8 Topics from Chapter 8: Water Cycle Changes

The Hydrologic Cycle (8.1 and 8.3)

Linh Vu

Water is essential to the livelihood and wellbeing of all life on Earth. It composes 70 percent of the Earth’s surface, yet only around two percent is available as freshwater. Of that two percent, ice sheets, glaciers, and snowpack take up 97 percent. That leaves about three percent of total freshwater to share for essential ecosystem functioning and human societal water needs. It is a vital natural resource on the planet, supporting a wide range of human activities from agriculture to industrial processes and even electricity generation. In theory, the three percent of freshwater we share is enough to meet the demands of human life and ecosystems around the planet. However, geographical and seasonal differences can play a role in the distribution of these resources, and they may not be enough to meet regional demands. Often, those who need the water the most contribute the least to global warming. Water scarcity arises when there is not enough fresh water available to meet demand. Over half of the world population experiences freshwater scarcity for at least one month of the year, with half a billion facing freshwater scarcity year-round. Roughly 80 percent of the world currently suffers from high levels of water insecurity.

Climate scientists can assess and attribute changes in the water cycle using observations, climate models, and theoretical understanding from physics and chemistry. For example, scientists can observe how the water cycle changes in response to natural variability and human influence using paleoclimate data, such as tree rings or ice core data.

The Clausius-Clapeyron relation is key for understanding the relationship between air temperature and water vapor pressure that allows scientists to understand and theorize global responses to climate change. It can be manipulated to understand the relationship between specific humidity and temperature, or in other words, how a warming climate will affect the moisture content of the atmosphere. Parts of the IPCC report on observed changes to the water cycle is summarized here; however, it should be referred to for greater detail if needed.

1. Observed changes
   • The global water cycle as a whole

Specific humidity is the amount of water vapor in a unit mass of air. Generally, warmer air can have higher specific humidity than colder air. Relative humidity is the ratio between the amount of water vapor in a unit mass of air and the maximum amount of water vapor it can hold. Sometimes we see relative humidity decrease or stay the same. In contrast, specific humidity increases because the air temperature rises fast enough to keep a similar ratio at a cooler air temperature. Since the 1970s, the climate has seen increases in global near-surface and tropospheric specific humidity caused by human-caused warming. In the terrestrial Northern Hemisphere, we have seen a decrease in relative humidity since air over land warms up faster than the ocean. We generally see higher moisture flux with moister air, leading to increased precipitation over wet tropical regions and “more extreme and persistent wet and dry weather events.” (Section 8.3.1.1)

The expected changes to the water cycle can best be summarized as intensification or strengthening. There is no clear definition of global water cycle ‘intensity’; however, a global increase in land precipitation and humidity is expected with human-caused global warming. This is due to the theoretical understanding that warmer air can hold more water. However, due to large observational uncertainties and a low signal-to-noise ratio, this has not yet been detected nor attributed to human activity.

The flux between the ocean and atmosphere can best be determined by the difference between precipitation and evaporation (P-E) and is equal to the moisture converging in the atmosphere. From the ocean’s perspective, this can also be thought of as the freshwater flux between the ocean and atmosphere. The ocean takes up about 70% of the planet’s surface, so there are relatively few locations where evaporation can be measured. Using model estimates, the IPCC provides “robust evidence of an amplified oceanic patterns in P-E” (Section 8.3.1.1), leading to higher sea surface salinities in regions where evaporation exceeds precipitation. Over land, a large amount of interannual variation is linked to the El Niño–Southern Oscillation (ENSO), a climate pattern that affects much of the rainfall occurring in the tropics and the Asian-Pacific. ENSO moves warm ocean waters between central America and southeast Asia, increasing rainfall. There is some evidence that P-E increases in wet regions and decreases in dry areas in tropical systems, but this also aligns with ENSO making it hard to make a conclusive assessment on the human influence of P-E over land.

• • “Wet gets wetter, dry regions persist and expand.”

There has been an increase in mean annual precipitation amounts in mid-latitude land areas in the Northern Hemisphere. However, there are still observational uncertainties when assessing precipitation trends at a regional or seasonal scale due to internal variability. Effects from greenhouse gases (GHGs) and aerosols, plus changes in precipitation intensity versus frequency, can also make it harder to quantify how much precipitation will change with warming. Aerosols can both suppress and intensify rainfall. In shallow clouds, larger aerosols—which are mainly produced from biomass burning and other human activity—suppress precipitation, while in deep convective clouds, smaller aerosols are known to strengthen precipitation. Higher CO₂ levels are correlated with changes in precipitation and frequency but have not been formally attributed to human activities. Instead, humans have contributed to “increased temporal variability of annual precipitation amount over land… which is most pronounced in annual mean daily precipitation intensity” (Section 8.3.1.3).

Annual mean potential snowfall area in high-latitudes and mountain watersheds is expected to decrease by 0.52 million km² per decade. In the Himalayan Alps, wintertime snowfall has increased while summertime snowfall has reduced. There is expected to be reduced mean annual snowfall in the Arctic despite a substantial precipitation increase, mainly in the summer and autumn when temperatures are close to the melting point. Snowfall directly influences the amount of runoff and melt discharging into streams that impact irrigation, drinking water, and even flooding, depending on the season.

