1. Jennifer Adam, Assistant Professor Civil and Environmental Engineering Soil Seminar Series January 31, 2011 E ARTH S YSTEM M ODELING : A F RAMEWORK FOR I NTERDISCIPLINARY R ESEARCH

2. O UTLINE  What is an “Earth System Model”?  A new Earth System Model: BioEarth  How can BioEarth serve as a framework for interdisciplinary research?

3. W HAT IS AN E ARTH S YSTEM M ODEL ?

4. W HAT IS NOT AN E ARTH S YSTEM MODEL ? A NSWER : “S TAND -A LONE ” M ODELS  Represents a single or limited number of components of the earth system  Can be very sophisticated (and computationally demanding!) for this single component  Example: A land surface hydrology model

5. W HAT IS AN E ARTH S YSTEM MODEL ?  Represents multiple components of the earth system  Example: Community Earth System Model (CESM)  Developed at National Center for Atmospheric Research (NCAR) and greater community  Global scale Coupler Atmosphere Ocean Land Land & Sea Ice

6. W HEN TO USE S TAND -A LONE VERSUS E ARTH S YSTEM M ODELS  Stand-Alone Models:  Only needing to understand 1-way impacts  When “forcing” the model with observed data is better than “forcing” the model with outputs from another model (historical simulations)  Earth System Models:  To explore 2-way effects or feedbacks between components of the earth system  To explore projected changes in a system when wanting to capture feedback effects, e.g., due to climate change (future simulations)

7. D ECIDING ON C OMPONENTS OF THE E ARTH S YSTEM M ODEL  Domain of interest  geographic location and extent  scale  Question of interest  Current understanding of most important processes

8. A NEW E ARTH S YSTEM M ODEL

9. B IO E ARTH VERSUS CESM BioEarthCESM Regional scale: Pacific Northwest (uses a GCM for boundary conditions) Global Scale (no boundary problems) Finer Spatial ResolutionCoarser Spatial Resolution No Ocean, sea ice, or glaciersHas ocean, sea ice, glaciers Some components more sophisticatedSome components necessarily less sophisticated *More explicit handling of human activities: cropping systems (irrigation, tillage, fertilization), forest management, reservoir operations *Integrated economic modeling; can be used for policy scenario investigation *Model outputs more directly relevant for local scale planning/decision making/informing policy Less explicit handling of human activities

10. M ODELING D OMAIN : T HE P ACIFIC N ORTHWEST

11. I NCEPTION OF B IO E ARTH  A Vision  Envisioned by faculty at WSU Laboratory for Atmospheric Research (LAR)  Projects used for leveraging  AIRPACT: near-term air quality forecasting for Pacific Northwest  WA Ecology project on Columbia River Basin long- term water demand forecasting  NSF IGERT: Nitrogen Systems: Policy-oriented Integrated Research and Education (NSPIRE)

12. T HE T EAM  6 Institutions with WSU as Lead  18 Faculty  3 Postdoctoral Scholars  10 PhD Students  5 Working Groups

13. C ONTRIBUTING F ACULTY Modeling TeamEarth System Component Jennifer Adam, WSUTerrestrial/Aquatic Serena Chung, WSUAtmospheric Alex Guenther, NCARAtmospheric/Terrestrial John Harrison, WSUTerrestrial/Aquatic Brian Lamb, WSUAtmospheric Ruby Leung, PNNLAtmospheric/Terrestrial Claudio Stockle, WSUTerrestrial Christina Tague, UCSBTerrestrial/Aquatic Joe Vaughan, WSUAtmospheric

14. C ONTRIBUTING F ACULTY Economics Team Michael Brady, WSU Yong Chen, OSU Jon Yoder, WSU Cyberinfrastructure Team Ananth Kalyanaraman, WSU Joe Vaughan, WSU Outreach/Education Team Chad Kruger, WSU Fok Leung, WSU Andy Perleberg, WSU Jennie Stephens, Clark U. Ecology Team Dave Evans, WSU John Harrison, WSU Christina Tague, UCSB

15. G OAL AND O BJECTIVES Overarching Goal: To improve the understanding of regional and decadal-scale C:N:H 2 O interactions in context of global change to better inform decision makers involved in natural and agricultural resource management. Specific Objectives: 1. Air to Land Linkage: To investigate the role that atmospheric processes play in land surface C:N:H 2 O cycles. 2. Coupled Air/Land: To explore how ecosystem changes in the PNW affect land/atmosphere interactions. 3. Coupled Air/Land/Human: To examine how potential policy changes might affect the interactions between C:N:H 2 O cycles and regional-scale climate. 4. Communication: To explore how to best communicate the model results to resource managers and policy makers.

