Progress 09/01/19 to 08/03/23
Outputs
Target Audience: During the duration of the project, the project has reached the following audiences: 1) Students, scientists, and professionals 2) Growers in Alabama and the Central Valley of California 3) Extension specialists and crop consultants 4) Private industry involved in irrigation technology and management 5) Public agencies involved in agricultural water management Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided? The training opportunities were provided to students at Auburn University. This includes both classroom teaching and one-to-one mentoring. For the classroom teaching, results from the project have been incorporated into a graduate-level Agroclimatology course at Auburn University. Besides classroom teaching, graduate students have been participating in this project under the project director's or co-PIs' mentorships. This project improved these students' climate research and data science skills. In addition, the project also provides professional development for a research fellow to further enhance her technical and communication skills. It also created opportunities for students and professionals to attend and present at professional conferences, such as AGU, AMS meetings, etc., and publish in peer-reviewed scientific journals. How have the results been disseminated to communities of interest? We collaborated with the Alabama Cooperative Extension System and University of California Cooperative Extension specialists that help with dissemination of our research findings e.g., at County Extension meetings and field days. Beyond traditional extension methods, we leverage social media, including Twitter and YouTube. What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
The major accomplishments of the project can be summarized as follows: 1. We have developed a new framework to improve weekly precipitation and reference evapotranspiration (ETo) forecasts. The framework includes the ensemble model output statistics (EMOS) algorithm to postprocess multi-model weekly precipitation and ETo forecasts from reforecast database of several numerical models. The postprocessed forecasts were evaluated in different climate regions over the US, and showed significantly improved performance compared to the conventional multi-model ensemble approach. 2. We have developed an optimized framework using MODIS-based vegetation indices to forecast corn yields from mid-season. We evaluated different schemes accounting for: different machine learning techniques with a mid-season composite or multi-temporal composites as inputs, four training domains, 16-day composites versus daily interpolated composites involving the day of pixels as predictors, and different MODIS predictors. The results showed the linear regression models driven by the single latest composite in mid-season often outperformed elastic net and random forest models driven by multi-temporal composites. The forecast performance decreased with longer subsets of EVI composites being used. Overall, the best forecasts consistently outperformed concurrent National Agricultural Statistical Service (NASS) forecasts. 3. Our research has improved our understanding of climate impacts on crop water use and demand. First, we developed a new climate downscaling approach, namely super-resolution deep residual network (SRDRN), by adapting a convolutional based deep learning algorithm in computer vision field. This approach showed outstanding performance compared to the widely used analog technique, and is readily applicable to downscale climate projections from Global Climate Models (GCMs). Second, we evaluated how much cover crops could mitigate the impacts of mean and extreme climates during historical and future periods using a crop simulation model and downscaled and bias-corrected GCMs. The results showed growing cover crops increased system water use efficiency and buffered the impacts of climate extremes on crop yields and water losses. Lastly, a method was developed to calibrate the CROPGRO-Peanut model to simulate historical peanut yields in seven counties in the peanut belt. The method included calibrating spatial soil and genetic parameters. The model will evaluate different irrigation strategies for future climate adaptations. 4. We developed and evaluated an improved ET estimation method based on Priestley-Taylor Jet Propulsion Laboratory algorithm (PT-JPL) with HLS optical reflectance imagery and ECOSTRESS land surface temperature (LST) and surface emissivity, in addition to MODIS surface albedo and ERA5-Land climate reanalysis data. The new approach can be used to develop denser times series of ET from HLS imagery, which can also be used to map ET at 30 m spatial resolution. The new ET estimates are evaluated against ground-based observations from the AmeriFlux network and compared with the performance of the original ECOSTRESS PT-JPL ET estimates across different ecosystems and landcover settings over the continental United States. The results showed improved performance compared to the original PT-JPL algorithm. 5. We utilized a widely used crop simulator, the Decision Support System for Agrotechnology Transfer (DSSAT), to evaluate the performance of a model predictive control (MPC) based irrigation control scheme. The total irrigation amount from MPC is compared to the rule-based irrigation plan from DSSAT, which is further evaluated based on the crop yields from different irrigation plans. It is shown that although the MPC model does not consider the water stress for the crop, which DSSAT uses for the rule-based irrigation plan, with a careful determination of a growth-stage-based moisture trajectory as the set-point, MPC can achieve a similar crop yield with less total irrigation water amount. 6. We developed a customized DL model by incorporating customized loss functions, multitask learning, and physically relevant covariates to bias correct and downscale hourly precipitation data. We designed six scenarios to systematically evaluate the added values of weighted loss functions, multitask learning, and atmospheric covariates compared to the regular DL and statistical approaches. We found that all the scenarios with weighted loss functions performed notably better than the other scenarios with conventional loss functions and a quantile mapping-based approach at hourly, daily, and monthly time scales as well as extremes. Multitask learning showed improved performance on capturing fine features of extreme events and accounting for atmospheric covariates, highly improved model performance at hourly and aggregated time scales, while the improvement is not as large as from weighted loss functions. 