Intro

Washington is famous for being rainy with parts of the state getting more than 200 days of rain a year. This is something that may only increase as climate change is increasing precipitation in some areas. Studying how this is vital to better understand Washington’s future. Understanding how the precipitation has already changed in Washington based on historic precipitation data is vital to not only understand how much wetter it’s getting but also where is it getting more rain. This is vital for informing how the region will use water as Washington is a state with some of the deadliest landslides in history something that can be caused by days of heavy rain.

Understanding where this excess rain is falling will be vital to protect communities in the state. Some of the critical questions I will look it answer are how much and where is it getting wetter in Washington. And are there areas of Washington trending towards having frequent droughts? Can the areas with the biggest increases in rainfall be correlated with years with the most landslides in areas?

This research will be vital in understanding both climate change as a whole as well as the risks involved in not doing something to stop it. Therefore I think this project is vital to everyone and can serve as a tangible example of what climate change is doing. I hope that showing direct evidence of how the state is changing can also help inform state planning, as if we know rainfall is increasing in an area we can better plan for possible landslides.

Background

Rainfall is a crucial part of the water cycle and determines much of our lives from what crops we can grow to where we build cities. Understanding rainfall is crucial for society to be able to properly plan and have our society function. “Generally precipitation is expected to increase as the world gets warmer” (Fountain and Kao, 2021), this is the consensus among climate scientists and is what we can expect for our future. One portion of understanding rainfall is understanding the water cycle and where the moisture comes from (figure 1).

Figure 1. Water cycle diagram. Blue arrows represent fluxes between reservoirs. (usgs.gov).

This cycle is being changed by climate change as radiative forcing adds more energy into the system causing more evaporation and more uncertainty. “A world that is around 4C warmer than the pre-industrial era would have around 28% more water vapour in the atmosphere” (Hausfather, 2018), this extra moisture means that the water cycle itself is changing. One way we can understand more about the water cycle is to look to the past and see how climate change has already begun to change the water cycle. In order to understand this more we must understand how much has already changed. “Overall, global land precipitation has increased by about 2% since the beginning of the 20th century” (Dore, 2005). This may not seem like a lot but 2% on a global level is a huge change. Changes like this can have lots of effects throughout ecosystems, for example, soil “bacterial community composition and function have been shown to change with experimental manipulation of precipitation timing” (Evans et Wallenstein, 2014). These changes can have huge impacts on the ecosystem as a whole as changes in the soil bacteria can drastically affect the plants trying to live there and then radiate up the trophic levels. Changes in soil bacteria could also affect what crops are able to be grown meaning impacts on society’s food supply, something that can have terrible outcomes. Understanding these impacts not only on a global level but also on a smaller scale is vital in order to adequately fight climate change.

Regionally, “under continued global warming, more intense moisture transport within atmospheric river events is projected to increase the magnitude of heavy precipitation events on the west coast of the USA” (IPCC WG 1 Ch. 8, 2021). This means that the west coast is expected to get more rain and a larger portion of it in extreme rainfall events such as atmospheric rivers. One way we can look to see how the rainfall is changing and how it will change is by using complex climate modeling software this allows us to hopefully get an idea of what the future will look like. Models make these predictions by doing complex calculations on how multiple variables such as wind moisture temperature and more change around the environment, these are run on supercomputers that can take weeks at a time to get results. Models have been run to predict future changes in rainfall in Washington and found that rainfall is predicted to increase in some areas and decrease in other parts of the state (figure 2).

Figure 2. Change in precipitation (millimeters/month) from 1970–1999 to 2030–2059. CCSM3–WRF (top row) and ECHAM5–WRF (bottom row) are two different styles of modeling for the four seasons (Salathé Jr et al., 2010).

These results are interesting because they show an increase in precipitation in most of the state with half of the predictions showing a decrease in rain over the cascade range. This makes some sense with one of the most assured predictions of climate change being that the climate will become more chaotic. In these maps, you can also see that the cascade range in the middle of the state is one of the areas with the biggest change this is something that aligned with what I found however in some cases into a different direction. Other models have found some similar results, with the IPCC finding that climate change “...is projected to increase the intensity of heavy precipitation events on the west coast of the USA…(high confidence)” (IPCC WG1 Ch. 8, 2021). This is something that makes sense given that more heat in the climate means more evaporation, especially as the oceans warm. This increased rainfall can have disastrous effects on things like flooding and landslides. Washington has had a history of landslides in the past with some being extremely deadly. Previous research has found that “rainfall has a direct effect on occurrence of landslides” (Finlay et al., 1997). Understanding where the rain is changing can help us understand where more monitoring can be done for landslides. Doing this analysis also can be used to show where there are higher risks for droughts in the future allowing for infrastructure to be built to alleviate future droughts.

Methods

I began my analysis by downloading data by hand from NOAA however I quickly learned that this was going to be too slow to do the kind of analysis I wanted to do. I ended up using R and the FedData package in order to download weather station data easier. I used the get_ghcn_daily() function in order to download all the stations and all of their history in Washington. Once this was done I cleaned up the data and got it into an easier format to work with. I then began by dividing the data into the four seasons based on December January and February being winter, March April May being spring, June, July, and August being summer, and September, October, and November being fall. Once the data had been divided I then averaged the data in a station to show the seasonal total averaged over a 10-year period (figure 3). This allowed me to show change over time but average out any anomalous years. This showed me that I had enough data to go back to 1900 but not any earlier than that as there were only a few stations active at that time. 

Figure 3. Plot of average yearly winter rainfall (in cm) for 2010-2019. Color of dot represents avg yearly total, stations with less than 7 complete years were removed.

