California’s drought: A result of natural hydrologic variability, climate change, or both?

Panel 1With exceptionally dry conditions continuing with really no end in sight, drought was the main topic at the San Gabriel Valley Water Forum held October 2nd in Pomona.  The forum brought together experts and local water officials to discuss the multiple impacts of drought and how region can best be prepared.

The forum opened with a panel of experts discussing the current drought conditions. Is the current drought a sign of climate change or merely just a reflection of the state’s natural variability? Glen MacDonald from UCLA’s Institute of Environment and Sustainability, Dan Cayan from Scripps Institute of Oceanography, Alex Hall from UCLA’s Climate Change Project, and JT Reager from JPL take a look at the science and the data behind the drought.

MacDonald 1Glen MacDonald

UCLA Institute of the Environment and Sustainability; Professor of Geography, Ecology, and Evolutionary Biology at UCLA

Panel1-1-MacDonaldG_Page_02Glen MacDonald began by showing a graphic of the latest U.S. Drought Monitor. “I know that earth colors are very popular for decorating right now, but really those reds, oranges, and yellows are not a good sign,” he said. “Take a look at California – virtually the entire state is in exceptional to severe drought, and we all know that, but this is not something that’s just affecting California, it’s winding across the southwest.”

He noted that the inset shows a reservoir map and pointed out that the state reservoir capacity is about 39% full right now. “We would be low this time of year anyway, but that’s roughly about 56% of where we actually should be,” he said. “Of course, some reservoirs are much, much below that. We’re a little bit better south of the Tehachapis, but things are grim in terms of our surface storage.”

Panel1-1-MacDonaldG_Page_03Is this exceptional drought a result of climate change? “In terms of precipitation, this is not exceptional,” he said, presenting a slide showing a time series of precipitation records for California. “Last year, from October to April 2013-2014, as you can see, is not exceptionally low. It’s low, but we’ve had years of lower precipitation here in the state of California. Our long term mean, the red trace, is not exceptionally low compared to some of the other low precipitation periods we’ve experienced, so in terms of precipitation alone, this is not a record breaker.”

Panel1-1-MacDonaldG_Page_04If we take a look at temperature, it’s a little bit of a different story,” he said, presenting a slide depicting temperatures. “What you can see is that we’ve been on a long term haul upwards. It’s more attenuated near the coast, and more severe as you move into the eastern portions of the state, but since the 1980s, we’ve been in a period of prolonged high temperatures. The temperatures last winter were very, very high. We’re headed for potentially a record breaking warm year in 2014, so in terms of precipitation, this wasn’t a record breaker, but in terms of prolonged high temperatures, we’re seeing something we haven’t seen over the instrumental period.”

Panel1-1-MacDonaldG_Page_05Mr. MacDonald then presented a slide of time series for the Palmer Drought Severity Index (PDSI), a complex measure of drought that incorporates temperature, precipitation and soil moisture data. “Absolutely, when we couple those high temperatures and evaporation rates with the low precipitation, which is what the Palmer Drought Severity Index does, than we are indeed off the charts,” he said. “We have not had an instrumental period, a winter or a year, which is a dry as what we’re experiencing right now. So indeed, for the Palmer Drought Severity Index, this is a record breaker over the instrumental period, but remember that it’s a combination of two things, precipitation and temperature.”

Panel1-1-MacDonaldG_Page_06He also noted that from about 2000-2001, the southwestern United States has been in a period of prolonged aridity. “The top map show temperatures from the beginning of the century to 2010; we’ve had high temperatures compared to the 20th century in the southwest and California. The middle frame is precipitation; we are a nation divided with precipitation. We’ve had higher than normal precipitation around the Great Lakes and northeastern North America, whereas in the west and southwest in particular, lower than normal precipitation. And then Palmer Drought Severity Index represents that.”

So we are in a period of a decade or more now of general aridity in the southwest,” Mr. MacDonald said. “There has not been one year since the beginning of the century, when a large portion of the southwest was not in drought conditions compared to average conditions over the 20thcentury.”

Panel1-1-MacDonaldG_Page_07It’s misleading to think of droughts as three, two, or one year cycles; we’re into something longer than that, he said. “The best evidence of this is to take a look at Lake Mead. Right now we’re at about 1081 feet, and since the beginning of the century, we have been on a downward spiral in terms of elevation and storage capacity. Six feet more, at 1075 feet, we will be in a first federal water shortage declaration, which will mean a 4.4% cut to the lower basin states. Now we have senior water rights in California, we would not take the cut at that point, but Nevada, Arizona and Mexico will. Will they do that quietly? I don’t think so.” He noted that the earliest that declaration could be put into place would probably be in January of 2016, unless there’s good precipitation in the Upper Colorado River Basin this winter.