16. Atmosphere

17. G LOBAL C LIMATE M ODEL (GCM): CCSM4  The General Circulation Model used in CESM  Community Climate System Model (CCSM) has four separate models:  Atmosphere general circulation model (AGCM)  Ocean general circulation model (OGCM)  Land surface model (soil moisture, vegetation roles, snow, …)  Sea ice model

18. Nested Simulations for the Pacific Northwest Outer domain: 36-km resolution 12-km resolution 4-km resolution Hourly Meteorology Output: Precipitation, Temperature, Relative Humidity,… Global Climate Model: Precipitation, Temperature, Windspeed, Radiation, … Regional Meteorology: WRF (Weather Research and Forecasting)

19. R EGIONAL A TMOSPHERIC C HEMISTRY : CMAQ  CMAQ: Community Multi-Scale Air Quality Model  Compatible with WRF, nested model domains  Solution to the pollutant conservation equation  50-100 chemical species, 100’s of reactions  Cloud chemistry  Detailed wet/dry deposition scheme

20. Terrestrial

21.  Regional to Global Scales: 1/16° to 2 ° resolution  Sub-grid variability (statistically)  1 to 24 hourly time step  Spatially distributed  Closes the water and energy budgets  Sub-models:  Snow  Frozen ground/permafrost  Lakes/wetlands UW/Princeton Variable Infiltration Capacity (VIC) Model L ARGE -S CALE H YDROLOGY : VIC (V ARIABLE I NFILTRATION C APACITY M ODEL )

22.  Developed by Claudio Stöckle at WSU  Crop development and growth (unstressed or stressed)  Simulation of growth under increased atmospheric CO 2 concentration  Water and nitrogen balance  Salinity  Residue fate  Soil erosion by water  Carbon sequestration  Greenhouse gas emissions (CO 2 and N 2 O) Cropping Systems: CropSyst

23. E COHYDROLOGY :RHESS YS (R EGIONAL H YDRO -E COLOGIC S IMULATION S YSTEM ) Slide courtesy of Christina Tague

24. T ERRESTRIAL M ODEL I NTEGRATION : I NTEGRATING D IFFERENT S CALES Crop 1 VIC grid cell (~ 40 km 2 ) Crop 2 Natural Vegetation CropSyst is invoked RHESSys is invoked

25. B IOGENIC E MISSIONS : MEGAN M ODEL OF E MISSIONS OF G ASES AND A EROSOLS FROM N ATURE Developed by Alex Guenther What emissions are needed for chemical transport model (CMAQ)? Point, area, mobile, fire, terrestrial biosphere = biogenic 134 emitted chemical species Isoprene Monoterpenes Oxygenated compounds Sesquiterpenes Nitrogen oxide

26. Aquatic

27. R IVER R OUTING AND R ESERVOIRS 1 2 3 4 VIC Hydrology Model Reservoir Model River Routing Model

28. Large river basins, annual scale, inputs at 0.5 degrees Multi-element (C,N,P,Si); Multi-form (dissolved, particulate, organic, inorganic) Estimates contributions of various nutrient inputs to export Slide borrowed from Kara Goodwin G LOBAL N UTRIENT E XPORT FROM WATERSHEDS (NEWS) Dissolved Inorganic N (DIN) = nitrate + nitrite + ammonia

29. Economics

30. E CONOMICS : CREM  Columbia River Basin Economic Model (CREM)  Human decision making under changes in  Resources (land, water, fertilizer)  Political environment  Agricultural management practices  Crop selection, irrigation, fertilization  Natural resources management decisions

31. M ODELING D OMAIN : T HE P ACIFIC N ORTHWEST

32. O UTREACH AND C OMMUNICATIONS R ESEARCH  WSU Extension will hold two meetings per year (for four years) with stakeholders in  Agriculture  Forestry  Tribes  Communications research question: How does regular interaction between BioEarth developers and land-use stakeholders influence the perceived relevance and perceived utility of a model to decision-making?