Once I had the data divided by season and decade I then began the process of Kriging the data so that I could interpolate it between the stations. I chose to do this as “weather monitoring stations tend to be closer together in the eastern and central states than in the western states” (EPA, 2021). Kriging the data thus allowed me to better see trends for the space in between the weather stations. This was done by creating a variogram and then using that to interpolate the data this allowed me to have a more complete data set of Washington and not just specific stations (figure 4).

Figure 4. Example of kriged plot for winter 2010-19. Black dots represents the stations that I used to make the interpolated map.

Results

Figure 5. Avg total rainfall during December, January, and February in Washington since 1900.

Figure 6. Avg total rainfall during March, April, and May in Washington since 1900.

Figure 7. Avg total rainfall during June, July, and August in Washington since 1900. Different color scale than other animated maps due to less rainfall during the summer months.

Figure 8. Avg total rainfall during September, October, and November in Washington since 1900.

Changes In Average Winter Rainfall 1900-1929 to 1990-2019

Figure 9. Changes in Avg Winter Rainfall from 1900-1929 to 1990-2019. Red represents a decrease and blue and increase. Dots represent landslides caused by some form of precipitation.

The maps show that rain is increasing throughout the state, especially along the cascade range (figure 5 and 9). The changes in winter rain showed increases along the Cascade Range and decrease along the central coast and around the Columbia river valley near Vancouver, WA (figure 5). Similarly, for spring there was also an increase in rain in the Cascades, especially over the last 40 years (figure 6). Fall also showed this trend with the largest increases coming in the last ~30 years (figure 7). Summer rainfall seems to have a less obvious trend but still showed an increase in the Cascade Range and slight decreases in other parts of the state (figure 8). Landslides did not seem to have a strong correlation with the areas of increased rainfall I suspect this may have to do with where landslides are monitored the most as they were clustered around Seattle the most (figure 9).

Discussion

These results were very intriguing because they showed some very distinct differences between what the models from Salathé Jr et al. 2010 predict and what my analysis shows. Specifically, along the Cascade range, I found that there will be an increase in precipitation, especially in the winter while the climate models predicted there would be a decrease (figure 2 and 5). There could be a couple of reasons for this, one being that there is some tipping point that will send the trend in the other direction and lead to drier cascades that just hasn’t happened yet. This could also be due to a difference in the data used and the resolution of that data and the starting point for their model. I believe that my analysis is accurate and the trend of the cascades receiving more rain does make logical sense. If more moisture is being blown inland from the ocean due to increased radiative forcing, it would make sense that as this extra moisture moves up and over the Cascades it will rain out and cause an increase in the rainfall in that area. This discrepancy is extremely important and illuminates a conflict between historical trends and predictive models something that should be resolved so we have a more solid basis for what the future will look like with climate change in Washington. A “study found that the average climate model likely underestimates how extreme precipitation will change in response to global warming” (Teirstein, 2022). This could be another explanation as to why my results differ from the trends predicted in some models.

These results help show how much climate change is already causing changes in our lives, something that can be hard for people that don’t study climate change to realize. Showing how things are already changing can help create a sense of urgency and move people to act more drastically and swiftly to combat climate change, giving us a better chance for the future. These results can also be used to help show where money needs to be spent in the fight against climate change as where things are changing the most is a good place to start. More analysis needs to be done to understand the societal implications of these patterns and what can be done to help alleviate these changes. This is the future of analysis of this kind is understanding how and where things are changing so that we can better serve these areas and hopefully create a more just and sustainable society.

Work Cited

Climate Change Indicators: Heavy Precipitation. (2021, April 1). EPA. https://www.epa.gov/climate-indicators/climate-change-indicators-heavy-precipitation

Dore, M. H. (2005). Climate change and changes in global precipitation patterns: what do we know?. Environment International, 31(8), 1167-1181.

Douville, H., K. Raghavan, J. Renwick, R.P. Allan, P.A. Arias, M. Barlow, R. Cerezo-Mota, A. Cherchi, T.Y. Gan, J. Gergis, D. Jiang, A. Khan, W. Pokam Mba, D. Rosenfeld, J. Tierney, and O. Zolina, 2021: Water Cycle Changes. In Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelekçi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 1055–1210, doi:[10.1017/9781009157896.010](doi:%5B10.1017/9781009157896.010){.uri}.

Evans, S. E., & Wallenstein, M. D. (2014). Climate change alters ecological strategies of soil bacteria. Ecology Letters, 17(2), 155-164.

Finlay, P. J., Fell, R., & Maguire, P. K. (1997). The relationship between the probability of landslide occurrence and rainfall. Canadian Geotechnical Journal, 34(6), 811-824.

Fountain, H., & Kao, J. (2021, May 12). There’s a New Definition of “Normal” for Weather. New York Times. https://www.nytimes.com/interactive/2021/05/12/climate/climate-change-weather-noaa.html

Hausfather, Z. (2018, January 19). Explainer: What climate models tell us about future rainfall. Carbon Brief. https://www.carbonbrief.org/explainer-what-climate-models-tell-us-about-future-rainfall/

Perlman, H., & Evans, J. (2019, October 16). The natural water cycle (JPG). The Natural Water Cycle (JPG) | U.S. Geological Survey. Retrieved December 5, 2022, from https://www.usgs.gov/media/images/natural-water-cycle-jpg

Teirstein, Z. (2022, July 20). Extreme rainfall will be worse and more frequent than we thought, according to new studies. Grist. https://grist.org/extreme-weather/extreme-rainfall-will-be-worse-and-more-frequent-than-we-thought-according-to-new-studies/
Salathé, E. P., Leung, L. R., Qian, Y., & Zhang, Y. (2010). Regional climate model projections for the State of Washington. Climatic Change, 102(1), 51-75.

R Code

Code for this analysis can be found here