Thomas, R. Q., Boettiger, C., Carey, C. C., Dietze, M. C., Johnson, L. R., Kenney, M. A., et al. (2023). The NEON Ecological Forecasting Challenge. Frontiers in Ecology and the Environment, 21(3), 112–113. https://doi.org/10.1002/fee.2616
We thank NEON for providing the freely available data and the EFI community for feedback on the design of the Challenge. This material is based upon work supported by the National Science Foundation under Grant DEB-1926388.
Page last updated on 2024-09-16
+
Thomas, R. Q., Boettiger, C., Carey, C. C., Dietze, M. C., Johnson, L. R., Kenney, M. A., et al. (2023). The NEON Ecological Forecasting Challenge. Frontiers in Ecology and the Environment, 21(3), 112–113. https://doi.org/10.1002/fee.2616
We thank NEON for providing the freely available data and the EFI community for feedback on the design of the Challenge. This material is based upon work supported by the National Science Foundation under Grant DEB-1926388.
diff --git a/search.json b/search.json
index 4208d136c3..554e47a4cc 100644
--- a/search.json
+++ b/search.json
@@ -1,59 +1,66 @@
[
{
- "objectID": "index.html",
- "href": "index.html",
- "title": "Forecasting Challenge",
+ "objectID": "instructions.html",
+ "href": "instructions.html",
+ "title": "How to forecast",
"section": "",
- "text": "We invite you to submit forecasts!\nThe NEON Ecological Forecasting Challenge is an open platform for the ecological and data science communities to forecast data from the National Ecological Observatory Network before they are collected.\nThe Challenge is hosted by the Ecological Forecasting Initiative Research Coordination Network and sponsored by the U.S. National Science Foundation."
+ "text": "We provide an overview of the steps for submitting with the details below:\n\nExplore the data (e.g., targets) and build your forecast model.\nRegister and describe your model at https://forms.gle/kg2Vkpho9BoMXSy57. You are not required to register if your forecast submission uses the word “example” in your model_id”. Any forecasts with “example” in the model_id will not be used in forecast evaluation analyses. Use neon4cast as the challenge you are registering for.\nGenerate a forecast!\nWrite the forecast output to a file that follows our standardized format (described below).\nSubmit your forecast using an R function (provided below).\nWatch your forecast be evaluated as new data are collected."
},
{
- "objectID": "index.html#why-a-forecasting-challenge",
- "href": "index.html#why-a-forecasting-challenge",
- "title": "Forecasting Challenge",
- "section": "Why a forecasting challenge?",
- "text": "Why a forecasting challenge?\nOur vision is to use forecasts to advance theory and to support natural resource management. We can begin to realize this vision by creating and analyzing a catalog of forecasts from a range of ecological systems, spatiotemporal scales, and environmental gradients.\nOur forecasting challenge is platform for the ecological and data science communities to advance skills in forecasting ecological systems and for generating forecasts that contribute to a synthetic understanding of patterns of predictability in ecology. Rewards for contributing are skill advancement, joy, and potential involved in manuscripts. We do not currently crown winner nor offer financial awards.\nThe Challenge is an excellent focal project in university courses.\n \n\n\n\n\n\n\n\n\n\nTotal forecasts submitted to the NEON Challenge\n76369\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nMost recent data for model training\n2024-09-14\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNumber of years of data for model training\n11.21\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNumber of variables being forecasted\n10"
+ "objectID": "instructions.html#tldr-how-to-submit-a-forecast",
+ "href": "instructions.html#tldr-how-to-submit-a-forecast",
+ "title": "How to forecast",
+ "section": "",
+ "text": "We provide an overview of the steps for submitting with the details below:\n\nExplore the data (e.g., targets) and build your forecast model.\nRegister and describe your model at https://forms.gle/kg2Vkpho9BoMXSy57. You are not required to register if your forecast submission uses the word “example” in your model_id”. Any forecasts with “example” in the model_id will not be used in forecast evaluation analyses. Use neon4cast as the challenge you are registering for.\nGenerate a forecast!\nWrite the forecast output to a file that follows our standardized format (described below).\nSubmit your forecast using an R function (provided below).\nWatch your forecast be evaluated as new data are collected."