There is low confidence in the assessment of trends in global river discharge due to impacts by land use, dam construction, and differences in geography and morphology. There is, however, “robust evidence and high agreement that warming has led to earlier spring discharge maximum… earlier break up of Arctic river ice, as well as indications that warming as led to increased winter flows and decreased summer flows” (Section 8.3.1.5). Regionally, anthropogenic climate change has altered streamflows, although globally, there has been no significant trend. Surface dimming caused by local aerosol emissions can affect evaporation and heavily influence streamflows. Land runoff variations correlate with ENSO variability and further complicate assessing trends in global streamflow. Discharge due to melting glaciers has already reached its maximum point and begun to decline due to ice mass loss. There is low confidence that climate change has already affected the frequency and magnitude of floods globally.

Evapotranspiration is a term specific to land processes defined as the sum of evaporation from the surface plus transpiration from plants. In the IPCC, evaporation is a more comprehensive term that includes all evaporative processes over the Earth’s surface, including the ocean. Evaporation has declined over most regions in the last 50 years, while evapotranspiration has increased since the ’80s. The main driver of the increase in evapotranspiration is plant transpiration and the increased atmospheric demand for water. Using ecosystem models and satellite observations, scientists found an increase in ‘greening’, or the percentage of Earth covered by leaves, coincided with observed evapotranspiration trends. Plant water use efficiency is expected to rise with increasing CO₂ levels, potentially canceling the associated increase in evapotranspiration. However, this may not be true across all ecosystems. For example, increased water use efficiency in some forests meant increased plant growth and more evaporative demand. Some studies even suggest that increased efficiency would not be enough to compensate for increased plant growth, and surface water availability would decline even further. Such drought conditions would offset the CO₂ effect and lead to a decline in water use efficiency. There is high confidence these trends are partly attributed to both GHG forcings and aerosol emission.

Climate model improvement and paleoclimate data inclusion have greatly increased the confidence in projected aridity changes with global warming. Scientists have discovered a unique link between anthropogenic forcings and global trends in aridity over the last 120 years. The most dominant trend is the drying of North and Central America and the Mediterranean. The growing contrast between the wet and dry regions and advancing continental aridification is found to be associated firstly with greenhouse gas emissions and anthropogenic aerosols. Increasing evapotranspiration is linked to a global trend in decreasing soil moisture and the water availability during dry seasons and is very likely attributed to human influence. The robustness of trend attribution can vary widely across regions. Droughts have increased with intensity and frequency in the Mediterranean, western North America, and southwestern Australia due to the effects of human influence. Paleoclimate data suggests that certain areas in North America and the Mediterranean have experienced similar or stronger droughts than those reconstructed over the past thousand years. There is less confidence in the trend attribution in southern Africa and southwestern South America; however, greenhouse gas and aerosol emissions are still likely to affect these areas.

• • Freshwater reservoirs and impacts on access to clean water

Perhaps the most pressing issue in water cycle intensification is the impacts on freshwater reservoirs and clean water. Glaciers lost more mass in the 2010s than in any other decade since the beginning of the observational record. The near-universal retreat and decline in glaciers are very likely to be attributed to anthropogenic influence. Glaciers have contributed to about 21 millimeters of sea-level rise in the past 50 years, with the most mass loss occurring in Alaska, Greenland, and Arctic Canada. Runoff from smaller glaciers has decreased, while runoff from larger glaciers has increased. In certain regions in the Himalayas, increased precipitation can offset the mass lost by glaciers, but this is unique and considered anomalous.

Seasonal snow cover has been steadily decreasing in the Northern Hemisphere since 1950, and the IPCC has reaffirmed anthropogenic influence. It is closely tied with temperature; the response in temperature explains roughly 40 to 85 percent of changes in snow cover extent. Snowfall as a fraction of precipitation has decreased considerably as well. Springtime snow cover is especially critical since it is a significant source of drinking water, insulates the ground, and reflects radiation. A decline in springtime snow cover extent, depth, and duration is consistent with global warming and attributed to anthropogenic influence. However, the main drivers of snow cover changes are still disputed.

Wetlands and lakes were not previously considered in IPCC reports, yet they have the potential to affect the climate through carbon and methane budgets. Efforts have been undertaken to take inventory of global surface water bodies using satellite sensors. Results have shown a substantial decrease in natural surface since the 1970s and an increase in interannual variability in surface water extent. In addition, human-made water bodies such as rice paddies have increased and will continue to, especially in Southeast Asia.

