in floodplains. Between 1990 and 2015, an astounding 110 million people moved into these perilous areas, adding to an already significant risk. Among them, the elderly population — those most susceptible to the ravages of nature — grew by 38 million, a staggering 86.4%. As these flood-prone regions have swelled in population, so too have the stakes, creating an urgent need for effective crisis management.

As a symbol of resilience, the aftermath of each flood holds stories of survival and struggle. In 2021, the “7-20” extreme rainfall and flooding event in Zhengzhou led to a revealing exploration of crisis management. Fragmentation in disaster response was laid bare for all to see, highlighting the pressing need for improved systems of communication and preparation as communities faced the realities of climate chaos. It served as a bitter reminder that our technological advancements — though considerable — often lag behind the complexities of natural disasters.

Turning our gaze back to the decades between 1991 and 2020, we witness another evolution — a rise in research efforts aimed at understanding and predicting landslides. With China emerging as a leader in this endeavor, the application of machine learning and deep learning techniques brought forth new hope in the fight against nature's unpredictability. As we harness the strengths of technology to figure out our place within the natural world, the relationship between humans and the landscape continues to change.

The Comprehensive Drought Index illustrates this evolving narrative well, merging the Standardized Precipitation Index and Standardized Runoff Index into a cohesive reflection of how both meteorological and hydrological patterns can interweave. In the Jiaojiang River Basin, this adaptability could serve as a model for regions grappling with similar challenges. In a world where climate patterns unravel unpredictably, this index stands as an anchor amid the storm.

Yet, the flooding in 2020 served as yet another ominous wake-up call. Authorities issued alerts for heavy rainfall for an astonishing 41 consecutive days, with average precipitation along the Yangtze River reaching its highest level since 1961. Streets once bustling with life fell silent, submerged beneath rushing waters, and agricultural fields turned to swamps. The rainfall’s intensity echoed through the nation’s economy, amplifying the enormous effects on commerce and agriculture alike.

As the storms wreaked havoc, farmers in coastal Bangladesh turned to innovation. Between 1991 and 2020, climate-smart agriculture technologies became vital lifelines. The use of thread and plastic pipes for irrigation emerged as a beacon of hope, bolstering community resilience against the rise of natural disasters and tides of salinity. As they fought to adapt to the changing times, the farmers demonstrated a will forged through generations of labor and struggle.

But the narrative of resilience does not end there. In the years that followed, governance in inland cities faced scrutiny. The severe flooding in Zhengzhou forced an examination of the disaster management structures — or lack thereof — within communities. With inadequate pre-disaster preparations and delayed responses, it became clear that to weather the storms of modernity, cities would need more than just historic experience; they needed foresight and collaboration.

In this time of turmoil, understanding the spatiotemporal patterns of precipitation became paramount. Between 1991 and 2020, a detailed examination using data from thousands of meteorological stations brought new insights into the changing climate. By analyzing the interplay between drought and flood, researchers offered hope for better predictive capabilities in the face of uncertainty. As the data flows like the rivers themselves, the answers lie within the storm clouds punctuating the skies.

As we stand at this crossroads, contemplating both the achievements and failures of the past few decades, a question lingers in the air. Will the lessons learned forge a new path of resilient adaptation, or will they fade into obscurity like the waters that once ravaged the land?

The future remains uncertain, yet the ongoing demographic shifts still promise an explosive growth in vulnerable populations living in flood-prone areas. In the coming decades, as flood memories decay and urban growth surges, what will become of these communities whose very lives are tethered to the whims of nature? The echoes of past storms serve as a reminder that amidst life’s relentless cycles, we must learn not only to endure but to thrive.

As we reflect on the resilience and fragility intertwined within the fabric of life on the Roof of the World, we grasp the stark truth of our existence. Water, once a giver of life, can also become a force of destruction. The mountain peaks gaze down, a mirror to our struggle, while the rivers carry forth tales of human persistence. In this realm of contrasts, perhaps it is not merely survival that we seek, but an understanding — a deeper connection to the world that sustains us, as we emerge from the turmoil with newfound wisdom, ready to confront the tides of calamity ahead.