7. We quantified the synergistic contributions of climate and management intensifications to maize yield trends from 1961 to 2017 in Iowa (United States) using a process-based modeling approach with a detailed climatic and agronomic observation database. We found that climate (management intensifications) contributes to approximately 10% (90%), 26% (74%), and 31% (69%) of the yield trends during 1961-2017, 1984-2013, and 1982-1998, respectively. However, the climate contributions show substantial decadal or multi-decadal variations, with the maximum decadal yield trends induced by temperature or radiation changes close to management intensifications induced trends while considerably larger than precipitation-induced trends. Management intensifications can produce more yield gains with increased precipitation but greater losses of yields with increased temperature, with extreme drought conditions diminishing the yield gains, while radiation changes have little effect on yield gains from management intensifications. Under the management condition of recent years, the average trend at the higher warming level was about twice lower than that at, the lower warming level, and the sensitivity of yield to warming temperature increased with management intensifications from 1961 to 2017. Due to such synergistic effects, management intensifications must account for global warming and incorporate climate adaptation strategies to secure future crop production. 8. We comprehensively assessed trends, spatiotemporal variability, and drivers of soil moisture (SM) and evaporative demand (ED) flash droughts over the contiguous United States (CONUS) during 1981-2018 using hierarchical clustering, wavelet analysis, and bootstrapping conditional probability approaches. Results show that flash droughts occur in all regions in CONUS, with Central and portions of the Eastern US showing the highest percentage of weeks in flash drought. ED flash drought trends are significantly increasing in all regions, while SM flash drought trends were relatively weaker across CONUS, with small significant increasing trends in the South and West regions and a decreasing trend in the Northeast. Rising ED flash drought trends are related to increasing temperature trends, while SM flash drought trends are strongly related to trends in weekly precipitation intensity besides weekly average precipitation and evapotranspiration. In terms of temporal variability, high-severity flash droughts occurred every 2-7 years, corresponding with ENSO periods. For most CONUS regions, severe flash droughts occurred most often during La Niña and when the American Multidecadal Oscillation was in a positive phase. Pacific Decadal Oscillation negative phases and Artic Oscillation positive phases were also associated with increased flash drought occurrences in several regions. These findings may have implications for informing long-term flash drought predictions and adaptations.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2022
Citation:
Lesinger, K., D. Tian. 2022. Trends, Variability, and Drivers of Flash Droughts in the Contiguous United States. Water Resources Research, 58, e2022WR032186.
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Zhen, X., Huo, W., Tian, D., Zhang, Q., Sanz-Saez, A., Chen, C.Y. and Batchelor, W.D., 2023. County level calibration strategy to evaluate peanut irrigation water use under different climate change scenarios. European Journal of Agronomy, 143, p.126693.
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Takhellambam, B.S., P. Srivastava, J. Lamba, R. McGehee, H. Kumar, D. Tian. 2023. Projected Mid-Century Rainfall Erosivity Under Climate Change Over the Southeastern United States, Science of The Total Environment, p. 161119.
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Medina, H., D. Tian. 2023. Synergistic contributions of climate and management intensifications to maize yield trends from 1961 to 2017. Environmental Research Letters, 18, 024020.
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Wang, F., D. Tian, and M. Carroll. 2023. Customized Deep Learning for Precipitation Bias Correction and Downscaling. Geoscientific Model Development, 16, 535556.
- Type:
Journal Articles
Status:
Published
Year Published:
2023
Citation:
Schillerberg, T. and Tian, D., 2023. Changes in crop failures and their predictions with agroclimatic conditions: Analysis based on earth observations and machine learning over global croplands. Agricultural and Forest Meteorology, 340, p.109620.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Lesinger, K, D. Tian, and H. Wang. Subseasonal forecast skill of US flash droughts from 2000 to 2022 in global dynamic models. 2nd National Flash Drought Workshop, Boulder, CO, May 2 to May 4, 2023
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Kumar, Sudhanshu, D. Tian. Understanding sub-seasonal predictability of soil moisture flash drought in the southeastern US based on causal analysis of large-scale climate patterns. 2nd National Flash Drought Workshop, Boulder, CO, May 2 to May 4, 2023
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2023
Citation:
Lesinger. K, D. Tian, and H. Wang. Subseasonal forecast skill of US flash droughts from 2000 to 2022 in global dynamic models. 103rd AMS Annual Meeting, Denver, 812 January 2023
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Lesinger. K, D. Tian, and H. Wang. Subseasonal flash drought forecast skill over the contiguous United States. AGU Fall Meeting 2022, Chicago, IL and online, 12 - 16 December 2022.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Wang, F., D. Tian, and M. Carroll. Customized Deep Learning for Precipitation Bias Correction and Downscaling. AGU Fall Meeting 2022, Chicago, IL and online, 12 - 16 December 2022.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Lesinger, K., D. Tian, and H. Wang. Flash drought forecast skill in subseasonal prediction models over the contiguous United States. Southeast Climate Adaptation Science Symposium, September 19-21, 2022, Gulf Shores, Alabama.
- Type:
Book Chapters
Status:
Published
Year Published:
2023
Citation:
Schillerberg, T., D. Tian. 2023. Climate Impacts on Crop Productions. In: Zhang, Q., Encyclopedia of Smart Agriculture Technologies. Springer, Cham.