33. H OW CAN B IO E ARTH SERVE AS A FRAMEWORK FOR INTERDISCIPLINARY RESEARCH ?

34. W HY IS I NTERDISCIPLINARY R ESEARCH N EEDED IN THE E ARTH S CIENCES ?  Human knowledge doubles every X years (X depends on discipline, but is somewhere between 2 and 21 years)  True “Renaissance Men” no longer exist; We must pool our expertise to solve problems that span the breadth of human knowledge  “Stationarity is Dead” (Milly et al., Science, 2008)  We are entering a new regime where we can no longer rely on relationships developed from historical data  Extrapolation is not valid; New relationships must be explored  The New Frontier in modeling of Earth processes  Many “Stand Alone” models have reached a high level of sophistication; The New Frontier is in integrating these models  This needs a team of people who are experts in each of the component models

35. S OME R OADBLOCKS TO I NTERDISCIPLINARY R ESEARCH  Physical location  How to overcome: Cyberinfrastructure (videoconferencing; document, data, and code sharing)  University politics  How to overcome: Fair distribution of credit  Jargon: terminology specific to a discipline (loosely expanded to also include discipline-specific concepts or ideas)  How to overcome? This may be the biggest challenge.

36. B IO E ARTH : A VENUES OF C OLLABORATION AND COMMUNICATION  Between experimentalists and modelers in a discipline  Commonality: jargon of specific scientific discipline  Between modelers (multiple disciplines)  Commonality: computer jargon, modeling jargon  Between biophysical, computer, and social scientists  Commonality: jargon related to any common data inputs, research/academic jargon in general  Between scientists and decision makers  Commonality: jargon related to scientific information needed for decision making  Between scientists and land managers (land owners, farmers, forest managers, etc…)  Commonality: jargon related to land attributes More Commonality Less Commonality

37. H OW CAN WE SUCCESSFULLY COMMUNICATE ?  Focusing on our commonality in jargon to  Learn something of each other’s jargon  Learn what the other person does NOT know of your jargon  Develop a jargon common ground  Learn how to educate newcomers to this jargon common ground  What not to do  Neglect to take the time to educate a newcomer to the jargon common ground  Emphasize model output above the jargon common ground  Underestimate the intellectual capabilities of those outside of academia

38. S UMMARY  An Earth System Model is important for exploring 2- way linkages and feedbacks between earth system components.  BioEarth is a regional Earth System Model that explicitly accounts human activities as well as linkages between earth system and economic processes.  A roadblock to interdisciplinary research is discipline-specific jargon. BioEarth can be used as a framework for developing a common ground jargon that can be used not only for research but also for communicating research results to stakeholders.

39. N EED F URTHER I NFORMATION ?  Please contact me:  Jennifer Adam: jcadam@wsu.edu  Or visit our webpage: http://www.cereo.wsu.edu/bioearth/ http://www.cereo.wsu.edu/bioearth/ (currently in development)

40. B IO E ARTH

41. A PPROACH AND R ATIONALE  Integrate or link existing sophisticated “stand alone” models that are in continuous development  Atmosphere: meteorology, atmospheric chemistry  Terrestrial: hydrology, soil/plant biogeochemistry in cropped and natural systems, biogenic emissions  Aquatic: river routing, reservoir modeling, nutrient export  Economics  As the “stand alone” components continue to improve by their developers, the BioEarth will also continue to develop

42. Model Coupling Terminology  Model Linking  One-way  Two-way  Model Partial Integration  Model Full Integration

43. C AND N – RELEVANT PROCESSES IN RHESS YS Slide courtesy of Christina Tague

44. Physical System of Dams and Reservoirs Reservoir Operating Policies Reservoir Storage Regulated Streamflow Flood Control Energy Production Irrigation Consumption Streamflow Augmentation VIC Streamflow Time Series T HE R ESERVOIR M ODEL (C OL S IM ) Slide courtesy of Alan Hamlet

45. T – Transpiration I P – Interception capacity I – Infiltration Q – Runoff Q 01 – Drainage from 0 to 1 Q 02 – Drainage from 0 to 2 Q b – Baseflow ET 0 – Penman Monteith Pot. Evap. W 1 – water content in 1 W 2 - water content in 2 QbQb Q 12 T IPIP updated Soil moisture Redistribute I, W 1 and W 2 to CropSyst layers Q Q 01 ET 0, I, W 1, W 2 T, I P I CropSyst VIC VIC – C ROP S YST INTEGRATION Terrestrial

46. Point Sewage Non-point Natural N fixation Atmospheric N dep. Fertilizer N-fixation by crops Manure Crop export (-) Runoff Sewage treatment DIN in surface waters DIN exported from basin Retention in reservoirs In-stream removal Consumptive use Terrestrial retention NEWS2-DIN Nitrogen Inputs Slide borrowed from Kara Goodwin