},
{
- "objectID": "index.html#the-challenge-is-a-platform",
- "href": "index.html#the-challenge-is-a-platform",
- "title": "Forecasting Challenge",
- "section": "The Challenge is a platform",
- "text": "The Challenge is a platform\nOur platform is designed to empower you to contribute by providing target data, numerical weather forecasts, and tutorials. We automatically score your forecasts using the latest NEON data. All forecasts and scores are publicly available through cloud storage and discoverable through our catalog.\n \nFigure from Thomas et al. 2023"
+ "objectID": "instructions.html#generating-a-forecast",
+ "href": "instructions.html#generating-a-forecast",
+ "title": "How to forecast",
+ "section": "2 Generating a forecast",
+ "text": "2 Generating a forecast\n\n2.1 All forecasting approaches are welcome\nWe encourage you to use any modeling approach to make a prediction about the future conditions at any of the NEON sites and variables.\n\n\n2.2 Must include uncertainty\nForecasts require you to make an assessment of the confidence in your prediction of the future. You can represent your confidence (i.e., uncertainty in the forecast) using a distribution or numerically using an ensemble (or sample) of predictions.\n\n\n2.3 Any model drivers/covariates/features are welcome\nYou can use any data as model inputs (including all of the forecast target data available to date). All sensor-based target data are available in with a 1 to 7 day delay (latency) from time of collection. You may want to use the updated target data to re-train a model or for use in data assimilation.\nAs a genuine forecasting challenge, you will need forecasted drivers if your model uses drivers as inputs. If you are interested in using forecasted meteorology, we are downloading and processing NOAA Global Ensemble Forecasting System (GEFS) weather forecasts for each NEON site. The NOAA GEFS forecasts extend 35-days ahead. More information about accessing the weather forecasts can be found here\n\n\n2.4 Forecasts can be for a range of horizons\nForecasts can be submitted for 1 day to 1 year-ahead, depending on the variable. See the variable tables for the horizon that is associated with each variable.\n\n\n2.5 Forecasts can be submitted everyday\nSince forecasts can be submitted everyday, automation is important. We provide an example GitHub repository that can be used to automate your forecast with GitHub Actions. It also includes the use of a custom Docker Container eco4cast/rocker-neon4cast:latest that has many of the packages and functions needed to generate and submit forecasts.\nWe only evaluate forecasts for any weekly variables (e.g., beetles and ticks) that were submitted on the Sunday of each week. Therefore we recommend only submitting forecasts of the weekly variables on Sundays."
},
{
- "objectID": "index.html#contact",
- "href": "index.html#contact",
- "title": "Forecasting Challenge",
- "section": "Contact",
- "text": "Contact\neco4cast.initiative@gmail.com"
+ "objectID": "instructions.html#you-can-forecast-at-any-of-the-neon-sites",
+ "href": "instructions.html#you-can-forecast-at-any-of-the-neon-sites",
+ "title": "How to forecast",
+ "section": "3 You can forecast at any of the NEON sites",
+ "text": "3 You can forecast at any of the NEON sites\nIf are you are getting started, we recommend a set of focal sites for each of the five “themes”. You are also welcome to submit forecasts to all or a subset of NEON sites . More information about NEON sites can be found in the site metadata and on NEON’s website"
},
{
- "objectID": "index.html#acknowledgements",
- "href": "index.html#acknowledgements",
- "title": "Forecasting Challenge",
- "section": "Acknowledgements",
- "text": "Acknowledgements\nThomas, R. Q., Boettiger, C., Carey, C. C., Dietze, M. C., Johnson, L. R., Kenney, M. A., et al. (2023). The NEON Ecological Forecasting Challenge. Frontiers in Ecology and the Environment, 21(3), 112–113. https://doi.org/10.1002/fee.2616 We thank NEON for providing the freely available data and the EFI community for feedback on the design of the Challenge. This material is based upon work supported by the National Science Foundation under Grant DEB-1926388. Page last updated on 2024-09-16"
