Saturday, January 25, 2014

Answer: How common are droughts in California?

Clearly, my Fridays are too full with other work-that-needs-doing.  I'm still learning how to write the answer AND get it pushed out by Friday.  Insight:  I need to still write it on Thursday morning, just as I've been doing for a while.  Next week I''ll try to get it right! 

Oh, this is a rich research topic, one that I could spend hours and hours gathering data and comparing various resources and insights.  On my desk I have about a dozen papers on this subject.  I could write a book by now... 

But I'm going to resist, and just try to focus on how one would search to answer this question.   

Remember I asked:  
1.  Can you find a record of the rainfall in California for the past two thousand years? And what does that tell us about the relative frequency of drought?    
In other words, if you CAN find some way to track that rain back into deep history, how often do droughts take place?  And how long do they last?  In particular, should we really be worried?   Is this just a temporary thing that will be gone soon?  Or is this a vision of a future to come? 

I like this as a research question because it's typical of the way many research questions are asked--there a lot of specific details.  The phrases "past two thousand years" and "rainfall in California" and "relative frequency of drought" all act to pin down the kind of results we seek.  

One thing to be cautious of:  Don't let the specifics trap you into using exactly that language.  Sometimes that can be over-specific and get you stuck.

How I approached this question:  

My first thought was "how do you even measure rainfall (or lack of it) over two thousand years?"  

And so my first task, and my first query, was to learn a bit about this area.  So I started with a background discovery query: 

     [ measuring rainfall over thousand years ] 

In reading just a few of the results, I learned all kinds of key terms that I could use in my subsequent searches.  For instance, I learned the terms: 

   dendrochronology -- using tree rings to measure rainfall far into the past 
    paleoclimate -- the science of reconstructing prehistoric climates using different methods
    reconstructed precipitation -- specifically, paleoclimate models of rainfall 
    megadrought -- long-term droughts 
    megaflood -- giant flood events 
    Holocene -- the current geologic epoch; started around 10,000 years ago when the glaciers retreated

I learned a few methods for reconstructing climate in the past (such as dendrochronology, using archaeological sites to measure population density, looking at pollen analysis to see what kinds of plants were growing at a particular time, examining stalactites for evidence of variation in rainfall, etc.)  

 I also found a few sites that looked pretty interesting, and just as important, believable.  These tended to be the sites that conducted the original research (usually government, university or academic publications), as opposed to sites that merely compile the data for the purposes of giving it their own (often politically based) interpretation.  

After spending a bit of time immersed in the background literature, learning some of the ideas and picking up the language, I was ready to do my actual search! 

Next step: Searching for the records 

Knowing now that there are trees in California that span several thousand years (both the redwoods, Sequoia species, and the Bristlecone pine, Pinus longaeva), I thought I'd start with dendrochronology as my first attempt to find 2,000 years worth of rainfall data.  

So my first real, topical query was:  

     [ dendrochronology western united states ] 

Note that I didn't use "California" here.  (I could have, but I realized that California is a 19th century arbitrary boundary, and I wanted to cast my net more widely.  I can always narrow later.)  

This was a very productive search.   If you ever want to learn about dendrochronology, let me point you to the University of Tennessee dendro site.  Aside from being a fascinating repository of all things dendrochronological, it also gave me a name of a researcher--Henri D. Grissino-Mayer--that would be useful later.  

Opening articles side-by-side:  When I find a rich set of links like this, I CMD-click to open several pages in new tabs.  (For PCs I think it's a control-click.) This lets me see many results side-by-side in different tabs without losing the original SERP that was so useful.  CMD-click.  CMD-click.  CMD-click.  3 new tabs are open. I quickly look through them to see ones that look plausibly good for more information.  In effect, I'm skimming through a lot of material rapidly.  

Interestingly, one of the links was a newspaper article in the LA Times.  Reading that gave me another name and a clue.  It mentioned "..paleoclimatologist Connie Woodhouse, a co-author of the study that will be published in the journal Geophysical Research Letters."  

Keeping notes:  What I'm doing is learning as I go--picking up (and often writing-down) the names of authors, researchers, titles of papers, and labs where good work is going on.  This often proves useful when I'm trying to expand my search.

If this was going to be a long piece of research, one that would last for weeks or months, I'd start keeping track of this information in a file (usually an online file) where I'd drop links, all those terms-of-art, and names.  I usually keep this notefile online where I can find it later if I ever need to return to a topic.  (As a Google Doc it's handily re-discoverable with a simple search.)  