Groundwater is the most extensive supply of freshwater, delivering “between a quarter and a third of the world’s annual freshwater withdrawals to meet agricultural, industrial, and domestic demands” (Section 8.3.1.7.4). It is somewhat challenging to attribute how groundwater has changed because of non-climate influences such as human withdrawals and land-use change. Groundwater recharge is found to be associated with heavy or extreme precipitation, especially in the tropics and subtropics. Precipitation intensities and their sensitivities to warming temperatures are often underestimated in climate models and may lead to underestimating their effect on recharging groundwater. Groundwater depletion is associated with the expansion of irrigated agriculture as a consequence of a growing population with higher water demands. Droughts and climate variability can also amplify depletion. This magnitude of this change is expected to contribute to global sea-level rise by 0.3 to 0.9 millimeters per year.  Changes in meltwater from glaciers and seasonal snowpacks also reduce the duration and magnitude of recharge. It is not well understood how receding glaciers will affect the groundwater system; however, shifts in timing and magnitude of groundwater levels are already observed. It is expected glacier loss will lead to reduced recharge.

2. Variations in large-scale phenomena and regional variability
   • Atmospheric circulations

Atmospheric circulations are primary drivers in transporting heat and moisture throughout the planet. One of the most frequently mentioned is the Intertropical Convergence Zone (ITCZ). The ITCZ is a region near the equator with abundant intense sun and warm water. With the help of trade winds, air converges here and rises, drastically affecting the precipitation in regions along the equator. With global warming, the IPCC expects “significant narrowing and strengthening of the Pacific ITCZ… but no change in the ITCZ location” (Section 8.3.2.1), leading to increased precipitation and expanding dry zones, particularly over land. These trends are further supported by ocean surface-salinity observations, where a freshening is expected in the core of the ITCZ and increased salinity in the margins.

Hadley cells are responsible for trade winds and control weather patterns in the low latitudes. On either side of the equator, warm moist air converges and rises. As it rises, it expands, cools, and travels poleward until it sinks in the subtropics, often bringing precipitation. The Hadley cell has “very likely widened and strengthened since at least the 1980s, mostly in the Northern Hemisphere” (Section 8.3.2.2), which has important implications on the poleward shift of tropical cyclone trajectories in both hemispheres. A poleward shift leads to a decline in precipitation at the poleward edge of the subtropics, where air descends. There is limited evidence on human influence, but a southward and widening in the Southern Hemisphere Hadley cell is a robust feature of many models. Greenhouse gases and stratospheric ozone contribute to the expansion of the Hadley cell. Meanwhile, Antarctic ozone depletion can cause a poleward shift in the Hadley cell, leading to poleward shifts in storm tracks and expanding dry regions.

• • Tropical cyclones

Since the last IPCC report, there has been extensive progress in understanding the changes behind tropical cyclones and their sensitivity to GHGs and aerosols. Tropical cyclones can cause local heavy rainfall events and flooding. In addition, urbanization can aggravate total rainfall and contribute to increased flood risk. Heavy rainfall events associated with tropical cyclones over the United States have both increased and intensified due to anthropogenic forcings, the IPCC concludes with medium confidence. These can also be important for regional freshwater supply and annual precipitation amounts. Local tropical cyclone rainfall amounts depend on the cyclone speed and precipitation rate. A slower cyclone with a higher precipitation rate will have a higher rainfall accumulation than a fast-moving one with lower precipitation rates. Currently, there is limited evidence that cyclone speed has changed since the 1900s due to little agreement among models. Thus, there is low confidence in increasing cyclone intensity but robust evidence for increased tropical cyclone rainfall with global warming.

3. Conclusion

Water cycle intensification is expected to have profound consequences on human life, agriculture, and industry. Increasing global temperatures mean higher contrasts between wet and dry regions on the planet. Wet areas will experience heightened, more frequent precipitation rates, and dry regions will expand. This has critical implications for glaciers and snow, which are closely connected with streamflow and groundwater. For humans, this can escalate water scarcity and exacerbate inequalities of those who are most vulnerable to the negative impacts of climate change. There is still work to be done in improving model confidence regarding cyclone intensification and assessing trends in global river discharge, a process that may be sped up with evident impacts of flooding and rainfall in many countries around the globe.

Why should we expect water cycle change?

Rose Schoenfeld

Why should we expect water cycle changes? (pg. 15 – 32)

Global precipitation and evaporation (essentially the water cycle) are directly determined by the energy balance of the Earth. However at smaller scales, say smaller than 4000 km, the changes in the water cycle are dominated by moisture transports, dependent on thermodynamic and dynamic processes. Understanding the energy budget and moisture budgets serves to aid in the understanding of changes in the water budget as the global climate warms.

The Clausius-Clapeyron Equation, an equation that demonstrates the relationship between air temperature and the amount of water vapor that the air can hold, determines that specific humidity will increase by around 7% per degree Celsius of warming, that is if relative humidity holds constant, generally a good first approximation on a global scale.

Hydrological sensitivity is the linear change in global mean precipitation with global surface air temperature, once adjustments of the hydrological cycle occur due to radiative forcing (radiative forcing is the change in energy flux in the atmosphere and Earth’s surface, and can be caused by natural or anthropogenic/human caused factors). It is dependent on the atmospheric net radiative cooling and thermal deeping of the troposphere. This is limited by the cooling effect of evaporation and latent heat release in the atmosphere. Evaporation of water off a surface leads to a cooling of the surface. This is because due to the nature of evaporation, the higher energy molecules of water “escape” from the surface into the air which thus lowers the average molecular energy (temperature) of the surface. Liquids have bonds that hold their molecules together, meaning energy is required for evaporation to be sustained.