+ "objectID": "instructions.html#forecast-file-format",
+ "href": "instructions.html#forecast-file-format",
+ "title": "How to forecast",
+ "section": "4 Forecast file format",
+ "text": "4 Forecast file format\nThe file is a csv format with the following columns:\n\nproject_id: use neon4cast\nmodel_id: the short name of the model defined as the model_id in your registration. The model_id should have no spaces. model_id should reflect a method to forecast one or a set of target variables and must be unique to the neon4cast challenge.\ndatetime: forecast timestamp. Format %Y-%m-%d %H:%M:%S with UTC as the time zone. Forecasts submitted with a %Y-%m-%d format will be converted to a full datetime assuming UTC mid-night.\nreference_datetime: The start of the forecast; this should be 0 times steps in the future. There should only be one value of reference_datetime in the file. Format is %Y-%m-%d %H:%M:%S with UTC as the time zone. Forecasts submitted with a %Y-%m-%d format will be converted to a full datetime assuming UTC mid-night.\nduration: the time-step of the forecast. Use the value of P1D for a daily forecast, P1W for a weekly forecast, and PT30M for 30 minute forecast. This value should match the duration of the target variable that you are forecasting. Formatted as ISO 8601 duration\nsite_id: code for NEON site.\nfamily name of the probability distribution that is described by the parameter values in the parameter column (see list below for accepted distribution). An ensemble forecast as a family of ensemble. See note below about family\nparameter the parameters for the distribution (see note below about the parameter column) or the number of the ensemble members. For example, the parameters for a normal distribution are called mu and sigma.\nvariable: standardized variable name. It must match the variable name in the target file.\nprediction: forecasted value for the parameter in the parameter column"
},
{
- "objectID": "learn-more.html",
- "href": "learn-more.html",
- "title": "Learn more",
- "section": "",
- "text": "Introductory tutorial for submitting to Challenge focused on aquatics theme: https://github.com/OlssonF/NEON-forecast-challenge-workshop. A webinar version of tutorial\nMore advanced tutorial for submitting to Challenge focused on terrestrial theme: https://github.com/rqthomas/FluxCourseForecast\nOther tutorial materials about ecological forecasting\n\n\n\n\n\nLewis, A., W. Woelmer, H. Wander, D. Howard, J. Smith, R. McClure, M. Lofton, N. Hammond, R. Corrigan, R.Q. Thomas, C.C. Carey. 2022. Increased adoption of best practices in ecological forecasting enables comparisons of forecastability across systems. Ecological Applications 32: e02500 https://doi.org/10.1002/eap.2500\nLewis, A. S. L., Rollinson, C. R., Allyn, A. J., Ashander, J., Brodie, S., Brookson, C. B., et al. (2023). The power of forecasts to advance ecological theory. Methods in Ecology and Evolution, 14(3), 746–756. https://doi.org/10.1111/2041-210X.13955\n\n\n\nThomas, R. Q., Boettiger, C., Carey, C. C., Dietze, M. C., Johnson, L. R., Kenney, M. A., et al. (2023). The NEON Ecological Forecasting Challenge. Frontiers in Ecology and the Environment, 21(3), 112–113. https://doi.org/10.1002/fee.2616\nThomas, R.Q, R.P. McClure, T.N. Moore, W.M. Woelmer, C. Boettiger, R.J. Figueiredo, R.T. Hensley, C.C. Carey. Near-term forecasts of NEON lakes reveal gradients of environmental predictability across the U.S. Frontiers in Ecology and Environment 21: 220–226. https://doi.org/10.1002/fee.2623\nWheeler, K., M. Dietze, D. LeBauer, J. Peters, A.D. Richardson, R.Q. Thomas, K. Zhu, U. Bhat, S. Munch, R.F Buzbee, M. Chen, B. Goldstein, J.S. Guo, D. Hao, C. Jones, M. Kelly-Fair, H. Liu, C. Malmborg, N. Neupane. D. Pal, A. Ross, V. Shirey, Y. Song, M. Steen, E.A. Vance, W.M. Woelmer, J. Wynne and L. Zachmann. Predicting Spring Phenology in Deciduous Broadleaf Forests: An Open Community Forecast Challenge.