In this particular case, I just followed-up on this clue by doing a search for: 

     [ Connie Woodhouse Geophysical Research Letters drought ] 

which quickly took me to her article, "Two Thousand Years of Drought Variability in the Central United States."  ( Bulletin of American Meteorological Society  79.12 (1998): 2.)  Fascinating article, but not quite specific enough to California, but with the insight that "multidecade droughts [ of the past ] were of much greater duration and severity than twentieth century droughts..."  Of course, I check on the author--who is she?  A quick search shows she's a professor at the University of Arizona Geosciences department with joint appointments in the Laboratory of Tree-Ring Research and Geosciences.  (As a side-note, I always do this kind of background check on authors.  It's useful to know where they come from and what their perspective is.)  

Uh oh.  This is getting interesting, and slightly scary. Here's another reference by the same author that I picked up from Google Scholar.  
Woodhouse, Connie A., et al. "A 1,200-year perspective of 21st century drought in southwestern North America." Proceedings of the National Academy of Sciences 107.50 (2010): 21283-21288.
All of these papers are coming from incredibly reputable sources.  The National Academy of Sciences is an incredibly prestigious non-profit organization that serves pro bono as "advisers to the nation on science, engineering, and medicine."  They regularly issue very authoritative reports on a variety of topics, including paleoclimate. 

Broadening your research by scanning:  But as I scanned through this paper I found it was about the southwest US in general.  However, I spotted references to analyses done on California.  When I Control-F through the paper looking for "California," I find this reference:  
Hughes, M. K., and P. M. Brown.  Drought frequency in central California since 101 B.C. recorded in giant sequoia tree rings. Climate Dynamics, 6, 161–167 (1992)
I also find an article by Grissino-Mayer (another name I recognize!) on a reconstruction of a 2,129-year dendrochronology for New Mexico.  I'm getting close!  So these documents are out there, just waiting for me to find them.  

Getting to the Hughes article is a little tricky.  I was able to read the first couple of pages, and it indeed does seem to be exactly what we're looking for, but it's hidden behind a paywall--I can't get to it without paying $37.  (But I'll write a post on ways to work around this early next week.)  

So I kept looking, and found another article by Scott Stine, this one is published in Nature (the very well-respected British science journal), AND it's available as a PDF.  In this case, I looked up the article in Scholar, and the link to the PDF handily appears on the right.  (Pay attention to these PDF links in Scholar--they're often your best friend in gaining access to these articles.)  
Stine, Scott. "Extreme and persistent drought in California and Patagonia during mediaeval time." Nature 369.6481 (1994): 546-549.

In this article he tell us that two major droughts happened in 892 - 1112 (220 years) and 1209 - 1350 (141 years).  This was at the same time as the "Medieval Climatic Disruption" that was affecting climate around the world.  It was a dry few centuries in California.  

But by now I'm convinced that specific dendrochronologies exist for California too (and not JUST the "Western United States"), so I change my search to:  

     [ dendrochronology California drought  paleoclimate ] 

From this search, I quickly find a few more examples: 
Hughes, M. K. and L. J. Graumlich. 1996. Climatic variations and forcing mechanisms of the last 2000 years. Volume 141. Multi-millenial dendroclimatic studies from the western United States. NATO ASI Series, pp. 109-124

When I repeat this same search in Google Books, I find this fascinating text.   
The West Without Water: What Past Floods, Droughts, and Other Climatic Clues Tell Us about Tomorrow. University of California Pr, 2013.  Ingram, B. Lynn, and Frances Malamud-Roam. 

Focusing your search by following an author:  As I read through much of this book (in Google Books), it seemed to be incredibly close to what I was seeking, so I thought I'd follow up on the work of some of the authors.  I started with Lynn Ingram (at UC Berkeley Geography Department), and in the process, I discovered a gold mine of her work.  

She describes her research as answering the following questions:  
(1) What was the natural variability of annual average inflow into the San Francisco Bay estuarine system during the Holocene and the last interglacial? (2) What was the amplitude and seasonal distribution of inflow during this period? (3) How did changes in inflow and sea level rise affect the ecosystems of the northern reach of the estuary? 4) What are the implications for future climate change and sea level rise for the San Francisco Bay system and other urban estuaries? (5) How has upwelling along California coast varied over the past several thousand years?

This is a great lead.  Her research questions are exactly what I was looking for.  

I went to Google Scholar (knowing that she's probably going to publish in academic journals) and ran a search just on her name.  

     [ Lynn B. Ingram

to find papers she's written on the topic.  Bingo!  

In her 2006 paper in Quaternary Science Reviews we find this diagram:    

This is one of the key diagrams from: "Holocene paleoclimate records from a large California estuarine system and its watershed region: linking watershed climate and bay conditions"  Quaternary Science Reviews 25 (2006) 1570–1598).  
Frances P. Malamud-Roama, B. Lynn Ingram, Malcolm Hughes, Joan L. Florsheim.

Excerpt from this paper's summary: 

"Climate over California during the last century has been relatively stable when examined in the context of the past 2000 years, even with the severe droughts of the, 1930s and the mid-1970s. ... Longer term records, including tree rings, and sediments from lakes and from in and around the San Francisco Bay, provide a means of extending records to cover paleoclimate for much of the Holocene. When these longer paleoclimate records are considered, current drought conditions experienced in the US Southwest do not appear out of the range of natural variability."