Hydrological sensitivity, the change in global mean precipitation associated with the change in global surface air temperature, is an important way to characterize the effect of climate change on the water cycle. Hydrological sensitivity is generally lower over land than over water. However, we do see land and ocean mean precipitation increases in projections. Precipitation increases with a warming climate can be offset some by cooling from aerosols and other sources, explaining why trends in global precipitation observed are small and difficult to interpret. Artificially reducing absorbed sunlight at the surface will not completely offset the precipitation changes from greenhouse gases however.

Apparent hydrological sensitivity (the global mean rate of precipitation change per observed degree Celsius change) is less than the hydrological sensitivity due to the influence of radiative forcing agents, that is rapid atmospheric adjustments to greenhouse gases and aerosols.

Scientists are highly confident that the global mean evaporation and precipitation will increase with warming temperatures, but the rate depends on the model, expected with a range of 2-3% per °C. This is however partly offset by the cooling effects of aerosols and atmospheric adjustments, and with greater effects over the ocean than land. Aerosols cool the atmosphere often by reflecting shortwave radiation back to space, rather than allowing the energy to be absorbed by the Earth surface or the atmosphere. Generally speaking the parts of the atmosphere circulation where air converges are expected to get wetter, and locations that involve much net evaporation are expected to become drier.

Increased moisture transport from ocean areas to regions of the atmospheric circulation with existing high precipitation will lead to salinity patterns strengthening over the ocean, however the effects over land are expected to be more complex. Overall higher levels of temperature increase over land causes shifts in atmospheric circulation and averages to a reduction in relative humidity near the surface, and a decrease in precipitation. Scientists are highly confident that the very wet and very dry seasons and patterns in weather will intensify with climate change and a warming planet. That is, wet gets wetter and dry gets drier.

Scientists predict that the differing atmospheric wind patterns due to radiative forcing and changing surface temperatures will affect the regional water cycle in most all regions. Tropical atmospheric circulation (the rising of air at the Equator and sinking of air at 30 degrees North and South) is expected to slow down in the future, but we may see some intermittent strengthening of the circulation due to internal variability. Generally there is an expected strengthening and narrowing of the ITCZ (intertropical convergence zone, a band of heavy and frequent rain approximately along the equator). Radiative forcing and warmer temperatures will also lead to a poleward expansion of storm tracks, however the mechanisms of drying in these regions is largely unknown.

Scientists are certain that warming will lead to the melting of ice water stores, everywhere where the temperature rises above freezing. Earlier snowmelt and less overall snow volume will cause changes in seasonal streamflow, additionally melting snowpack and glaciers contribute to an increase in nearby streamflow. It is expected that heavy precipitation events will intensify in the future, and increased moisture will lead to more rainfall during wet events like tropical cyclones and storms (around 7% per degree C of warming).

It is also likely that this increased intensity of rainfall, but lower frequency overall, will lead to greater surface runoff issues and flooding issues. Surface runoff is the flow of water on the surface of the ground, not underground. It occurs when the source of water is unable to infiltrate the ground, either due to the soil being saturated or slow absorption speeds of the ground. Surface runoff drives erosion of landscapes, and it also is the major cause of urban flooding. It can also lead to agriculture issues.

Warming climate leads to higher evaporative demand, and thus lowers soil moisture. Then as the levels of CO₂ increase, plants will increase their water use efficiency, and changes in evapotranspiration from plants can also affect soil moisture. Additionally deforestation will lead to a decrease in precipitation, and increase streamflow, and urbanization contributes to increased local rainfall and runoff potency. Then as temperature increases globally, the human need and consumption of water will likely deplete diminishing groundwater resources in the dry regions. Aerosols, often anthropogenic, can have an effect on precipitation. In addition to influencing balances of shortwave and longwave radiation, they can have an effect on cloud microphysics, often serving as cloud condensation nuclei (CCN).

What are the projected water cycle changes? (pg. 63 – 96)
Scientists are confident that the global water cycle will increase in intensity as global warming continues. The global water cycle consists of mean precipation, evaporation and runoff. It is projected that the global annual precipitation over land will increase by 2.4% with a low carbon emission scenario (SSP1-1.9), all the way up to 8.5% in the highest emission scenario (SSP5-8.5), by 2081-2100 compared to 1995-2014.

It is also expected that there will be persistent increases in global mean column integrated water vapor and near-surface specific humidity over land masses. It is also expected with medium confidence that we will see some decrease in near-surface relative humidity over land because of the many physical processes at play. Overall, scientists are expecting an increase in moisture transport into storm systems, monsoon rainfall, and into the higher latitudes.

Scientists are expecting an increase in precipitation overall with temperature rise because of the increase of greenhouse gasses and a decrease in shortwave reflecting air pollution. Thus it is expected that total precipitation will increase greatly in the high latitude regions, with the logical transition from snowfall to rainfall with higher temperature, excluding of course the coldest areas and times of year.

