Precipitation Changes in The Great Plains Region
Ally Barry
Author
08/02/2021
Added
24
Plays
Description
As impacts of climate change continue, the Great Plains Region's agricultural industry is at risk. To best determine means of adaptation, it's important to have an understanding of how our climate might change. The research project focuses on future mean temperature and precipitation changes in the Great Plains using CMIP data to generate our climate models.
Searchable Transcript
Toggle between list and paragraph view.
- [00:00:00.000]Hello everybody! My name is Ally Barry,
- [00:00:02.130]and I'm a sophomore double majoring in
- [00:00:04.160]environmental studies and political
- [00:00:05.690]science from right here in Lincoln, NE.
- [00:00:07.624]This summer, I was afforded the
- [00:00:09.234]opportunity to gain experience with
- [00:00:11.529]climate projection models under the
- [00:00:13.679]guidance of assistant professor Ross Dixon
- [00:00:16.063]in the Depertment of Earth and Atmospheric
- [00:00:18.071]Sciences. Throughout this summer, I
- [00:00:19.539]encountered many learning curves, crashed
- [00:00:21.899]a lot of code, but, throughout it all, I
- [00:00:22.821]gained a further appreciation for the work
- [00:00:25.603]that modelers do to assist in the
- [00:00:27.315]preparation for impacts of climate change.
- [00:00:29.979]Whether it be a shift in your local food
- [00:00:31.819]supply or the compounding heat of summer,
- [00:00:33.981]we all will experience impacts of climate
- [00:00:35.522]change. While rapid change might feel
- [00:00:37.847]distant from ourselves, it matters a lot
- [00:00:40.494]to the Great Plains Region as we face
- [00:00:42.359]impending food insecurity. In their fourth
- [00:00:44.819]report, the National Climate Assessment
- [00:00:46.627]states, "projected changes in
- [00:00:48.515]precipitation, coupled with rising extreme
- [00:00:50.778]temperatures before mid-century, will
- [00:00:53.176]reduce Midwest agricultural productivity."
- [00:00:55.320]We are a massive player in agriculture,
- [00:00:57.744]however, our projected climate is a
- [00:00:59.961]serious threat.
- [00:01:02.452]Current questions for the Great Plains
- [00:01:04.122]point primarily towards methods of
- [00:01:06.018]adaptation. If climate conditions are
- [00:01:07.929]going to change, how will agriculture need
- [00:01:09.898]to change with it?
- [00:01:11.153]If we want to know what the future might
- [00:01:13.493]look like, we need a method to predict it.
- [00:01:15.101]Globally, modelers have compiled the
- [00:01:17.350]Coupled Model Intercomparison Project
- [00:01:19.064](also known as CMIP) which is an open
- [00:01:21.474]source, comprehensive collection of model
- [00:01:23.817]output which reflects historical, present,
- [00:01:25.790]and future climates. In the lower left
- [00:01:28.796]side of the screen, you'll see a map of
- [00:01:30.614]the participating nodes which each
- [00:01:32.955]generate and release data. Boulder, CO
- [00:01:35.359]hosts the nearest node producing the
- [00:01:37.380]Community Earth Systems Model version 2
- [00:01:39.129](CESM2) which can be used to generate
- [00:01:42.346]global scale models and provide boundary
- [00:01:44.878]conditions for regional scale models, like
- [00:01:46.892]the Weather Research and Forecasting Model
- [00:01:49.079](WRF). These two different climate model
- [00:01:52.068]types, regional and global, are both
- [00:01:54.429]useful for addressing different aspects of
- [00:01:56.662]climate prediction. While global scale
- [00:01:59.071]models are beneficial for understanding
- [00:02:01.419]wide scale change, they are generally run
- [00:02:03.754]at a lower resolution (for example 100-200
- [00:02:07.640]km), making it difficult to use them for
- [00:02:09.848]application such as forcing agricultural
- [00:02:12.322]and hydrological models. Regional models
- [00:02:15.553]have much higher resolutions (often
- [00:02:17.864]between 4-25 km), but they must be forced
- [00:02:22.092]by boundary conditions (temperature,
- [00:02:24.340]momentum, and moisture).