Highlights

  • In 1991–2020, precipitation fluctuations during crop-growing seasons and natural disasters significantly reduced grain production and yield stability in China, with irrigation shown to mitigate risks from precipitation variability. - By 2020, floods in China had killed more than 282,737 people since 1950, with the 2020 floods affecting 27 provinces and causing economic losses estimated at US$25 billion, primarily due to extreme precipitation along the Yangtze River. - From 1991 to 2020, nitrogen fertilizer application in China was found to negatively impact grain stability, highlighting ecological degradation as a vulnerability to climate variability, especially precipitation fluctuations. - The Jiaojiang River Basin in southeastern China demonstrated a strong response relationship between drought indices and El Niño-Southern Oscillation (ENSO) events, with drought characteristics varying significantly during different ENSO phases from 1991 to 2020. - Between 1990 and 2015, the population living in Chinese floodplains increased by 110 million, with the elderly population in these areas growing by 38 million (86.4%), exacerbating flood risk in the context of climate change and rapid development. - In 2021, the “7-20” extreme rainfall and flooding event in Zhengzhou, Henan Province, revealed fragmentation problems in disaster crisis management, underscoring the need for improved emergency response and risk communication in inland cities. - From 1991 to 2025, China emerged as the leading contributor to landslide susceptibility research, with machine learning and deep learning techniques increasingly employed to model and predict landslide risks. - The Comprehensive Drought Index (CDI), combining the Standardized Precipitation Index (SPI) and Standardized Runoff Index (SRI), effectively reflected the combined meteorological and hydrological wet-dry characteristics of the Jiaojiang River Basin from 1991 to 2020. - In 2020, authorities in China issued alerts for heavy rainfall for 41 consecutive days, with average precipitation in areas along the Yangtze River reaching the highest level since 1961, contributing to severe flooding. - Between 1991 and 2020, higher farmers’ incomes in China strengthened their incentives for agricultural engagement, underscoring the critical role of agricultural subsidies and policy support in enhancing grain production stability. - From 1991 to 2020, the use of thread pipe/plastic pipe for irrigation ranked first among climate-smart agriculture technologies adopted by coastal farmers in Bangladesh, reflecting adaptation strategies to frequent natural disasters and salinity intrusion. - In 2021, the severe precipitation-induced flooding in Zhengzhou, Henan Province, was analyzed using random forest and XGBoost algorithms to identify risk factors and optimize flood risk assessment models. - Between 1991 and 2020, the spatiotemporal patterns of typhoon storm surge disasters in China were categorized into three clusters based on tropical cyclone track clusters, supporting disaster mitigation strategies. - From 1991 to 2020, the population in rainstorm hazard areas of China increased by 14.6%, with the elderly population in these regions growing by 86.4%, highlighting demographic shifts in disaster exposure. - In 2020, the average intensity, quantity, and duration of precipitation in southern China were among the highest since 1961, contributing to severe flooding and significant economic losses. - Between 1991 and 2020, the adoption of climate-smart agriculture technologies by coastal farmers in Bangladesh was significantly influenced by annual agricultural income, extension contact, training exposure, knowledge of CSA, and attitude towards CSA. - In 2021, the governance capacity of inland cities in North China for managing extreme precipitation and flooding events was found to be low, with insufficient pre-disaster preparation and prevention, poor risk communication, and slow emergency response. - From 1991 to 2020, the spatiotemporal variation of precipitation regime in China was investigated using the Standardized Precipitation Index (SPI) calculated from monthly data from 3804 meteorological stations, providing insights into drought and flood prediction. - In 2020, the population in floodplains of China was expected to increase rapidly in the coming decades, with flood memory effects on controlling population growth decaying over time, exacerbating flood risk. - Between 1991 and 2020, the use of artificial intelligence methods, particularly machine learning and deep learning, became prominent in landslide susceptibility modeling in China, with significant contributions from Chinese research institutions.

Sources

  1. https://onlinelibrary.wiley.com/doi/10.1002/fes3.70096
  2. https://www.mdpi.com/2071-1050/17/6/2616
  3. https://doi.apa.org/doi/10.1037/tra0001928
  4. https://www.hanspub.org/journal/paperinformation?paperid=109129
  5. https://onlinelibrary.wiley.com/doi/10.1111/tgis.70099
  6. https://www.mdpi.com/2076-3417/15/15/8660
  7. https://www.nepjol.info/index.php/JAAR/article/view/85132
  8. https://saudijmph.com/index.php/pub/article/view/76
  9. https://www.banglajol.info/index.php/IJARIT/article/view/82752
  10. https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-025-23981-w