|
Progress 09/01/21 to 08/31/22
Outputs
Target Audience:During the project reporting period, the project produced research and outreach products that reached the following audiences: 1) Growers in Alabama. 2) Extension specialists and crop consultants 3) Private industry involved in irrigation technology and management. 4) Students, scientists, and professionals Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided?The training opportunities were provided to students at Auburn University. This includes both classroom teaching and one-to-one mentoring. For classroom teaching, results from the project have been incorporated into a graduate-level Agroclimatology course at Auburn University. The students were introduced to evapotranspiration, weather and climate data and analysis, and remote sensing in agricultural water management. Besides classroom teaching, three graduate students have been participating in this project under the mentorships of the project director or co-PIs. These students have improved their climate research and data science skills through this project. In addition, the project also provides professional development for a research fellow to further enhance her technical and communication skills. It also created opportunities for students and professionals to attend and present at professional conferences, such as AGU, ASABE, ASA-CSSA-SSSA meetings, etc., and publish in peer-reviewed scientific journals. How have the results been disseminated to communities of interest?During the project reporting period, the project produced research and outreach products that reached the following audiences: 1) Growers in Alabama. 2) Extension specialists and crop consultants 3) Private industry involved in irrigation technology and management. 4) Students, scientists, and professionals What do you plan to do during the next reporting period to accomplish the goals?During the next reporting period, we will finish improving the remote sensing-based ET estimation approach, and continue working on developing a deep learning-based framework to estimate daily ET at 30-m resolution. We will continue developing the hybrid human-artificial intelligence-based framework and test it with DSSAT model simulations from point to county scales.
Impacts
What was accomplished under these goals?
During the past project reporting year, we have accomplished the following tasks to achieve the project goal: 1) We developed and evaluated an improved ET estimation method based on Priestley-Taylor Jet Propulsion Laboratory algorithm (PT-JPL) with HLS optical reflectance imagery and ECOSTRESS land surface temperature (LST) and surface emissivity, in addition to MODIS surface albedo and ERA5-Land climate reanalysis data. The new approach can be used to develop denser times series of ET from HLS imagery, which can also be used to map ET at 30 m spatial resolution. The new ET estimates are evaluated against ground-based observations from the AmeriFlux network and compared with the performance of the original ECOSTRESS PT-JPL ET estimates across different ecosystems and landcover settings over the continental United States. The results showed improved performance compared to the original PT-JPL algorithm. 2) We comprehensively evaluated the Super Resolution Deep Residual Network (SRDRN) deep learning model for climate downscaling and bias correction. The SRDRN model sequentially stacked 20 GCMs with single or multiple input-output channels, so that the biases can be efficiently removed based on the relative relations among different GCMs against observations, and the intervariable dependences can be retained for multivariate bias correction. It corrected biases in spatial dependences by deeply extracting spatial features and making adjustments for daily simulations according to observations. For univariate SRDRN, it considerably reduced larger biases of Tmean in space, time, as well as extremes compared to the quantile delta mapping (QDM) approach. For multivariate SRDRN, it performed better than the dynamic Optimal Transport Correction (dOTC) method and reduced greater biases of Tmax and Tmin but also reproduced intervariable dependences of the observations, where QDM and dOTC showed unrealistic artifacts. 3) We contributed to an overview of subseasonal predictability for case studies of some of the most prominent extreme events across the globe using the ECMWF S2S prediction system: heatwaves, cold spells, heavy precipitation events, and tropical and extratropical cyclones. The considered heatwaves exhibit predictability on time scales of 3-4 weeks, while this time scale is 2-3 weeks for cold spells. Precipitation extremes are the least predictable among the considered case studies. Tropical cyclones, on the other hand, can exhibit probabilistic predictability on time scales of up to 3 weeks, which in the presented cases was aided by remote precursors such as the Madden-Julian oscillation. For extratropical cyclones, lead times are found to be shorter. These case studies clearly illustrate the potential for event-dependent advance warnings for a wide range of extreme events. 4) We generated novel 15-minute precipitation datasets from hourly precipitation datasets obtained from five NA-CORDEX downscaled climate models under RCP 8.5 scenario for the historical (1970-1999) and projected (2030-2059) years over the Southeast United States using a modified version of the stochastic method. The results showed the conservation of mass of the precipitation inputs. Furthermore, the probability of zero precipitation, variance of precipitation, and maximum precipitation in the disaggregated data matched well with the observed precipitation characteristics. The generated 15-minute precipitation data can be used in all scientific studies that require precipitation data at that resolution. 5) We utilized a widely used crop simulator, the Decision Support System for Agrotechnology Transfer (DSSAT) to evaluate the performance of a model predictive control (MPC) based irrigation control scheme. The total irrigation amount from MPC is compared to the rule-based irrigation plan from DSSAT, which is further evaluated based on the crop yields from different irrigation plans. It is shown that although the MPC model does not consider the water stress for the crop, which DSSAT uses for the rule-based irrigation plan, with a careful determination of a growth-stage-based moisture trajectory as the set-point, MPC can achieve a similar crop yield with less total irrigation water amount.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2022
Citation:
Wang, F., D. Tian. 2022. On deep learning-based bias correction and downscaling of multiple climate models simulations. Climate Dynamics, pp. 1-18.
- Type:
Journal Articles
Status:
Published
Year Published:
2022
Citation:
Takhellambam, B.S., Srivastava, P., Lamba, J., McGehee, R.P., Kumar, H. and Tian, D., 2022. Temporal disaggregation of hourly precipitation under changing climate over the Southeast United States. Scientific Data, 9(1), pp.1-14.
- Type:
Journal Articles
Status:
Published
Year Published:
2022
Citation:
Domeisen, D.I. and 39 coauthors including *H. Medina and D. Tian. 2022. Advances in the subseasonal prediction of extreme events: Relevant case studies across the globe. Bulletin of the American Meteorological Society, 103(6), E1473-E1501.