- [00:02:25.891]Before running our simulations, we
- [00:02:27.879]hypothesized that precipitation changes
- [00:02:30.094]across the Great Plains Region would
- [00:02:32.254]experience a North-South dipole based on
- [00:02:34.551]previous analysis of CMIP6 models as shown
- [00:02:37.524]in the top right figure from a study by
- [00:02:39.219]Almazroui et al. in 2020 where green
- [00:02:42.821]indicates precipitation increase, tan
- [00:02:45.762]indicates decrease, and the hatchings
- [00:02:48.220]indicate model agreement. The lack of
- [00:02:50.827]model agreement in presents a unique
- [00:02:52.984]challenge in the Southern Great Plains- a
- [00:02:54.891]challenge regional models could assist
- [00:02:56.519]with. Our objective was to narrow down
- [00:02:58.230]this domain to only the Great Plains and
- [00:03:00.434]generate a higher resolution, regional
- [00:03:02.508]simulation of future temperature and
- [00:03:04.763]precipitation change under a strong carbon
- [00:03:07.506]forcing scenario.
- [00:03:08.742]To generate these simulations, we first
- [00:03:10.792]started by selecting a domain to encompass
- [00:03:13.100]the entire Great Plains region with extra
- [00:03:15.543]space available on the edges to account
- [00:03:18.073]for any edge effect that might produce
- [00:03:19.562]inaccurate data. Thus, our simulations
- [00:03:21.691]were run at 20km horizontal resolution, 5x
- [00:03:25.721]higher than that of CESM2. Regional models
- [00:03:28.821]do require data for the boundary forcings
- [00:03:31.634]of temperature, momentum, and moisture
- [00:03:33.557]which we derived from CESM2 to force our
- [00:03:36.955]regional model, WRF. When setting up our
- [00:03:39.967]model runs, we selected a historical and
- [00:03:42.945]future time period of 2000-2009 and
- [00:03:45.395]2085-2094 respectively. The historical
- [00:03:49.586]period allows us to have a control period
- [00:03:51.636]in which we can explore the accuracy and
- [00:03:53.779]biases of the model. While the future
- [00:03:56.012]period is what we are most interested in
- [00:03:58.574]as it projects mean changes for the region
- [00:04:00.808]which will help inform future decisions in
- [00:04:03.138]agriculture that might be necessary to
- [00:04:04.923]adapt to climate change. To process our
- [00:04:07.519]data, we used high performance computing
- [00:04:09.208]systems from the Holland Computing Center.
- [00:04:12.144]Although we produced a massive amount of
- [00:04:14.406]outputs, our results focus on mean
- [00:04:16.590]temperature and precipitation. While we
- [00:04:19.160]initially wanted to consider the entire
- [00:04:21.184]summer, computing time was limited so we
- [00:04:23.094]displayed only the month of June after
- [00:04:24.981]discarding the month of May as a spin up
- [00:04:26.907]period.
- [00:04:27.519]Our first result shows the historical
- [00:04:29.605]period and future projection of
- [00:04:31.544]temperature for the Great Plains produced
- [00:04:33.117]by the global CESM2 for the month of June
- [00:04:35.639]with the colorbar measuring in degrees
- [00:04:37.781]Celsius. On the left is the historical
- [00:04:40.342]period of 2000-2009 which shows a gradient
- [00:04:43.511]of higher temperatures in the south to
- [00:04:46.238]colder temperatures in the north with
- [00:04:48.414]exception to the Rocky Mountains located
- [00:04:49.642]on western portion of the domain. On the
- [00:04:52.091]right is displayed the mean temperature
- [00:04:53.808]change from 2085-2094 which projects
- [00:04:56.405]widespread increase in temperatures across
- [00:04:58.930]the full region with the largest change
- [00:05:01.584]in the norther Great Plains- an expected
- [00:05:04.035]result of polar amplification.
- [00:05:06.004]Using the global models as a reference,
- [00:05:08.758]we can then compare our regional and
- [00:05:11.233]global results. Generally, the WRF
- [00:05:13.017]simulation in the panel of the right-hand
- [00:05:15.444]side is clear and statistically
- [00:05:17.764]significant across the entire region: the
- [00:05:19.833]Great Plains will experience warming.
- [00:05:21.629]However, the historical temperatures do
- [00:05:24.497]extend warming further into the central
- [00:05:27.792]Great Plains as well as exhibit a greater
- [00:05:30.267]increase in mean temperature for our
- [00:05:32.674]future period. This bias is small, but
- [00:05:35.101]could be a result of many factors like
- [00:05:37.525]the chosen domain, the forcing from the
- [00:05:39.984]global model, or specific physical
- [00:05:41.251]parameters that would be further explored
- [00:05:43.913]in future research.