Panel1-1-MacDonaldG_Page_08Using tree ring records, we see that what we’re experiencing is not a rare event when we go back over 1000 years or so,” said Mr. MacDonald, presenting a slide showing data derived from tree ring records. “Through natural forces, you can get a drought that is this severe and is prolonged, so we can’t say that this is a smoking gun for climate change; however, the high temperatures which are really exacerbating this can be blamed on climate change, so my sense is that we are looking at a climate change drought when we consider that it is the high temperatures that is really driving the PDSI here.”

Anywhere in the state of California we have the potential for these prolonged droughts, so we have to plan for something more than a three or five year cycle,” he said. “Is there any reason why this drought should end in three years or should end in five years? We know that due to natural causes, these droughts can actually be prolonged more than that. It doesn’t mean that it has to be, but we know that it is, and typically when they are more prolonged than that, it’s because we have very warm conditions.”

Panel1-1-MacDonaldG_Page_10We can’t necessarily look to El Niño to save us, Mr. MacDonald said, presenting a slide depicting both ENSO and California precipitation. “There is a relationship when we have very, very vigorous, warm eastern Pacific waters during an El Niño event, but in general, El Niño explains a small amount of the variability in precipitation across the state,” he said. “There can be El Niño years where we have just normal precipitation, and there could be small El Niño years where we have below normal precipitation.”

In order to break the drought, it needs to be a strong El Niño, such as the one in 1997, he said. He noted that in 1997, there was a warming of about 2.5 to 2.7 degrees Celsius. “Right now, projections for El Nino are 60-65% projection that by December, we’ll be in El Niño,” he said. “Right now we’re still in a neutral state. The projection for how warming is about 1 to 1.25 degrees, or less than half of what we need to put us into one of those bumper crop El Niño’s. Now things may change, but right now the projections are not super great that we’re going to have a big El Niño year.”

Panel1-1-MacDonaldG_Page_11He then presented NOAA’s projected long term outlook for precipitation and temperature for the next three months, noting that they are projecting more precipitation in parts of the southwest, but dry in the north, along with higher temperatures. “I don’t think we can look at the moment at the data the we have that it’s necessary going to be a big break coming our way,” Mr. MacDonald concluded. “It could happen; it’s very episodic here in terms of precipitation, but I don’t think right now we should be sanguine.”

Cayan 1Dan Cayan

Research meteorologist with the USGS and Scripps Institution of Oceanography

Dan Cayan began by saying he would talk about the broader scale, focusing largely on climate change. “My key points are that we are essentially moving into a future that is not informed by our past. We’re going to see climate that really takes us beyond the envelope of variability, if we can believe the climate model projections,” he said. “The other point I want to make is that California is naturally extremely volatile climate in terms of precipitation with ups and downs from year to year. That’s going to expose us to droughts and to the opposite side, floods. And finally, the incidence of the extremely heavy precipitation events in our state is really critical as far as generating the water supply. A few large events, whether they come or not, really make a difference.”

Pages from Panel1-2-Cayan_Page_02He then presented a slide with a time series of temperature, with the right hand side being a swarm of model projections for future temperatures. He noted that they don’t look at just one model, but an ensemble of models and simulations because of the uncertainties due to model imperfections, emission scenarios, and natural variability. “If you compare the future to the historical past, we are looking at a future by the end of the century whose temperature will likely be at least 3 degrees Fahrenheit if not more than 5 degrees Fahrenheit greater than levels that we see today,” he said.

He noted that this chart is for the month of July. “What’s emerging from climate models is the fact that summer temperature warming is greater than cool season warming and of course, that has a lot of impact as far as water demand, human health, the ecosystem and so forth.”

Pages from Panel1-2-Cayan_Page_03He then presented a slide depicting projected temperature and precipitation changes for Southern California, explaining that the horizontal axis is temperature change and the vertical axis is precipitation change, with the slice across the middle of the graph representing zero precipitation change.  “If you’re below that, it means you’re getting drier; if you’re above, you’re getting wetter,” he said.  “This is the early part of the 21st century according to a set of about 30 climate models.”

Pages from Panel1-2-Cayan_Page_04He then presented another chart depicting model results for temperature for the middle part of the century, noting that the amount of warming is on the X axis and the amount of precipitation is shown on the Y axis. “Roughly half of the model solutions are getting wetter, and roughly half, perhaps a little more than half show getting drier, and of course that’s discomforting,” he said.