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Saminathan, S., Medina, H., Mitra, S. and Tian, D., 2021. Improving short to medium-range GEFS precipitation forecast in India. Journal of Hydrology, 598, p.126431.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Jang, J., D. Tian, and Q. He. Accepted. Model-based Irrigation Control using Model Predictive Control and DSSAT Crop Simulator. 2022 American Control Conference, June 6, 2022, Atlanta, Georgia.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Rashid, T., D. Tian. Improving ECOSTRESS-based Evapotranspiration Estimates Using Harmonized Landsat Sentinel-2 Imagery. EGU General Assembly 2022, May 23-27, 2022, Vienna, Austria
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Tian, D. AI-driven environmental analytics for data-informed climate adaptation and resilience. Envisioning 2050 in the Southeast: AI-Driven Innovations in Agriculture, March 9-11, 2022. Auburn, Alabama.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Rashid, T., D. Tian. Improving ECOSTRESS-based Evapotranspiration Estimates Using Harmonized Landsat Sentinel-2 Imagery and Deep Learning. Envisioning 2050 in the Southeast: AI-Driven Innovations in Agriculture, March 9-11, 2022, Auburn, Alabama.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2022
Citation:
Wang, F., Di Tian. On deep learning-based bias correction and downscaling of multiple climate models simulations. Envisioning 2050 in the Southeast: AI-Driven Innovations in Agriculture, March 9-11, 2022, Auburn, Alabama.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
Wang, F., Di Tian. Bias Correction of Multi-model GCMs Daily Temperature based on Deep Learning. 2021 AGU Fall Meeting, 13-17 December 2021, New Orleans, LA and online.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
Zhen, X., W. Huo, D. Tian, Q. Zhang, A. Sanz-Saez, C. Chen, W. D. Batchelor. County-Level Calibration Strategy to Evaluate Peanut Irrigation Water Use Under Different Climate Change Scenarios. 2021 ASA, CSSA, SSSA International Annual Meeting, November 7-10, Salt Lake City, UT.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
Wang, F., D. Tian. On deep learning bias correction for GCMs daily temperature outputs. The 3rd NOAA Workshop on Leveraging AI in Environmental Sciences, September 1317, 2021, Boulder, Colorado & Virtual
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
Rashid, T., D. Tian. Predicting Field-scale Daily Evapotranspiration Using Multi-Source Spaceborne Remote Sensing Imagery and Deep Learning. Alabama Water Resources Conference, September 8-10, 2021, Gulf Shore, AL.
|
Progress 09/01/20 to 08/31/21
Outputs
Target Audience:During the project reporting period, the project produced research and outreach products that reached the following audiences: 1) Growers in Alabama and Central Valley of California. 2) Extension specialists and crop consultants 3) Private industry involved in irrigation technology and management. 4) Students, scientists, and professionals Changes/Problems:In the original proposal, we propose to develop 10-meter resolution ET and soil moisture. 10-m resolution optical reflectance data from Sentinel-2 is only available every 5 days. We decided to use Harmonized Landsat Sentinel-2 (HLS) dataset, which is at 30-m resolution but is available every 2-3 days. In addition, because the other datasets, such as land cover, topography, and digital soil data, are available at 30-m resolution, we decided to make our final ET and soil moisture product at 30-m resolution instead of 10-m. What opportunities for training and professional development has the project provided?The training opportunities were provided to both students in Auburn University and University of California Davis. This includes both classroom teaching and one-to-one mentoring. For the classroom teaching, results from the project have been incorporated into a graduate-level Agroclimatology course at Auburn University and an undergraduate agricultural water management course at the University of California Davis. The students were introduced to evapotranspiration, weather and climate data and analysis, and remote sensing in agricultural water management. Besides classroom teaching, six graduate students have been participating in this project under the mentorships of the project director or co-PIs. These students have improved their research and data science skills through this project. In addition, the project also provides professional development for two postdocs and one research scientist to further enhance their technical and communication skills. It also created opportunities for students and professionals to attend and present in professional conferences, such as AGU, ASABE, ASA-CSSA-SSSA meetings, etc., and publish in peer-reviewed scientific journals. How have the results been disseminated to communities of interest?We collaborated with the Alabama Cooperative Extension System and University of California Cooperative Extension specialists that help with dissemination of our research findings e.g., at County Extension meetings and field days. Beyond traditional methods of extension, we leverage social media including Twitter and YouTube. What do you plan to do during the next reporting period to accomplish the goals?During the next reporting period, we will finish developing and evaluating remote sensing-based ET estimation approach, and continue working on developing a deep learning-based framework to estimate daily soil moisture at 30-m resolution. We will continue developing the hybrid human-artificial intelligence-based framework and test it with DSSAT model simulations from point to county scales.
Impacts
What was accomplished under these goals?