Precipitation is expected to decrease in some areas however, including areas such as the Mediterranean, southern Africa, Amazonia, Central America, southwestern South America, southwestern Australia and coastal West Africa, which has severe implications for people and ecosystems in these regions. On the flip side, monsoon precipitation is expected to intensify over South Asia, East Asia and central-eastern Sahel. Overall daily mean precipitation intensity will likely increase over much of the globe, however, the number of days without any precipitation is expected to increase in some regions like the subtropics, Amazonia, and Central America. All these effects can be summarized by an increase in precipitation variability over most of Earth’s land regions.

On the subject of monsoons, it is predicted that there will be delayed onsets and extinction of summer monsoons in North America. Also it is predicted that monsoon precipitation in East Asia will continue to increase in intensity and length. Additionally, in South America, it is likely that there will be delayed onset of monsoon seasons, but it is unclear how it will affect the total amount of precipitation in this region. Lastly, scientists project that in Australia monsoons precipitation will increase, also with an increase in rainfall variability and increased intensity of rainfall.

Scientists are medium confident that the yearly range of precipitation has elevated since the 1980s, but it is unclear if this is due to human influence. However there is high confidence the human caused decrease in springtime snow cover and glacier melt, have contributed to differences in streamflow, and human activities have had effects on the seasonality of water availability, including drier dry seasons.

Climate projects show that human caused forcings will lead to increased global mean evaporation over most ocean areas of Earth’s surface. There is a great elevation of atmospheric demand for water and thus an increase in evapotranspiration from plants over most land, save for a few moisture-limited areas. Granted there are many uncertainties in prediction of evapotranspiration, especially as scales that are smaller like seasonal and regional scales.

Scientists predict that the global runoff will largely elevate with increased global temperature, but there will be decent regional and seasonal variability in this. It is expected that runoff will increase in regions like the northern high latitudes and will decline in areas like the Mediterranean and southern Africa. It is likely that runoff will increase in areas of central and eastern Africa, but decrease in Central America, and areas in southern South America. The magnitude of the runoff change is dependent on emission increases. It is expected that the seasonal variations of runoff and streamflow will intensify with increased global surface temperatures in the subtropics.

Snowy areas are predicted to see that peak flows of spring snowmelt will start to occur sooner in the year, although the amount of runoff due to snow will decrease with lower amounts of total snow accumulation, except for places where the runoff is caused by glacier melt. Overall flooding amounts will increase, but this again varies based on location and the type of flooding. The prediction of effects of climate change are contingent of the human land-use and land-cover changes, which have the ability to greatly affect these trends.

It is predicted that the soil moisture will decrease in semi-arid locations where rainfall is most prevalent in the winter, this includes places like the Mediterranean, southwestern North America, southwest Australia, Central America, and the Amazon. Overall, these areas will likely become drier as a result of lower precipitation and an increase in the atmosphere’s evaporative demand for water (more water will be evaporated into the atmosphere). These areas will also see increased likelihood of drought and the droughts they do see will be higher severity. The extent of these changes depends on the amount of emissions, but even with the lower emission scenarios, scientists expect to see significant changes in drought and dryness, which will have effects on the availability of freshwater. The future holds changes in aridity for some locations that have not been seen in the last 1,000 years.

Glaciers are a point of major concern for climate scientists. They are projected to continue to melt at rapid rates, even with the lowest emission scenarios. The runoff from these glacier melts has been modeled to peak differently depending on the climate scenario, however the rate of the most glacier mass loss in low latitudes areas is consistently projected to occur within the next few decades for all climate scenarios. Runoff from the large glaciers will increase with higher temperatures as more of the glacier melts, that is until enough glacier mass is depleted that the melted water volume decreases. This would result in a peak of runoff and then a decrease. Glaciers in the poles are continuing to melt and lose mass over the next decades, and perhaps continue even after 2100.

Climate scientists are virtually certain that the snow coverage and duration in the Northern Hemisphere will continue to fall with rising global temperatures. It is likely that the studies on the snow in the Northern Hemisphere will translate to the Southern Hemisphere as well, and we will likely see much of the same effects. Snow is also predicted to continue to melt earlier and earlier in the year. These altered timing and volume of snowmelt will affect the water cycle in these regions, impacting runoff, soil moisture, and evapotranspiration.

In the wetland regions, it is expected that there will be decrease in precipitation and evaporation will increase, and a rise in seawater will lead to higher saltwater intrusion into wetlands on the coast. However it is largely unclear how much the sea level rise will influence the coastal wetlands extent. For lakes, scientists are predicting a decrease in overall ice.

In the tropics, the tropical atmospheric circulation is expected to weaken, which would result in the narrowing ITCZ (intertropical convergence zone) and the core of the band will intensify. The shifting of the ITCZ will be included by regional changes. The Hadley cell is projected to weaken overall, and thus the cells are predicted to expand poleward, thus changing the same and likely having climatic effects on the regions affected by this circulation. The Walker circulation (which is an atmospheric air flow cycle over the equatorial Pacific Ocean) is predicted to weaken and this will lead to reduced precipitation over the western tropical Pacific. Smaller and shorter-term strengthening of the Walker circulation may be attributed to internal variability.