Pages from Panel1-2-Cayan_Page_05Mr. Cayan then presented the same graph layout depicting the projections for the latter part of the century, and noted that the results for potential precipitation are even less clear. “I think this is a function of the fact that precipitation is highly variable in this state and it occurs episodically, so we don’t really have great clarity if we’ll get on the average wetter or drier,” he said, noting that these are 30 year averages from the climate models. “So unmistakably according to climate physics, we will be warming in our future. What I take from the precipitation is that we’ll have considerable volatility and we have to be prepared.”

Pages from Panel1-2-Cayan_Page_07California uses the snowpack as a reservoir, so what are the projections for snowpack in the future as the climate warms? Mr. Cayan presented a slide depicting projected changes in the snowpack moving forward through the 21st century. “The blue shading depicts changes that are not great compared to today; as it gets redder, it represents greater and greater change,” he said. “Red means 30% of today’s average or less, so as you can see, the early part of the 21st century, we’re just starting to erode the snowpack around the edges. By middle century, the lower elevations are losing considerable amount of snow – about one-third; at the end of the century, we have approximately half of what we would have had in today’s climate.”

Mr. Cayan noted that this is a relatively conservative estimate that is driven by a climate model on the lower end of climate warming. “Were I to show you a different model, we would see losses that exceed two-thirds of today’s snowpack, so this is going to be a factor that really confronts water management in California and the west,” he said. “It doesn’t mean we’ve lost water; it means we’ve lost water in the form of storage in the snowpack, so that’s something that’s on the horizon.”

Pages from Panel1-2-Cayan_Page_08He then showed a set of projections of soil moisture for the month of June from a hydrological model simulation, noting that the red shading represents drier conditions. “You can see the drying of the western landscape in comparison to a normal that we would have seen over the last 30 to 40 years, and that drying is largely attributed to essentially starting the warm season earlier, exposing the landscape to greater potential for evaporation,” he said.

Pages from Panel1-2-Cayan_Page_11He then presented another slide of model simulations and said, “What we’ve done here is we’ve counted in this set of model simulations the fraction for every year of the decadal incidence of dry years. The sliding bar across the bottom of each of these charts is a running tally of decadal dryness, and what you see is that in the future, we are certainly not saved from dry episodes. We’re expecting from climate models that again we will see prolonged episodes of dryness, much as we’ve seen in our past.”

Pages from Panel1-2-Cayan_Page_10He then presented a graph of water year precipitation over time, and noted that the top bar graph with the black line is the total precipitation in the Bay Delta watershed. “You can see that over time, it wobbles up and down and of course, we’re in a down right now,” he said. “What I want to draw your attention to is the red trace below that and that’s the amount of precipitation that’s contributed by only the upper five percent of precipitation in each year. 95% of the precipitation is contributed by daily events that are the green trace in this chart, which contributes a lot of the annual total. But the variability from one year to the next and from one decade to the next, is strongly driven by these very heavy events, so a small handful events are really the tail that wags our dog in California as far as precipitation.”

So to get out of these dry conditions, we’re really going to need some very large storms,” he said.

Hall 1Alex Hall

Faculty director for UCLA’s Center for Climate Change Solutions

Alex Hall began by saying that his presentation would address whether recent low precipitation levels are a sign of things to come. “The main tools that I’m going to be using to explore this question are global climate models,” he said. “These are basically computer simulations of the climate system that are used to project future conditions.”

There are roughly three dozen or so of these climate models that have been developed around the world, and in this particular study we looked at the output from 34 of the most recent crop of these global climate models,” he said. “We looked at the simulations of the 20th century, and also projections of 21st century climate change and the greenhouse gas emissions scenario that we presumed was a business as usual emissions scenario where we keep increasing the greenhouse gases at the same rate, we have been which is to say quite rapidly.”

Pages from Panel1-3-Hall_Page_03He then presented a chart showing historical precipitation, noting that the black curve is the observed precipitation from 1900 to approximately present day, and on the right hand side are the model projections for precipitation. “These trends are actually rather weak. We don’t see huge changes in precipitation projected by the models. They are also a little bit all over the map. Some of them are positive, some of them are negative, and the picture that you get is actually rather cloudy or fuzzy when you look to the future as far as precipitation changes go.”

The observed trend is actually very weakly positive; it’s not significant in any way,” he said. “The models project on average very little trend with the caveat that some of them do project some moistening and some project some drying. There’s a little bit of uncertainty there, but the picture you get from the type of analysis is that we don’t project huge changes in precipitation coming in on average.”