During the past project reporting year, we have accomplished following tasks to achieve the project goal: 1) Evapotranspiration (ET) is a fundamental variable for water management. We have modified the Priestley-Taylor Jet Propulsion Laboratory (PT-JPL) algorithm to improve field-scale evapotranspiration estimations using the harmonized Landsat Sentinel-2 (HLS) optical reflectance data, the ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS) land surface temperature data, and the ERA5 Land reanalysis data. The algorithm is being evaluated against observed ET at several eddy covariance sites from AmeriFlux network and compared with the original approach. 2) Precipitation is another fundamental variable for water management. We have developed a new framework to improve weekly precipitation forecasts with up to four weeks lead time. The framework includes the left censored ensemble model output statistics (EMOS) algorithm to postprocess multi-model weekly precipitation forecasts from the Subseasonal Experiment (SubX) database. The postprocessed forecasts were evaluated in different climate regions over the US, and showed significantly improved performance compared to the conventional multi-model ensemble approach. This work has been published in the Journal of Hydrometeorology. 3) Crop yield forecast information can be useful for informing water and crop management. We have been developing an optimized framework using MODIS-based vegetation indices to forecast corn yields from mid-season. We evaluated different schemes accounting for: different machine learning techniques with a mid-season composite or multi-temporal composites as inputs, four training domains, 16-day composites versus daily interpolated composites involving the day of pixels as predictors, and different MODIS predictors. The results showed the linear regression models driven by the single latest composite in mid-season often outperformed elastic net and random forest models driven by multi-temporal composites. The forecast performance decreased with longer subsets of EVI composites being used. Overall, the best forecasts consistently outperformed concurrent National Agricultural Statistical Service (NASS) forecasts. This work has been published in the Journal of Applied Earth Observations and Geoinformatics. 4) Understanding climate impacts on crop water use and demand is the prerequisite step for developing strategies to adapt to the changing climate. We have accomplished several activities to achieve this objective. First, we developed a new climate downscaling approach, namely super resolution deep residual network (SRDRN), by adapting a convolutional based deep learning algorithm in computer vision field. This approach showed outstanding performance compared to the widely used analog technique, and is readily applicable to downscale climate projections from Global Climate Models (GCMs). This work has been published in the journal Water Resources Research. Second, we evaluated how much cover crops could mitigate the impacts from mean and extreme climates during both historical and future periods using a crop simulation model and downscaled and bias corrected GCMs. The results showed Growing cover crops increased system water use efficiency and buffered impacts of climate extremes on crop yields and water losses. This work has been published in Agricultural Water Management. Lastly, a method was developed to calibrate the CROPGRO-Peanut model to simulate historical peanut yields in seven counties in the peanut belt. The method included calibrating spatial soil and genetic parameters. The model will be used to evaluate different irrigation strategies for future climate adaptations. 5) To achieve the research objective of developing a hybrid human-artificial intelligence-based framework for irrigation scheduling, a graduate student went through extensive training on Decision Support System for Agrotechnology Transfer (DSSAT), a software application program that comprises dynamic crop growth simulation models for over 42 crops. To formulate the sequential optimization of irrigation scheduling as an economic model predictive control (MPC) problem, Mr. Jang is currently learning the fundamentals and implementation of MPC. 6) A new web-based software for irrigation scheduling: We have established processing tomato and almond precision irrigation experimental research sites. We also developed algorithms for optimizing irrigation scheduling based on yield or profit. We are working on integrating these advanced algorithms into the FARMs web app that is available at https://ciswma.lawr.ucdavis.edu/.
Publications
- Type:
Journal Articles
Status:
Accepted
Year Published:
2021
Citation:
Li, Yanzhong, D. Tian, and H. Medina. 2021. Multi-model Subseasonal Precipitation Forecasts over the Contiguous United States: Skill Assessment and Postprocessing. Journal of Hydrometeorology, in press
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Asadi, P., and D. Tian. 2021. Estimating leaf wetness duration with machine learning and climate reanalysis data. Agricultural and Forest Meteorology, 307, p.108548.
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Li, Yizhuo, D. Tian, G. Feng, W. Yang, L. Feng. 2021. Climate change and cover crop effects on water use efficiency of a corn-soybean rotation system. Agricultural Water Management, 255, p.107042
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Wang, F., D. Tian, L. Lowe, L. Kalin, and J. Lehrter. 2021. Deep learning for daily precipitation and temperature downscaling. Water Resources Research, 57, e2020WR029308
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Tian, D., X. He, P. Srivastava, and L. Kalin. 2021. A hybrid framework for forecasting monthly reservoir inflow based on machine learning techniques with dynamic climate forecasts, satellite-based data, and climate phenomenon information. Stochastic Environmental Research and Risk Assessment, pp.1-23.
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Medina, H., D. Tian, and A. Abebe. 2021. On optimizing a MODIS-based framework for in-season corn yield forecast. International Journal of Applied Earth Observation and Geoinformatics, 95, p.102258.
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Xue, J., Z. Huo, I. Kisekka. 2021. Assessing impacts of climate variability and changing cropping patterns on regional evapotranspiration, yield and water productivity in Californias San Joaquin watershed, Agricultural Water Management, Volume 250, 2021, 106852.
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Liu, J., U. Gull, D.H Putnam, I. Kisekka. 2021. Variable Rate Irrigation Uniformity Model for Linear Move Sprinkler Systems. Transactions of the ASABE. doi: 10.13031/trans.14313.
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Kim, J.S., I. Kisekka. 2021. FARMs: A Geospatial Crop Modeling and Agricultural Water Management System. ISPRS Int. J. Geo-Inf. 2021, 10, 553. https://doi.org/10.3390/ijgi10080553.
- Type:
Journal Articles
Status:
Published
Year Published:
2021
Citation:
Jimenez, A.F., Ortiz, B.V., Bondesan, L., Morata, G. and Damianidis, D., 2021. Long Short-Term Memory Neural Network for irrigation management: a case study from Southern Alabama, USA. Precision Agriculture, 22(2), pp.475-492.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2020
Citation:
Wang, F., D. Tian, L. Lowe, L. Kalin, J. Lehrter, and B. Dzwonkowski. Deep Learning for Daily Precipitation and Temperature Downscaling. 2020 AGU Fall Meeting, Dec. 1-17. Virtual.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2020
Citation:
Tian, D., 2020. Developing an Analytical Framework Based on Earth System Observations Forecasts: Towards Data-Informed Climate-Smart Management for Agriculture and Water Resources. 2020 ASA-CSSA-SSSA Annual International Annual Meetings. Nov. 9-13. Virtual.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2021
Citation:
Kisekka, I. 2021. Data-driven site-specific zone irrigation management ASABE Annual International Meeting. Virtual.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2020
Citation:
Peddinti, S. R. and I. Kisekka. 2020. Evapotranspiration estimation with high-resolution aerial imagery over almond orchards using TSEB model. AGU 2020 Virtual.