- [00:05:46.158]The next set of results displays
- [00:05:47.933]precipitation starting with our global
- [00:05:49.741]CESM2 still only for the month of June.
- [00:05:51.770]Precipitation rates are shown using our
- [00:05:54.167]colorbar where blue represents less
- [00:05:55.731]precipitation and red indicates higher
- [00:05:57.903]levels of precipitation in mm/day. In
- [00:06:02.051]these historical simulations, the CESM2
- [00:06:04.321]produces greater precipitation towards the
- [00:06:06.491]east as you move away from the Rocky
- [00:06:08.932]Mountains. This is a result of the rain
- [00:06:11.352]shadow effect.
- [00:06:12.976]The future change in precipitation across
- [00:06:15.209]the Great Plains (shown on the right-hand
- [00:06:17.690]side) displays a dipole pattern in which
- [00:06:20.549]precipitation increases in the south and
- [00:06:22.818]the northern portion experiences a net
- [00:06:25.075]loss in precipitation.
- [00:06:26.264]Moving to our regional WRF simulation, we
- [00:06:29.686]can still observe the rain shadow effect
- [00:06:31.788]with greater levels of precipitation in
- [00:06:33.895]the east.
- [00:06:35.276]Both the global and regional simulations
- [00:06:36.955]share a general pattern of historical
- [00:06:38.659]precipitation, however, the regional model
- [00:06:41.513]suggests much drier conditions than that
- [00:06:43.626]of the global scale.
- [00:06:44.999]Moving to the future projections, we see a
- [00:06:47.630]similar general increase in precipitation,
- [00:06:49.823]however, the north-south dipole does not
- [00:06:53.318]carry the strong presence we expected.
- [00:06:55.375]Additionally, statistical significance in
- [00:06:57.879]the future projection is lacking in both
- [00:06:59.968]the CESM2 and our our regional runs.
- [00:07:02.215]As mentioned with temperature, this could
- [00:07:05.165]be improved by reevaluating our physical
- [00:07:07.065]parameterizations, domain, and months ran.
- [00:07:10.210]In conclusion, we developed high
- [00:07:12.280]resolution, regional simulations of
- [00:07:14.530]historical and future projects of
- [00:07:16.313]temperature and precipitation in the Great
- [00:07:18.216]Plains Region. The output from our
- [00:07:20.012]simulations can be altered and expanded
- [00:07:21.826]upon beyond our study to explore longer
- [00:07:24.893]time periods and more fields that would
- [00:07:26.977]assist in studying heavy precipitation
- [00:07:29.067]events, the diurnal cycle, or heat waves
- [00:07:31.237]in the Great Plains. Since we did not
- [00:07:33.277]achieve much statistical significance, it
- [00:07:35.497]would be beneficial to expand our time
- [00:07:37.327]period to include June, July, and August-
- [00:07:39.156]still using May as a spin up. Additionally
- [00:07:42.135]we could manipulate the model parameters
- [00:07:44.669]and domain.
- [00:07:46.279]However, we still assert the importance
- [00:07:48.179]and benefit of high resolution, regional
- [00:07:49.559]models for the Great Plains Region as they
- [00:07:52.004]provide information to the agricultural
- [00:07:54.059]community that global models could not.
- [00:07:55.709]I'd like to thank Ross Dixon for serving
- [00:07:58.219]as my mentor and advisor on this project,
- [00:08:00.710]the Holland Computing Center for providing
- [00:08:03.230]computing resources, and the UCARE program
- [00:08:05.862]for funding this experience.
- [00:08:07.376]Thank you so much!
The screen size you are trying to search captions on is too small!
You can always jump over to MediaHub and check it out there.
Log in to post comments
Embed
Copy the following code into your page
HTML
<div style="padding-top: 56.25%; overflow: hidden; position:relative; -webkit-box-flex: 1; flex-grow: 1;">
<iframe
style="bottom: 0; left: 0; position: absolute; right: 0; top: 0; border: 0; height: 100%; width: 100%;"
src="https://mediahub.unl.edu/media/17563?format=iframe&autoplay=0"
title="Video Player: Precipitation Changes in The Great Plains Region"
allowfullscreen
></iframe>
</div>
Comments
0 Comments