Pages from Panel1-3-Hall_Page_05Now mean changes in precipitation over the course of a century are one thing, but changes in extremes are quite another, so we wanted to look also at the changes in the extremely wet and the extremely dry years in the models,” he said, presenting a plot and acknowledging that it’s a bit complicated. “Basically, every column on the X axis is a 20 year period of the 21st century, from 2000-2020 on the left, to 2080-2100 on the right, and every row on the Y axis is one of the global climate models. The colors here represent the numbers of extremely wet years; if the box is colored white, that means that the model is projecting the number of extremely wet years that roughly match those in the historical period; the bluer these boxes become, the more the numbers of extremely wet years exceed the expectation that was set by the historical period.”

He noted that the boxes get bluer as you move to the right. “As you move forward in time, we are seeing more and more extremely wet years occurring in these projections, so the models are saying that the numbers of extremely wet years will probably increase in the future.”

Pages from Panel1-3-Hall_Page_06He then presented a similar plot of model projections for extremely dry years and noted that the models don’t see any increase in extremely dry years compared to the historical record until later in the century. “These extremely dry years also tend to increase, but the numbers of extremely dry years don’t exceed natural variability levels until the end of the 21st century, so for the next 50 years or so, we don’t really see any significant change in the numbers of extremely dry years.”

The message here is that by the end of the 21st century, we have roughly 50% more extremely dry years; we have twice as many extremely wet years by the middle of the 21st century, and three times as many extremely wet years by the end of the 21st century,” he said. “Extremely dry and wet years do increase in general in the future, but the extremely wet years seem to be the ones that are more likely to occur going forward in time, but really these signals don’t’ emerge until the middle of the 21st century.”

So what does this all mean for the recent lack of precipitation? “The recent low levels of precipitation are indeed severe but they are not really unprecedented, and really the lack of precipitation in the current drought does appear to be highly consistent with natural variability,” he said. “If you go back to the paleo record, we see even greater instances of extremely dry periods, so it does seem that the recent drought is quite consistent with what we expect from natural variability of this system.”

The models say something similar, Mr. Hall said. “There is no expectation of an increase in extremely dry years really until the middle of the 21st century when we do see a modest increase in the number of dry years, according to some of the models, so both the observational and model evidence really do support the idea that the recent precipitation deficit is attributable to natural climate variability.”

Pages from Panel1-3-Hall_Page_11However, precipitation is not the only factor that leads to drought, he said. “We also have other factors like loss of water to evaporation,” he said, presenting a graph of the Palmer Drought Severity Index, noting that the index is going down and the trend over the past century is significant. “This PDSI doesn’t measure precipitation; it’s meant to be a surrogate for soil moisture conditions so incorporates the effects not only of precipitation but also of evaporation and evaporative losses. Probably the downward trend in the PDSI is due to the fact that the temperatures have gone up over this period of time, and that has enhanced evaporation; the combined influence of the recent deficit precipitation and probably the accelerated evaporative losses have meant that the most recent few years are really unambiguously the most severe in the instrumental record from the perspective of this PDSI.”

So I hope collectively we have conveyed a little bit more nuanced perspective on what drought means,” Mr. Hall said. “It’s not just a precipitation deficit; it’s also related to conditions on the ground. Anthropogenic climate change may make droughts worse through warmer temperatures, and these warmer temperatures will increase evaporation because evaporation is highly sensitive to temperature. And we can get much more severe droughts through warmer temperatures, rather than through any change in precipitation.”

So although recent precipitation levels are severely low, they are not completely unprecedented, said Mr. Hall. “We really don’t see anthropogenic signals increase extremely dry years until the mid century of later, so this current drought that we’re experiencing is within the envelope of what we’ve already experienced. We also see, interestingly, more extremely wet years in the future, so we have to think about that as we’re planning for the future; extremely wet years can be problematic from a flood control perspective.”

Reager 1JT Reager

Research hydrologist for NASA’s Jet Propulsion Laboratory

JT Reager began by giving some facts about groundwater. “Groundwater accounts for 96% of unfrozen fresh water globally, it’s a major source of water for over 2 billion people globally, and it supplies 45% of the water for irrigation globally,” he said. “Groundwater supplies over half of the drinking water and over 60% of water used for irrigation in the U.S. It is the strategic water reserves in times of extreme drought; when there’s no water coming from rain or snowmelt, we turn to groundwater as our main resource.”