- Type:
Theses/Dissertations
Status:
Published
Year Published:
2020
Citation:
Hanoi Media, PhD Dissertation: Data-driven agroclimate modeling and forecasting based on earth observations and predictions: A study of evapotranspiration, precipitation, and crop yields , Auburn University, 2020.
|
Progress 09/01/19 to 08/31/20
Outputs
Target Audience:During the reporting period, the project produced research and outreach products that reached the following audiences: 1. Farmers that grow row crops in the north Alabama and Central Valley of California. 2. Extension specialists and farm advisors working on irrigation related issues. 3. Private industry involved in irrigation technology and management (established collaborations with Netafim and Jain Irrigation). 4. Public agencies involved in agricultural water management. Changes/Problems:Due to COVID-19, the University closed and research activities were ramped down to contain the virus. It affected the fieldwork components of this project.We plan to start field experiments next year in May 2021. In addition, we plan to search for any existing high-quality experimental datato support the work on model parameterization and validation. What opportunities for training and professional development has the project provided?The training opportunities were provided to both students in Auburn University and University of California Davis. This includes both classroom teaching and one-to-one mentoring. For the classroom teaching, the project has been incorporated into a graduate-level agroclimatology and modeling course in Auburn. The enrollment during the reporting period was 10 graduate students. Students have been learning modeling evapotranspiration and analyzing climatedata in the class.A module on model-based irrigation management in irrigation systems and water management has beenadded to upper undergraduate class at the University of California Davis. The enrollment during the reporting period was 15 students (mix of undergraduate and graduate students). The students were introduced to crop modeling and remote sensing in agricultural water management. Besides classroom teaching, three graduate students have been working on this project under the mentorships of the project director or co-PI. These students have improved their research skills by conducting this project. The project also created opportunities for the investigators to attend and present in different conferences, such as AGU, ASABE, ASA-CSSA-SSSA, etc. How have the results been disseminated to communities of interest?
Nothing Reported
What do you plan to do during the next reporting period to accomplish the goals?We plan to combine our reference evapotranspiration forecast products with satellite remote sensing to forecast potential evapotranspiration and evaluate the forecast with ground observations. We also plan to work on field-specificsoil moisture forecasts and improve crop model calibration algorithms. Besides that, we will start implementing artificial intelligence-based optimization algorithms to improve irrigation decisions.In addition, we will continue parametrizing the fresh tomato model and peanut model in DSSAT based on CROPGRO. After parameterizing the model, we plan to test it against available field data. We also plan to improve the FARMs web app by adding gSSURGO, and linking it to DSSAT using existing Schema's in FARMs. We plan to continue extension and outreach activities related to advanced Climate Smart Irrigation management.
Impacts
What was accomplished under these goals?
Following tasks have been accomplished to achieve the project goal: 1. Developed reference ETforecasts with numerical weather predictions and statistical postprocessing methods. The reference ET forecasts were computed using FAO-56 equation withThemaximum and minimum temperature, solar radiation, wind speed, and dew point temperaturefrom using retrospective forecasts from three The International Grand Global Ensemble (TIGGE) models: including the European Centre for Medium-Range Weather Forecasts (ECMWF), the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS), and the United Kingdom Meteorological Office (UKMO) forecasts. Three postprocessing methods includingthe nonhomogeneous Gaussian regression (NGR), affine kernel dressing (AKD), and Bayesian model averaging (BMA) techniques. The three probabilistic approaches generally outperform conventional procedures based on the simple bias correction of single-model forecasts, with the NGR post-processing of the ECMWF and ECMWF-UKMO forecasts providing the most cost-effective reference ET forecasting. 2. Developingmethods to calibrate the DSSAT-Cropgro-Peanut model to simulate county level yields in the peanut belt. We have conducted calibrations for Houston County, Alabama, which is primarily a non-irrigated county. We used the Geospatial Simulation optimizer in QGIS to calibrate soil water holding capacity and the root distribution function to minimize error between simulated and observed yield for 2006-2017. The model was able to explain 79% of annual yield variability. We have initiated calibrations for Collinsworth County, Texas. 3. Start parameterizing the CROPGRO tomato model in DSSAT.
Publications
- Type:
Journal Articles
Status:
Published
Year Published:
2020
Citation:
Medina, H., D. Tian. 2020. Comparison of probabilistic post-processing approaches for improving numerical weather prediction-based daily and weekly reference evapotranspiration forecasts. Hydrology and Earth System Sciences, 24(2).
- Type:
Journal Articles
Status:
Published
Year Published:
2019
Citation:
He, X., L. Estes, M. Konar, D. Tian, D. Anghileri, K. Baylis, T. Evans, J. Sheffield. 2019. Integrated approaches to understanding and reducing drought impact on food security across scales. Current Opinion in Environmental Sustainability, 40, pp. 43-54.
