Wherein we review a study on projected climate impacts to irrigated agriculture and the CIRCulator’s Managing Editor says goodbye.
It is now generally acknowledged that warming tied to human-induced climate change is going to hit our species where it will most assuredly count: in the fields that produce our foods, potentially impacting such tabletop essentials as our daily bread, corn tortillas, and beds of rice. This is especially true for irrigated agriculture that relies on melting snow.
As we have reported extensively on this blog, mountain snow, or snowpack, is very important for irrigated agriculture here in the Pacific Northwest and the American West generally. We currently rely on snow accumulating on our mountains during the colder months, melting and flowing downstream during the warmer months, and irrigating our crops during their growing season in the process. However, the American West has also seen a decline in our snowpack, a decline that is expected to continue in the decades ahead under projected future warming. This decline in mountain snow is going to impact how we grow crops here in the Pacific Northwest. And as it turns out, we are not alone.
Worldwide reliance on irrigation derived from snowmelt and how that reliance will be, well…less reliable under projected future warming is the subject of a recent paper in the journal Nature Climate Change.
The study, led by Yue Qin and including CIRC’s John Abatzoglou, examines the projected impacts that a world that is either 2 or 4 degrees Celsius (3.6 or 7.2 degrees Fahrenheit respectively) warmer than preindustrial levels will have on worldwide snowmelt-dependent irrigated agriculture. (If you’re curious, globally we’re currently already over 1°C warmer than preindustrial levels.)
All of this matters, because while irrigated agriculture accounts for only abut 20–25% of all harvested crops, those crops make up about 30–40% of all the food calories consumed by our species, including wheat, rice, and corn, all of which rely to a considerable degree on snowmelt-derived irrigation, according to this study’s analysis.
As you might expect, some regions and crops are currently more dependent on snowmelt-derived water than others and, thus, are more susceptible to future spikes in the mercury. This is what Qin and colleagues determined as well. The researchers also concluded—again, as you would expect—that impacts to irrigation will be felt worldwide whether we see 2°C of warming or 4°C of warming, but that impacts under 4°C are projected to be the worse of the two scenarios. And while none of this is exactly a newsflash for many of our readers, what makes Qin and colleagues’ study worth your time are the details they delve into. For instance, to say that Qin and colleagues’ study is global in its scope is perhaps a bit misleading. The study is also, in a sense, hyper-local.
Basin by Basin
To determine the impact of future warming on snowmelt-dependent agriculture worldwide, Qin and colleagues had to first examine the problem one region at time, which in this case meant one hydrologic basin at a time. The researchers did this by first determining which basins—from the Columbia River to the Syr Darya Basin (think Kazakhstan and Uzbekistan)—are currently most reliant on snowmelt for irrigation. The researchers then used data from future climate projections to determine how 2°C of warming or 4°C of warming could affect each basin’s snowmelt in the future.
Then the researchers went the extra mile by assessing each snowmelt-dependent basin’s probable ability to store the water they would need to meet their irrigation demands in the future when their snowmelt becomes less reliable.
And, as if that weren’t enough, the researchers also assessed the current snowmelt-dependence of key crops. This was done not only for staples, including wheat, rice, and corn, but also for crops that provide the raw ingredients for our daily liquid vices, including coffee, grapes, barley, and sugar.
But before we get the study’s conclusions, let’s consider the essentials you will need to know about snowmelt and agriculture. Like a good rock show, it really comes down to timing and volume.
Snowmelt Timing and Volume
A large number of previous studies on the impacts of climate change to irrigated agriculture have tended to examine projected changes to annual precipitation. But as any farmer—or for that matter, any would-be gardener—will tell you, the right time to water your crops is when they need water. It’s a bit of a no-brainer, but we’ll just spell it out anyway: growing crops is a seasonal thing, and as a consequence, so is the demand for irrigated water in agriculture.
With the importance of irrigation timing in mind, Qin and colleagues chose to examine projected seasonal variations in crop water use, precipitation, and runoff. We should note here that the researchers were only interested in surface water,meaning water that falls to Earth as either rain or snow, water that is either used directly to meet crop water needs, runs off, or is stored as snowmelt for later use.