\n\n\n\nDietze, M., R.Q. Thomas, J. Peters, C. Boettiger, A. Shiklomanov, and J. Ashander. 2023. A community convention for ecological forecasting: output files and metadata v1.0. Ecosphere 14: e4686 https://doi.org/10.1002/ecs2.4686\n\n\n\nMoore, T.N., R.Q. Thomas, W.M. Woelmer, C.C Carey. 2022. Integrating ecological forecasting into undergraduate ecology curricula with an R Shiny application-based teaching module. Forecasting 4:604-633. https://doi.org/10.3390/forecast4030033\nPeters, J. and R.Q. Thomas. 2021. Going Virtual: What We Learned from the Ecological Forecasting Initiative Research Coordination Network Virtual Workshop. Bulletin of the Ecological Society of America 102: e01828 https://doi.org/10.1002/bes2.1828\nWillson, A.M., H. Gallo, J.A. Peters, A. Abeyta, N. Bueno Watts, C.C. Carey, T.N. Moore, G. Smies, R.Q. Thomas, W.M. Woelmer, and J.S. McLachlan. 2023. Assessing opportunities and inequities in undergraduate ecological forecasting education. Ecology and Evolution 13: e10001. https://doi.org/10.1002/ece3.10001\nWoelmer, W. M., Bradley, L. M., Haber, L. T., Klinges, D. H., Lewis, A. S. L., Mohr, E. J., et al. (2021). Ten simple rules for training yourself in an emerging field. PLOS Computational Biology, 17(10), e1009440. https://doi.org/10.1371/journal.pcbi.1009440\nWoelmer, W.M., T.N. Moore, M.E. Lofton, R.Q. Thomas, and C.C. Carey. 2023. Embedding communication concepts in forecasting training increases students’ understanding of ecological uncertainty Ecosphere 14: e4628 https://doi.org/10.1002/ecs2.4628"
+ "objectID": "instructions.html#representing-uncertainity",
+ "href": "instructions.html#representing-uncertainity",
+ "title": "How to forecast",
+ "section": "5 Representing uncertainity",
+ "text": "5 Representing uncertainity\nUncertainty is represented through the family - parameter columns in the file that you submit.\n\n5.0.1 Parameteric forecast\nFor a parametric forecast with a normal distribution, the family column would have the word normal to designate a normal distribution and the parameter column must have values of mu and sigma for each forecasted variable, site_id, depth, and time combination.\nParameteric forecasts for binary variables should use bernoulli as the family and prob as the parameter.\nThe following names and parameterization of the distribution are currently supported (family: parameters):\n\nlognormal: mu, sigma\nnormal: mu,sigma\nbernoulli: prob\nbeta: shape1, shape2\nuniform: min, max\ngamma: shape, rate\nlogistic: location, scale\nexponential: rate\npoisson: lambda\n\nIf you are submitting a forecast that is not in the supported list, we recommend using the ensemble format and sampling from your distribution to generate a set of ensemble members that represents your forecast distribution.\n\n\n5.0.2 Ensemble (or sample) forecast\nEnsemble (or sample) forecasts use the family value of ensemble and the parameter values are the ensemble index.\nWhen forecasts using the ensemble family are scored, we assume that the ensemble is a set of equally likely realizations that are sampled from a distribution that is comparable to that of the observations (i.e., no broad adjustments are required to make the ensemble more consistent with observations). This is referred to as a “perfect ensemble” by Bröcker and Smith (2007). Ensemble (or sample) forecasts are transformed to a probability distribution function (e.g., dressed) using the default methods in the scoringRules R package (empirical version of the quantile decomposition for the crps).