- Type:
Journal Articles
Status:
Published
Year Published:
2020
Citation:
Xue, J., KM Bali, S Light, T Hessels, I Kisekka. 2020. Evaluation of remote sensing-based evapotranspiration models against surface renewal in almonds, tomatoes and maize. Agricultural Water Management. Volume 238:106228.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2020
Citation:
Tian, D. P. Asadi, H. Medina, B. Ortiz, and I. Kesikka. A Climate Smart Framework for Forecasting Field-level Potential Evapotranspiration and Irrigation Requirement with Numerical Weather Predictions and Satellite Remote Sensing. 2020 European Geosciences Union Generally Assembly, Online, 48 May 2020
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2020
Citation:
Medina, H., D. Tian. A dynamic-statistical approach for probabilistic forecasting of daily soil moisture in the United States. 2020 European Geosciences Union Generally Assembly, Online, 48 May 2020
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Medina, H., D. Tian. Bayesian Crop Model Optimization for Corn Yield Forecasting Over the U.S. Corn Belt. 2019 AGU Annual Meeting, Dec. 9-13, San Francisco, California
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Li, Y., D. Tian, G. Feng, and L. Feng. Does cover crop mitigate climate-induced nitrogen loss from a Maize-soybean Cropping System? 2019 ASA-CSSA-SSSA Annual International Annual Meetings. Nov. 10-13. San Antonio, Texas.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Bill Rice and Isaya Kisekka. Automated Basin-wide ET Estimation Using the SEBS Method to Improve Groundwater Sustainability Plan Development. AGU Fall Meeting 2019.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Iael Raij-Hoffman, Sushant Mehan, Kenneth Miller, George Paul, Yohannes Yimam, John Dickey, Tim Hartz, Thomas Harter, Isaya Kisekka. A Multi-scale Modeling Assessment of Nitrogen Leaching From Central Valley Irrigated Processing Tomatoes. CDFA FREP Conference 2019.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Isaya Kisekka. Sustainable Agricultural Water Management for Enhanced Productivity and Environmental Outcomes, California State University, Chico California Invited Guest Lecture in the department of Geology and Environmental Engineering. November 4. Chicco California.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2020
Citation:
Isaya Kisekka. Sustainable Water Management in Agriculture for improved Productivity, Economic and Environmental Outcomes, FFAR Foster our Future, Washington DC Feb. 4-5th 2020.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Li, Y., D. Tian, G. Feng, and L. Feng. Does Cover Crop Mitigate Climate Extremes Impact on System Water Use Efficiency in a Maize-Soybean Cropping System? 2019 ASABE Annual International Meeting, July 7-10, Boston, Massachusetts
- Type:
Journal Articles
Status:
Published
Year Published:
2019
Citation:
Schillerberg, T., D. Tian, and R. Miao. 2019. Spatiotemporal patterns of maize and winter wheat yields in the United States: predictability and impact from climate oscillations. Agricultural and Forest Meteorology, 275 (2019): 208-222.
- Type:
Journal Articles
Status:
Accepted
Year Published:
2020
Citation:
Schillerberg, T., D. Tian. 2020. Changes of crop failure risks in the United States associated with large-scale climate oscillations in the Atlantic and Pacific Oceans. Environmental Research Letters, in press.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Asadi, P., D. Tian. Developing Improved Leaf Wetness Duration Models with Machine Learning and Climate Reanalysis Data. 2019 AGU Annual Meeting, Dec. 9-13, San Francisco, California
|
Progress 06/01/19 to 05/31/20
Outputs
Target Audience:During the reporting period, the project produced research and outreach products that reached the following audiences: 1. Farmers that grow row crops in the north Alabama and Central Valley of California. 2. Extension specialists and farm advisors working on irrigation related issues. 3. Private industry involved in irrigation technology and management (established collaborations with Netafim and Jain Irrigation). 4. Public agencies involved in agricultural water management. Changes/Problems:Due to COVID-19, the University closed and research activities were ramped down to contain the virus. It affected the fieldwork components of this project.We plan to start field experiments next year in May 2021. In addition, we plan to search for any existing high-quality experimental datato support the work on model parameterization and validation. What opportunities for training and professional development has the project provided?The training opportunities were provided to both students in Auburn University and University of California Davis. This includes both classroom teaching and one-to-one mentoring. For the classroom teaching, the project has been incorporated into a graduate-level agroclimatology and modeling course in Auburn. The enrollment during the reporting period was 10 graduate students. Students have been learning modeling evapotranspiration and analyzing climatedata in the class.A module on model-based irrigation management in irrigation systems and water management has beenadded to upper undergraduate class at the University of California Davis. The enrollment during the reporting period was 15 students (mix of undergraduate and graduate students). The students were introduced to crop modeling and remote sensing in agricultural water management. Besides classroom teaching, three graduate students have been working on this project under the mentorships of the project director or co-PI. These students have improved their research skills by conducting this project. The project also created opportunities for the investigators to attend and present in different conferences, such as AGU, ASABE, ASA-CSSA-SSSA, etc. How have the results been disseminated to communities of interest?
Nothing Reported
What do you plan to do during the next reporting period to accomplish the goals?We plan to combine our reference evapotranspiration forecast products with satellite remote sensing to forecast potential evapotranspiration and evaluate the forecast with ground observations. We also plan to work on field-specificsoil moisture forecasts and improve crop model calibration algorithms. Besides that, we will start implementing artificial intelligence-based optimization algorithms to improve irrigation decisions.In addition, we will continue parametrizing the fresh tomato model and peanut model in DSSAT based on CROPGRO. After parameterizing the model, we plan to test it against available field data. We also plan to improve the FARMs web app by adding gSSURGO, and linking it to DSSAT using existing Schema's in FARMs. We plan to continue extension and outreach activities related to advanced Climate Smart Irrigation management.
Impacts
What was accomplished under these goals?