It is predicted that rainfall will increase for eastern-central African regions, but decrease in the West, with a delay on the onset of the wet season. The uncertainty however will increase for this prediction for the higher emission climate scenarios. There is high confidence that strong precipitation from tropical cyclones will increase, this is because of processes that are related to increased low-level moisture conference and water vapor in the environment. It is also predicted that precipitation from extratropical storms will increase as temperatures rise in most locations.

In the Southern Hemisphere, we will see the storm track shift poleward, but the same effect is not expected for the North Pacific storm track or the North Atlantic Storm Track.

Atmospheric rivers are expected to increase in magnitude and duration in the future. This leads to overall increased precipitation. This would have the most effects for places such as the west coast of the United States, and western Europe. Atmospheric rivers for locations such as the United States west coast contribute up to 50% of the total rainfall for some of these regions, so an increase in both magnitude and duration in these events would have great consequences on the water cycle for these areas.

The ENSO (El Niño–Southern Oscillation, an irregular but periodic variation in winds and sea surface temperatures over the tropical eastern Pacific Ocean, has effects on the climate of the tropics and subtropics), is expected to have altered influence on precipitation. The rainfall variability due to the El Niño–Southern Oscillation is expected to increase by the 2nd half of the 21st century in many climate models. The MJO (Madden-Julian Oscillation which affects the mid-latitudes), is expected to intensity with global warming, with greater amounts of associated precipitation.

In summary, global warming is expected to lead to continued strengthening of the global water cycle, which includes its variability, monsoon precipitation, and the extent and intensity of wet and dry climatic and weather events. Evidence indicates that precipitation and surface water runoff will become more variable with time over land areas, both seasonally and with each passing year. Average mean precipitation over global land is predicted to increase from 0-5% with the low greenhouse gasses emissions scenario (SSP1-1.9), 1.5-8% for the intermedia emission scenario (SSP2-4.5), and 1-13% for the highest emission scenario (SSP5-8.5) by 2081–2100 compared to 1995-2014.

These changes in the water cycle will have severe effects on ecosystems, human life, and the Earth system as we know it. The increased severity of events such as droughts, atmospheric rivers, runoff, flooding, etc have drastic consequences. As the wet gets wetter and dry gets drier, it has implications for global food production, fresh water availability, urban flooding, agricultural flooding, wildfire risk due to dry conditions, heatwaves, and hurricanes. In addition, the melting of glaciers leads to runoff issues, but also contributes to sea level rise, which has consequences for many populations.

It is imperative that the global community acknowledges the gravity of these projected changes in the water cycle and the changes we have already observed, and disregards the interests of Capitalism and economic pursuits that aim to drain the Earth of its resources and balance, and take the needed policy actions to best mitigate climate change.

Stories Beyond the Models: Understanding Limitations in Hydrologic Forecasts with Climate Change

adi stein

IPCC 6 Chapter 8 – Summary Part 3
Understanding changes in the water cycle (also known as the hydrologic cycle) with climate change requires an understanding of how the models used to describe these changes function on a fundamental level. While it is important to have a grasp of the hydrologic cycle to parse the science behind these models, knowing what limits the models have and how they are handled allows for a dialogue with the IPCC report instead of an idolization of it. Engaging with the literature is part of the scientific process of questioning published results. Then we can better understand how these models change under extreme events that test the limits of the models or challenge their assumptions of the future. Questioning and pushing results is not a direct refutation of science that has been develop by professionals for many years but a curiosity to comprehend the science instead of meaninglessly regurgitate it or accidentally misquote it.

This paper will provide context for and summarize “What are the limits for projecting water cycle changes?” (section 8.5) and “What is the potential for abrupt change?” (section 8.6) in the Intergovernmental Panel on Climate Change (IPCC) Sith Assessment Report Chapter 8. It will describe the goals of modeling, complications with climate change, limits in hydrologic modeling, how abrupt change in the future may alter models, and how extreme hydrologic events will change with climate change.

Introduction to Modeling
In general, modeling is the development of a process to describe one event given some information. Often scientists use mathematics as a tool to create models through rigorous testing in experiments and understandings of the world. There are two main types of models: an empirical model and a theoretical model. Empirical models use data to describe relationships between two or more measures. An example is using rainfall to describe streamflow in a nearby stream. Understanding the properties of the soil, current moisture conditions, how much rainfall there is, and other hydrologic processes can allow hydrologists to develop an empirical model that allows them to compute streamflow for a given precipitation. This might take the form of a statistical estimation (in winter X mm of precipitation results in Y cubic feet per second of streamflow on average), a derived equation (0.4*X mm of precipitation = Y cubic feet per second of flow), or some other relationship. A theoretical model on the other hand aims to describe how and why a process occurs. In hydrology, theoretical models often rely on a physical understanding of a phenomenon, such as knowing that rainfall can run off the land and into the stream channel to create streamflow. Theoretical models can be the basis for empirical models and vice versa.