\n\n\n5.1 Example forecasts\nHere is an example of a forecast that uses a normal distribution:\n\ndf <- readr::read_csv(\"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/forecasts/raw/T20231102190926_aquatics-2023-10-19-climatology.csv.gz\", show_col_types = FALSE)\n\n\ndf |> \n head() |> \n knitr::kable()\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nmodel_id\ndatetime\nreference_datetime\nsite_id\nfamily\nparameter\nvariable\nprediction\n\n\n\n\nclimatology\n2023-10-20\n2023-10-19\nARIK\nnormal\nmu\noxygen\n4.542862\n\n\nclimatology\n2023-10-20\n2023-10-19\nARIK\nnormal\nsigma\noxygen\n1.448393\n\n\nclimatology\n2023-10-20\n2023-10-19\nARIK\nnormal\nmu\ntemperature\n8.070854\n\n\nclimatology\n2023-10-20\n2023-10-19\nARIK\nnormal\nsigma\ntemperature\n1.330059\n\n\nclimatology\n2023-10-21\n2023-10-19\nARIK\nnormal\nmu\noxygen\n4.194895\n\n\nclimatology\n2023-10-21\n2023-10-19\nARIK\nnormal\nsigma\noxygen\n1.448393\n\n\n\n\n\nFor an ensemble (or sample) forecast, the family column uses the word ensemble to designate that it is a ensemble forecast and the parameter column is the ensemble member number (1, 2, 3 …)\n\ndf <- readr::read_csv(\"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/forecasts/raw/T20231102190926_aquatics-2023-10-19-persistenceRW.csv.gz\", show_col_types = FALSE)\n\n\ndf |> \n dplyr::arrange(variable, site_id, datetime, parameter) |> \n head() |> \n knitr::kable()\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nmodel_id\ndatetime\nreference_datetime\nsite_id\nfamily\nparameter\nvariable\nprediction\n\n\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n1\nchla\n3.795652\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n2\nchla\n3.963322\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n3\nchla\n2.053637\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n4\nchla\n3.294723\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n5\nchla\n1.847344\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n6\nchla\n7.829229"
},
{
- "objectID": "learn-more.html#tutorials",
- "href": "learn-more.html#tutorials",
- "title": "Learn more",
- "section": "",
- "text": "Introductory tutorial for submitting to Challenge focused on aquatics theme: https://github.com/OlssonF/NEON-forecast-challenge-workshop. A webinar version of tutorial\nMore advanced tutorial for submitting to Challenge focused on terrestrial theme: https://github.com/rqthomas/FluxCourseForecast\nOther tutorial materials about ecological forecasting"
+ "objectID": "instructions.html#submission-process",
+ "href": "instructions.html#submission-process",
+ "title": "How to forecast",
+ "section": "6 Submission process",
+ "text": "6 Submission process\n\n6.1 File name\nSave your forecast as a csv file with the following naming convention:\ntheme_name-year-month-day-model_id.csv. Compressed csv files with the csv.gz extension are also accepted.\nThe theme_name options are: terrestrial_daily, terrestrial_30min, aquatics, beetles, ticks, or phenology.\nThe year, month, and day are the year, month, and day the reference_datetime (horizon = 0). For example, if a forecast starts today and tomorrow is the first forecasted day, horizon = 0 would be today, and used in the file name. model_id is the id for the model name that you specified in the model metadata Google Form (model_id has no spaces in it).\n\n\n6.2 Uploading forecast\nIndividual forecast files can be uploaded any time.\nTeams will submit their forecast csv files through an R function. The csv file can only contain one unique model_id and one unique project_id.\nThe function is available using the following code\n\nremotes::install_github(\"eco4cast/neon4cast\")\n\nThe submit function is\n\nlibrary(neon4cast)\nneon4cast::submit(forecast_file = \"theme_name-year-month-day-model_id.csv\")"
},
{
- "objectID": "learn-more.html#research-from-the-ecological-forecasting-initiative-research-coordination-network.",
- "href": "learn-more.html#research-from-the-ecological-forecasting-initiative-research-coordination-network.",
- "title": "Learn more",
- "section": "",