Another paper by these same authors: 

Malamud-Roam, F. et al. "Holocene climates and connections between the San Francisco Bay estuary and its watershed: a review." San Francisco Estuary and Watershed Science 5.1 (2007).

pushes back their analyses of rainfall, drought, and other climate variables back to 13,000 years ago.  To get that far back in time they use a variety of data sources (tree rings only get you 5000 years back), but the story is pretty consistently the same:  droughts happen fairly frequently; sometimes they're very, very long.  

"If you go back thousands of years, you see that droughts can go on for years if not decades, and there were some dry periods that lasted over a century, like during the Medieval period and the middle Holocene. The 20th century was unusually mild here, in the sense that the droughts weren’t as severe as in the past. It was a wetter century, and a lot of our development [ in California and the West ] has been based on that."

Search lessons:  There are many lessons to draw from this example.  

1.  What's authoritative?  We looked at a large number of articles, trying to stick to well-known sites that I considered authoritative.  How do you know what's reliable?  Good question.  The answer is tricky, but it's basically a reputation question that comes from your belief system.  For instance, in searching for articles on this topic, I stayed away from "climate change skeptic" sites.  (There are a lot out there.)  Their data could be fine, particularly about paleoclimate, but I've found that they tend to cherrypick the data they present.  I tend to trust articles (and authors) where the data isn't quite so black-and-white.  Deciphering the past is a tricky business.  In many of the papers I cite above, you'll see contradictory evidence ("this site indicates these years were in drought, but this other site seems to be well-watered...").  That lets me know they're not out to sell you something, but are genuinely trying to present data for evaluation.  Ambiguity isn't always a problem, sometimes it means the answer just isn't that simple.  

2.  Learn the terminology.  As you see in my very first step, often the best first thing to do is to look around a bit and learn what the key terms are.  This is always the first thing I do when starting a research in a new (to me) area.  What special words and phrases do I need to know?  Who are the big players in the field?  What institutions participate in this kind of work?  And very importantly:  What common terms do I need to learn have special meanings in this area?  It's just as important to learn that some ordinary words (such as "drought") have very specific meanings in this area.  To avoid misunderstandings, be sure to notice how writers use everyday words in special, sometimes subtle ways.  

3.  Follow an author. As I mentioned, as I research, I keep notes.  I take names.  I write down institutions, places, and temporary results. I write down fragments of things found along the way that might help to find the next bit.  Keeping notes of authors is really useful.  Then, once you find an author that has an interest along the lines of your research, check out their publications.  This is especially true for academic writers, but this also works well for news reporters, essayists, and book authors as well.  

4. Use Scholar and Books for their intended purposes.  I can't tell you how often I would find a paper, then zip over to Scholar to find the citation of that paper AND find a PDF.  Same thing with Books--I'd find a book reference, then use Google Books to (usually) search inside of the book for more information.  Both of these are special purpose tools, well worth understanding for what they are.  

Thanks to all of the SearchResearch commenters!   You make reading the blog a real search joy! 

Search on! 


  1. Thanks to you, Dr. Russell. It is excellent and easy how we can learn with the challenges, the comments and the answers.

    I enjoyed this challenge and the answer very much. Specially, the process to get it.

    With your answer, I learned new things and new ways to search. Honestly I didn't think to search many of the things you did. For example, [ measuring rainfall over thousand years ]

    For me, reading the post that my peers make it is also great; because, they give another perspective and also clues to do new searches. Thanks to all of you!

    In this challenge, I learned more stuff thanks to Fred, who shared with me. Thanks Fred!

    There is much to learn and try! Your answer Dr. Russell has so much parts and knowledge that I need to "eat" it by little.

  2. My take away from this challenge is to build the foundation first. Be in a gathering knowledge mode as opposed to a quick solution. Validate your sources. Compare your sources. Follow your sources. Then begin the evaluation process.

    This challenge and Dr. Dan's answer has given me more insight than any other challenge. To Fred and Ramon- you kept reaching for more information not being satisfied with what you had found. I was really impressed by your efforts.

  3. Remmij I may not always understand some links you provide but your contribution is always interesting. You really do think outside the box. I would really enjoy knowing more of your thoughts along with the links. You added to this challenge through out by providing many links that I would not have considered. I feel like we all enjoy each others contributions.

  4. In my previous post; I forgot to say that I believed the chart that answers this challenge would be the one that comes in site that Rosemary, Fred and others found too. And, no! There is more than one chart with information about this topic.

  5. Dan Russell has got another fan. My brother is a senior investigator in meteorology and his laconic comment was "Excelente !!!". Yes, three exclamation points, he went wild.