A key aspect of the scientific process is being able to rigorously test results and ensure they can be repeated. In our streamflow example, hydrologists can measure the actual streamflow (observed streamflow) and compare it to their model’s prediction to see how well their model did. The difference between the predicted value and the observed value is called bias. Accurate predictions have a bias close to zero, meaning they were close to the true value that they were predicting. Precise predictions have a small spread, or difference among predictions when computed multiple times. A model should be accurate and precise before being used to make decisions. But we may not always be able to gather observations to determine the accuracy of our models.

Not every stream has a stream gauge to measure flow. Various factors can cause this such as the inability to afford more sensors or personnel, trying to minimize environmental disturbance, or measuring streamflow is not a community priority due to other factors. Scientists may then focus on larger scales where data does exist or strive for a process-based approach that makes their model more adaptive by using the underlying physics. While understanding the underlying processes may be more desirable, there may not be enough known about the system to describe such processes. The underlying processes themselves may also be too complicated to predict or still being researched and scientists do not currently know how to describe all the processes. This can result in scientists acknowledging the complexity of certain processes and leaving them out of their model in favor of not misrepresenting what they cannot accurately describe. Hydrologists may do this by assuming there are no dams in the watershed they are modeling, which still allows them to gather insight about the natural process of the watershed but fails to describe how damming impacts stream temperature for example. There is still knowledge gained in the process of modeling, but we cannot apply the model without understanding the assumptions that went into it. Due to the complexity of our world, it can be difficult to near impossible to capture every process, but we still want to model it and gain some understanding instead of no understanding by refusing to try. Being able to acknowledge the flaws and still have a discourse is where George E.P. Box’s famous quote “All models are wrong, but some are useful” shines in climate science.

Climate Change & Modeling
While climate change is occurring today and currently impacting people, many climate models are interested in how human caused climate change will impact our world in the future. Yet this poses a problem when considering the need to rigorously test our models as we cannot compare forecasts (predictions about the future) with observations until the observations happen. Historically scientists have assumed that climate data gathered in the past will inform us how the climate behaves in the future. This assumes that the historical climate is the same as future climate, or that climate itself will not change with time, often referred to as the stationarity assumption. Using this assumption with climate change is illogical however as we know the climate is changing with time. But at the same time, we need information to guide the decisions that are taken with respective to climate change.

Developing models that do not rely on the stationarity assumption involves a variety of methods, from flexible statistical approaches to more process-based understandings. Creating flexibility allows models to respond to greenhouse gas emission projections and effects for various emission scenarios as described in CIMP6. Yet there are many different complex parts to modeling the climate that scientists are still working to understand. Therefore, we need to consider what limitations exist in these models for hydrology to acknowledge why different models have different forecasts and where the models can differ from reality.

Hydrological Modeling Limitations
IPCC 6 Chapter 8 finds atmospheric convection, cloud-aerosol interactions, and land surface processes to be the limiting factors in their hydrologic modeling. These limitations prescribe a wider range of possibilities by understanding how they impact the models instead of simply discounting the models because of them.

Atmospheric convection, how heat, moisture, and energy move throughout the atmosphere, is key to water transportation in the hydrologic cycle. As air rises it cools, resulting in moisture condensing (changing from a gas to liquid phase), and forming small droplets or ice crystals that compose clouds. On a global scale, this rising motion occurs near the equator and moves towards the cooler poles. Yet moisture does not stay in the atmosphere long and eventually falls out of the atmosphere as rain, snow, hail, sleet, or other forms of precipitation. This convective motion is further governed by the Hadley Cells, circular patterns of moving air in the atmosphere. Modeling is restricted by the computational cost of computing small-scale cloud processes, difficulty simulating the day-night (diurnal) cycle over land, and difficulty accurately representing tropical precipitation. This limitation lowers result confidence in the tropics, but can be improved through weighing multiple models and focusing on specific weather events.

Aerosols, a solid or liquid particle so small it is suspended in the air, impact cloud formation and lifetime. Tiny particles, such as dust or pollution, change how liquid drops or ice crystals form clouds in the atmosphere and can allow water to accumulate more before precipitation. The aerosols provide more surfaces for condensation, spreading out how much is gathered in one spot and distributing the weight of condensed water. Spreading out the water leads to longer times before enough water is gathered in a single cloud condensation nucleus (CCN) to cause precipitation. While aerosols are known to impact precipitation, representing impacts on deep clouds is complex and difficult to model accurately. Cloud microphysics is not as well understood in the contexts needed for global climate models (GCMs). Forecasting aerosol impacts on clouds in the future is therefore limited until more advanced techniques are developed. Integrating high-level knowledge of aerosol effects on clouds allows for hypothesizing about future precipitation impacts not captured in the models.