Groundwater is any water stored beneath the surface of the earth. “There are two basic types of groundwater: recharging groundwater that is in unconfined aquifers that interact with the earth’s surface, and confined aquifers which are at a deeper level that don’t really interact that much with earth’s surface,” he said. “In California, we have a combination of both of these. We have a lot of fossil aquifers that are here from the last glacial maximum, so we are using a resource that is, by definition, not being recharged and not sustainable.”

Pages from Panel1-4-Reager-1_Page_03Groundwater is hard to monitor, especially in places where there is political or personal resistance to monitoring, so at JPL, we’ve been trying to find a way to monitor groundwater from space using satellites called Gravity Recovery And Climate Experiment or GRACE, Mr. Reager said. “These satellites measure changes in the earth’s gravity fields that are associated with changes in water storage in different places. It sounds complicated but it’s not really; these GRACE satellites measure changes in water storage beneath the path of the satellites in any form. Water storage means any water stored in snow, reservoirs, rivers, lakes, soil moisture, and mostly in groundwater.”

Satellites work well from space but they see things over big scales so the precision of the GRACE satellites is about 1.5 cm and monthly storage anomalies at about 150,000 square-kilometers. “That’s a big scale, that’s about the size of the Central Valley, so we’re seeing how water storage change is happening in that region since about 2002 when the satellite was launched.”

Pages from Panel1-4-Reager-1_Page_04The Central Valley is one of the most important agricultural regions in the world, with more than 250 different crops produced with a value of $17 billion per year in 2002 dollars, Mr. Reager noted. “Eight percent of the food produced in the U.S. by value; one-sixth of the irrigated land in the US and one fifth of the demand for groundwater in the U.S. Groundwater depletion and subsidence have been documented here for decades,” he said, noting the picture of subsidence due to groundwater use dates back to the 70s.

Pages from Panel1-4-Reager-1_Page_05He then presented a slide from a recent cover of Science Magazine that featured an article by Dan Cayan about how water storage is changing. “They call it the drought you can’t see because what we’re really talking about is groundwater changes,” he said. “These are the satellite data that we see superimposed on this map of California, and the most recent drying has been the most severe.”

Pages from Panel1-4-Reager-1_Page_06He then presented a time series of groundwater levels from 2002 with the pictures placed at the corresponding data points. “From 2002 up until the present, the level goes up and down, but generally we just keep getting drier and drier,” he said.

Pages from Panel1-4-Reager-1_Page_07There are different events here, Mr. Reager said. “We see the recent drought, which has been drying for a few years now. There’s a longer period drought, and in fact, the whole GRACE record from 2002 to present is a negative trend, so you could argue that since we’ve been observing this with satellites, California has been drying, and the recent drying is just an acceleration of that.”

Pages from Panel1-4-Reager-1_Page_08Using satellites, they can measure how much water is missing from normal conditions in the Central Valley, he said. “In March, we measured the worst month so far of the drought tor the Sacramento, the San Joaquin and the Tulare Basins,” he said, explaining that the bottom represents the groundwater deficit. “That peak month, we were missing 40 cubic kilometers of water from normal conditions. The volume of Lake Mead is 32 cubic kilometers, so we’ve gone way past using up the entire Lake Mead in the recent drought.

Pages from Panel1-4-Reager-1_Page_11GRACE gives the total water storage of everything below the satellite, and we can separate out the soil moisture, surface water, snow, and groundwater elements, he said. He presented a slide showing groundwater levels plotted against water project allocations, and noted that there’s a bit of a time lag. “There’s about a year lag before we really see the drought in groundwater; we need to go through a whole growing season,” he said. “Also we’re getting GRACE data about 3 to 4 months behind present time, so with the zero allocations that were basically given in January of 2014, we’re expecting that groundwater line to basically go off the charts by spring of 2015.”

Pages from Panel1-4-Reager-1_Page_12He then lastly presented a slide of cumulative groundwater depletion in the Central Valley, noting that the orange line is USGS well data and the green line is GRACE satellite data, available since 2002. “The general pattern here is that there are ups and downs. During wet periods, shown in blue, we see ups. During dry periods, shown in white, we see bigger downs, so the downs are bigger than the ups, and we’re entering a massive dry period, as you know.”

For more information …

Note:  The San Gabriel Valley Water Forum, not surprisingly, had a Southern California focus; some of the presentations are somewhat outside the focus of this blog.  Therefore, only some of the presentations, those with more of a statewide perspective, will be covered.

Daily emailsGet the Notebook blog by email and never miss a post!

Sign up for daily emails and get all the Notebook’s aggregated and original water news content delivered to your email box by 9AM. Breaking news alerts, too. Sign me up!

Print Friendly, PDF & Email