Surface water, as the name suggests, is water that literally resides on the Earth’s surface in streams, rivers, and snowy peaks. It is distinct from groundwater, which the authors didn’t look at for their study. Whether surface water originates as rain or snow matters a great deal to both the timing and volume of water available for irrigation. Understanding why this is so will make what follows easier to understand. Consider the Pacific Northwest.
(For readers familiar with our posts, what follows is our typical snow-is-so-very-important-for-agriculture-and-ecosystems-in-the-Pacific-Northwest-and-American-West-generally-and-climate-change-is-making-that-snow-less-reliable shtick. Feel free to skip ahead if you like.)
Here in the Pacific Northwest, we tend to have wet, dark, cool winters followed by dry, bright, warm summers. These cool winter temps mean that while our lower elevation cities get drenched with rain, our higher elevation wild lands in our mountains and foothills tend to get snow. This snow is a natural, albeit a seasonal, reservoir.
By slowly melting over the spring and summer months, snowpack provides water during the Pacific Northwest’s warmest and driest time of year when we grow the majority of our crops. We have come to rely on this snowmelt for exactly this reason. However, due to rising temperatures, snow is not only more likely to melt earlier in the calendar year, but precipitation that used to fall as snow is also becoming more likely to fall as rain.
When this happens, our regional snow reservoir not only gets smaller but all that previously stored water gets added to the hydrologic system earlier in the calendar year. (If you think of it like a bank account, we are essentially forced into making a withdrawal when we really should be making a deposit.)
For the Pacific Northwest under projected future warming, less snow means we will likely see larger flows of water in our rivers and streams during the colder months—when we really don’t need the water for irrigation—and smaller flows in our rivers and streams during our warmer months—when we really very certainly do need that water (and not only for irrigation, but also to protect engendered fish species).
Because crops need water when they need it, this means our water supply is likely to get out of synch with our agricultural demand on that supply. From the Columbia River to Syr Darya, this supply-and-demand-out-of-synch problem is what Qin and colleagues sought to calculate.
Snowmelt Supply and Demand Out of Synch
So, how does current worldwide snowmelt-dependent water supply and demand line up with future worldwide snowmelt-dependent water supply and demand?
To figure this out, Qin and colleagues determined each basin’s supply of and demand for irrigated surface water over the historical period 1985–2015. This set their basins’ baselines. Of key importance to each basin was not the total historical volume of surface water supply (though that is important), but what proportion of that water originated as either rain or snow.
Because that still wasn’t enough to understand the role snowmelt has played in irrigated agriculture, the researchers then determined basin by basin how much of historical demand has been met by snowmelt-derived and rain-derived surface water, respectively.
As you might imagine, over the historical period, there is a more-or-less straightforward nearly one-to-one match up for when in the calendar year a supply of rain- or snowmelt-derived water is added to a basin’s hydrologic system and when it is consumed for irrigated agriculture.
Consider an example from the American West: California’s heavily irrigated San Joaquin Basin. During the summer months, the basin’s farmers rely on surface water that comes almost entirely from snowmelt from the Sierras to irrigate their crops. San Joaquin’s current snowmelt-derived water peaks in June and correspondingly gets consumed in June. All this gets out of synch in a warmer future as mountain snowpack reservoirs decline and melt earlier in the year, according to Qin and colleagues’ analysis. And this is basically the pattern that the researchers found in multiple basins.
Using their historical baseline analysis, Qin and colleagues then used a series of databases and climate projection models to determine how far out of synch from their historical baselines some basins might get under either 2°C or 4°C of warming above preindustrial (in this case 1850–1879) levels.
As noted above, Qin and colleagues determined that the timing of water supply and water demand would get out of synch under both 2°C and 4°C when compared to the historical period. However, the researchers determined that the asynchronous effects would be far worse under 4°C of future warming when compared to 2°C of future warming.
Consider again the San Joaquin Basin. Under projected warming the researchers estimated that the peak snowmelt directly available for irrigation will be reduced and be delivered earlier in the calendar year. If we spell this out, we see that the timing of San Joaquin’s peak snowmelt dominated irrigation will be out of synch with future snowmelt runoff by as much as several weeks earlier under 2°C of warming and as much as a month earlier under 4°C of warming.