Land surface processes describe not only how water moves into rivers but also the evaporation of it into the atmosphere as well as how carbon storages change at the surface. Current GCMs are including new and improved processes such as soil freezing, permafrost, glaciers, surface water modeling, vegetation, and the soil-plant interface. While progress has been made describing root dynamics, allowing for better understanding of plant roles, dynamic vegetation models need further work to describe plant feedbacks. Through evapotranspiration, plants draw water from the ground and expel it as water vapor. Yet plants being impacted by wildfires, insects, soil chemistry, and atmospheric ozone content is missing from models.

Plant responses to increased carbon dioxide concentrations is included however. Most models underestimate the role of plants in the hydrologic cycle, consequentially misrepresenting the relationships between precipitation and evapotranspiration. Unfortunately groundwater is not accurately represented in many models, limiting understandings of interactions between the surface and subsurface. Land management is also missing from models even knowing land use increasingly impacts the water cycle. Despite improvements in modeling, missing these key understandings of land surface processes still leaves much room for improvement in GCM forecasts of the hydrologic cycle.

While limitations these processes may describe hydrologic models as insufficient, understanding them allows decisions to factor in what is missing from forecasts. Current models represent a world without the limitations described. When we draw stories from them, we can theorize how longer cloud lifetimes or changes in vegetation impacts our forecasts.

Potential for Abrupt Change
Making decisions not only requires insight on modeling limitations, but also how abrupt changes impact our forecasts. Abrupt changes often arise from systems reach an unstable point and cross a threshold that rapidly changes the system from one to another. Different climate scenarios aim to describe the different possible regimes our climate could end up in based on different levels of human-caused warming. IPCC 6 Chapter 8 outlines concerns about abrupt changes in the Atlantic Meridional Overturning Circulation, land surface, and solar radiation modification.

The Atlantic Meridional Overturning Circulation (AMOC), a large system of ocean currents driving oceanic circulation, impacts temperature movement as a key aspect of the hydrologic cycle. AMOC severely weakening or collapsing will cause global changes in precipitation, storm tracks, and monsoon intensity. Greenhouse gas emissions will weaken the AMOC by 2100 but are not likely to cause an abrupt collapse before then. Collapse could result in the drying of the Sahel, Mesoamerica, northern Amazonian regions, and Europe.

Changes in the Amazon forest, vegetation of the Sahara and the Sahel, and dust in the atmosphere can serve as land surface tipping points. Deforestation of the Amazon increases the possibility of disastrous fires that, in combination with drier conditions, can destabilize the rainforest’s ecosystems. The Amazon rainforest being the largest forest in the world serves as a major recycler of atmospheric moisture, heavily influencing the hydrologic cycle. Meanwhile the Sahara and Sahel in North Africa serve as a dry surface reflecting atmospheric radiation. The Sahara and Sahel reduce warming absorbed by the surface by reflecting it back into the atmosphere and space. Increased precipitation in the region due to human-caused climate change can increase precipitation growth and lower the reflectivity of the surface. Unfortunately the probability of a green Sahara or Sahel cannot be accurately modeled due to the land surface limitations previously described. Similarly limited is how dust, as aerosols, will impact future precipitation. Aerosols can affect precipitation but how is not well understood nor modeled. All three have the theoretical potential to vastly, and suddenly, impact the hydrologic cycle, but how is limited by our models. Therefore decisions made or avoided surrounding them should be approached with caution beyond what the models entail.

Solar radiation modification (SRM) is an engineering technique that aims to change either how much radiation is reflected or how much heat is lost to space. Such techniques could include injecting aerosols into the atmosphere, increasing cloud reflectivity, altering cirrus clouds, or increasing the reflectivity of the surface. While aerosols may reduce radiation, they can have detrimental effects on precipitation. In the starting or stopping of these methods does not solve the presence of human-elevated levels of carbon dioxide, but does impact hydrology. Rapid changes can occur from failures to implement SRM or political conflict surrounding their use.

Abrupt changes and their possibilities should be considered in junction with climate forecasts. Different warming scenarios can capture how some abrupt changes will influence the atmosphere, but changes limited in modeling cannot be numerically described. Therefore narrating potential futures with human-caused climate change should take additional steps to include these changes similar to model limitations.

Beyond Modeling
Models provide scientists, decision makers, and the public with tools to understand what can happen with human-caused climate change. Due to the complex nature of the environment however, models cannot capture everything. The development of science and technology is increasing at an increasing rate, but some processes are still not well understood. Climate change is a unique problem because we do not have a historical record of how warming occurring at such a rapid rate impacts the world and the people that now live in it. Yet climate change still impacts us today and will only become more demanding. So models are still formed to estimate the future with the knowledge we have and the knowledge of what we do not know.

Limitations and abrupt changes to forecasts invite a more nuanced analysis of the IPCC 6 report. The report does not stop at model outputs, but narratives of how the outputs describe our future. By considering what limits the models and what can cause the systems they represent to suddenly change, we can indue decisions with the caution they deserve. Engaging with the public about the challenges scientists face lowers the communication barrier between the two. Discussing concerns sooner presents less surprise should the models fail to capture a future and reassure that scenarios beyond the models have been considered. Models may not be right, but they can be useful when approached critically. The hydrologic cycle impacts all of us, so we cannot ignore climate change.