Following tasks have been accomplished to achieve the project goal: 1. Developed reference ETforecasts with numerical weather predictions and statistical postprocessing methods. The reference ET forecasts were computed using FAO-56 equation withThemaximum and minimum temperature, solar radiation, wind speed, and dew point temperaturefrom using retrospective forecasts from three The International Grand Global Ensemble (TIGGE) models: including the European Centre for Medium-Range Weather Forecasts (ECMWF), the National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS), and the United Kingdom Meteorological Office (UKMO) forecasts. Three postprocessing methods includingthe nonhomogeneous Gaussian regression (NGR), affine kernel dressing (AKD), and Bayesian model averaging (BMA) techniques. The three probabilistic approaches generally outperform conventional procedures based on the simple bias correction of single-model forecasts, with the NGR post-processing of the ECMWF and ECMWF-UKMO forecasts providing the most cost-effective reference ET forecasting. 2. Developingmethods to calibrate the DSSAT-Cropgro-Peanut model to simulate county level yields in the peanut belt. We have conducted calibrations for Houston County, Alabama, which is primarily a non-irrigated county. We used the Geospatial Simulation optimizer in QGIS to calibrate soil water holding capacity and the root distribution function to minimize error between simulated and observed yield for 2006-2017. The model was able to explain 79% of annual yield variability. We have initiated calibrations for Collinsworth County, Texas. 3. Start parameterizing the CROPGRO tomato model in DSSAT.
Publications
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2020
Citation:
Tian, D. P. Asadi, H. Medina, B. Ortiz, and I. Kesikka. A Climate Smart Framework for Forecasting Field-level Potential Evapotranspiration and Irrigation Requirement with Numerical Weather Predictions and Satellite Remote Sensing. 2020 European Geosciences Union Generally Assembly, Online, 48 May 2020
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2020
Citation:
Medina, H., D. Tian. A dynamic-statistical approach for probabilistic forecasting of daily soil moisture in the United States. 2020 European Geosciences Union Generally Assembly, Online, 48 May 2020
- Type:
Journal Articles
Status:
Published
Year Published:
2020
Citation:
Medina, H., D. Tian. 2020. Comparison of probabilistic post-processing approaches for improving numerical weather prediction-based daily and weekly reference evapotranspiration forecasts. Hydrology and Earth System Sciences, 24(2).
- Type:
Journal Articles
Status:
Published
Year Published:
2019
Citation:
He, X., L. Estes, M. Konar, D. Tian, D. Anghileri, K. Baylis, T. Evans, J. Sheffield. 2019. Integrated approaches to understanding and reducing drought impact on food security across scales. Current Opinion in Environmental Sustainability, 40, pp. 43-54.
- Type:
Journal Articles
Status:
Published
Year Published:
2020
Citation:
Xue, J., KM Bali, S Light, T Hessels, I Kisekka. 2020. Evaluation of remote sensing-based evapotranspiration models against surface renewal in almonds, tomatoes and maize. Agricultural Water Management. Volume 238:106228.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Medina, H., D. Tian. Bayesian Crop Model Optimization for Corn Yield Forecasting Over the U.S. Corn Belt. 2019 AGU Annual Meeting, Dec. 9-13, San Francisco, California
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Li, Y., D. Tian, G. Feng, and L. Feng. Does cover crop mitigate climate-induced nitrogen loss from a Maize-soybean Cropping System? 2019 ASA-CSSA-SSSA Annual International Annual Meetings. Nov. 10-13. San Antonio, Texas.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Bill Rice and Isaya Kisekka. Automated Basin-wide ET Estimation Using the SEBS Method to Improve Groundwater Sustainability Plan Development. AGU Fall Meeting 2019.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Iael Raij-Hoffman, Sushant Mehan, Kenneth Miller, George Paul, Yohannes Yimam, John Dickey, Tim Hartz, Thomas Harter, Isaya Kisekka. A Multi-scale Modeling Assessment of Nitrogen Leaching From Central Valley Irrigated Processing Tomatoes. CDFA FREP Conference 2019.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Isaya Kisekka. Sustainable Agricultural Water Management for Enhanced Productivity and Environmental Outcomes, California State University, Chico California Invited Guest Lecture in the department of Geology and Environmental Engineering. November 4. Chicco California.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2020
Citation:
Isaya Kisekka. Sustainable Water Management in Agriculture for improved Productivity, Economic and Environmental Outcomes, FFAR Foster our Future, Washington DC Feb. 4-5th 2020.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Li, Y., D. Tian, G. Feng, and L. Feng. Does Cover Crop Mitigate Climate Extremes Impact on System Water Use Efficiency in a Maize-Soybean Cropping System? 2019 ASABE Annual International Meeting, July 7-10, Boston, Massachusetts
- Type:
Journal Articles
Status:
Published
Year Published:
2019
Citation:
Schillerberg, T., D. Tian, and R. Miao. 2019. Spatiotemporal patterns of maize and winter wheat yields in the United States: predictability and impact from climate oscillations. Agricultural and Forest Meteorology, 275 (2019): 208-222.
- Type:
Journal Articles
Status:
Accepted
Year Published:
2020
Citation:
Schillerberg, T., D. Tian. 2020. Changes of crop failure risks in the United States associated with large-scale climate oscillations in the Atlantic and Pacific Oceans. Environmental Research Letters, in press.
- Type:
Conference Papers and Presentations
Status:
Published
Year Published:
2019
Citation:
Asadi, P., D. Tian. Developing Improved Leaf Wetness Duration Models with Machine Learning and Climate Reanalysis Data. 2019 AGU Annual Meeting, Dec. 9-13, San Francisco, California
|
|