Snowmelt-Dependent Regions and Crops
To determine which basins were most at risk to current warming, Qin and colleagues identified snowmelt-dependent basins by defining them on a scale that quantified just how much of the total volume of irrigation demand per basin was currently being met by snowmelt. Again, this was done the world over. This allowed the researchers to identify the world’s “hotspots” of snowmelt-dependent agriculture. (Yes, you read that right, snowmelt hotspots.)
The most snowmelt-dependent basins—the hottest snowmelt hotspots, if you will—were those that currently receive up to 30% of their irrigation surface water demand from snowmelt. This includes such notables as the Colorado River Basin and—you guessed it!—the San Joaquin Basin, not to mention absurdly huge regions of Eurasia. These were followed by basins that currently receive up to 20%, of their surface water irrigation from snowmelt, and so on down the scale. The twenty-percenters include the Columbia River Basin, as well as a scattering of basins from Eurasia to South America.
The researchers than used a similar snowmelt-dependency scale to determine the vulnerability of specific crops to future warming. But here they added a slight twist: the crop scale calculated the current percentage of snowmelt demand for each crop month by month. For instance, worldwide wheat production currently gets 45% of its water needs from snowmelt during April and 49% of its water needs from snowmelt in May.
(By the way, the visualization of these data is absolutely beautiful. The yellowish green-and-blue-toned boxes created by the designer look like tile work you might find in a hip modern kitchen. The plus is, if you were to organize said titles in your own hip kitchen exactly as the boxes are shown in the study, you could be forever reminded that your morning coffee is not really reliant on snowmelt at all, while the sugar you put in your morning brew gets a whopping 30% of its water from snow in June worldwide…and let’s not even talk about those cotton pajamas. Yikes!)
Basins at Risk of Lost Snowmelt
Of course whether a basin is currently dependent on snowmelt for irrigation doesn’t tell the whole story of how that basin might be impacted under future warming. What matters is not only how dependent a basin currently is on a snowmelt-derived water and not only how far out of synch future supplies of snowmelt are projected to be from current snowmelt demand, but also whether or not a basin can realistically find new ways to store more water when their natural snowpack mountain reservoirs drain earlier in the year or disappear entirely. This combination of factors is essentially what Qin and colleagues call their snowmelt hazard index.
In a severely simplified nutshell, the researchers calculated their snowmelt hazard index by combining all their previous analyses together (current snowmelt dependence and projected changes in supply and demand under future warming, etc.) with a key new extra bit, namely storage capacity, measured in the aptly named reservoir storage index. (Note: while the researchers acknowledge the importance of ground water pumping to solve—even if only temporarily—the problem of surface water shortages, they did not explore this subject in their analysis.)
Not surprisingly, regions that scored high in two or more of the three categories earned large snowmelt hazard index ratings. These included the Syr Darya and Amu Darya Basins of Asia, the Rhone-Ebro basin in Europe, and both the San Joaquin and Colorado River Basins of the American West, to name a few.
The Columbia River Basin, which scored better than its neighbors to the south and east (or maybe we should say less worse), got a middling-but-still-worrisome snow hazard scale rating, scoring next to the Tigris/Euphrates Basin. (Yes, you read that correctly as well. The Pacific Northwest’s future water woes were calculated to be on par with a key basin in the Middle East.)
Qin and colleagues end their paper by calling for both more research to better understand the changes and vulnerabilities to snowmelt-dependent irrigation they identified and for reductions of greenhouse gases in order to avoid the impacts to irrigation they outlined.
Considering the human impacts to our daily bread, corn, and rice, as well as our daily coffee, wine, beer, and all kinds of sugary sweets, we completely agree.
Publication: Qin, Yue, John T. Abatzoglou, Stefan Siebert, Laurie S. Huning, Amir AghaKouchak, Justin S. Mankin, Chaopeng Hong, Dan Tong, Steven J. Davis, and Nathaniel D. Mueller. “Agricultural risks from changing snowmelt.” Nature Climate Change 10, no. 5 (2020): 459-465. https://www.nature.com/articles/s41558-020-0746-8.
Featured Image: Featured Image: “Mt. Hood from Timberline Lodge.” (Photo Credit: Mike Fisher, some rights reserved.) (Yeah…the photo isn’t that relevant to the subject of the post. But it does have snow.)
Nathan Gilles is the managing editor of The Climate Circulator, and oversees CIRC’s social media accounts and website. When he’s not writing for CIRC, Nathan works as a freelance science writer. Other posts by this author.