- "text": "Lewis, A., W. Woelmer, H. Wander, D. Howard, J. Smith, R. McClure, M. Lofton, N. Hammond, R. Corrigan, R.Q. Thomas, C.C. Carey. 2022. Increased adoption of best practices in ecological forecasting enables comparisons of forecastability across systems. Ecological Applications 32: e02500 https://doi.org/10.1002/eap.2500\nLewis, A. S. L., Rollinson, C. R., Allyn, A. J., Ashander, J., Brodie, S., Brookson, C. B., et al. (2023). The power of forecasts to advance ecological theory. Methods in Ecology and Evolution, 14(3), 746–756. https://doi.org/10.1111/2041-210X.13955\n\n\n\nThomas, R. Q., Boettiger, C., Carey, C. C., Dietze, M. C., Johnson, L. R., Kenney, M. A., et al. (2023). The NEON Ecological Forecasting Challenge. Frontiers in Ecology and the Environment, 21(3), 112–113. https://doi.org/10.1002/fee.2616\nThomas, R.Q, R.P. McClure, T.N. Moore, W.M. Woelmer, C. Boettiger, R.J. Figueiredo, R.T. Hensley, C.C. Carey. Near-term forecasts of NEON lakes reveal gradients of environmental predictability across the U.S. Frontiers in Ecology and Environment 21: 220–226. https://doi.org/10.1002/fee.2623\nWheeler, K., M. Dietze, D. LeBauer, J. Peters, A.D. Richardson, R.Q. Thomas, K. Zhu, U. Bhat, S. Munch, R.F Buzbee, M. Chen, B. Goldstein, J.S. Guo, D. Hao, C. Jones, M. Kelly-Fair, H. Liu, C. Malmborg, N. Neupane. D. Pal, A. Ross, V. Shirey, Y. Song, M. Steen, E.A. Vance, W.M. Woelmer, J. Wynne and L. Zachmann. Predicting Spring Phenology in Deciduous Broadleaf Forests: An Open Community Forecast Challenge.\n\n\n\nDietze, M., R.Q. Thomas, J. Peters, C. Boettiger, A. Shiklomanov, and J. Ashander. 2023. A community convention for ecological forecasting: output files and metadata v1.0. Ecosphere 14: e4686 https://doi.org/10.1002/ecs2.4686\n\n\n\nMoore, T.N., R.Q. Thomas, W.M. Woelmer, C.C Carey. 2022. Integrating ecological forecasting into undergraduate ecology curricula with an R Shiny application-based teaching module. Forecasting 4:604-633. https://doi.org/10.3390/forecast4030033\nPeters, J. and R.Q. Thomas. 2021. Going Virtual: What We Learned from the Ecological Forecasting Initiative Research Coordination Network Virtual Workshop. Bulletin of the Ecological Society of America 102: e01828 https://doi.org/10.1002/bes2.1828\nWillson, A.M., H. Gallo, J.A. Peters, A. Abeyta, N. Bueno Watts, C.C. Carey, T.N. Moore, G. Smies, R.Q. Thomas, W.M. Woelmer, and J.S. McLachlan. 2023. Assessing opportunities and inequities in undergraduate ecological forecasting education. Ecology and Evolution 13: e10001. https://doi.org/10.1002/ece3.10001\nWoelmer, W. M., Bradley, L. M., Haber, L. T., Klinges, D. H., Lewis, A. S. L., Mohr, E. J., et al. (2021). Ten simple rules for training yourself in an emerging field. PLOS Computational Biology, 17(10), e1009440. https://doi.org/10.1371/journal.pcbi.1009440\nWoelmer, W.M., T.N. Moore, M.E. Lofton, R.Q. Thomas, and C.C. Carey. 2023. Embedding communication concepts in forecasting training increases students’ understanding of ecological uncertainty Ecosphere 14: e4628 https://doi.org/10.1002/ecs2.4628"
+ "objectID": "instructions.html#post-submission",
+ "href": "instructions.html#post-submission",
+ "title": "How to forecast",
+ "section": "7 Post-submission",
+ "text": "7 Post-submission\n\n7.1 Processing\nAfter submission, our servers will process uploaded files by converting them to a parquet format on our public s3 storage. A pub_datetime column will be added that denotes when a forecast was submitted. Summaries are generated of the forecasts provide descriptive statistics of the forecast.\n\n\n7.2 Evaluation\nAll forecasts are scored daily using new data until the full horizon of the forecast has been scored. Forecasts are scored using the crps function in the scoringRules R package. More information about the scoring metric can be found at here\n\n\n7.3 Comparison\nForecast performance can be compared to the performance of baseline models. We are automatically submitting the following baseline models:\n\nclimatology: the normal distribution (mean and standard deviation) of that day-of-year in the historical observations\npersistenceRW: a random walk model that assumes no change in the mean behavior. The random walk is initialized using the most resent observation.\nmean: the historical mean of the data is submitted for the beetles theme.\n\nOur forecast performance page includes evaluations of all submitted models.\n\n\n7.4 Catalog\nInformation and code for accessing the forecasts and scores can be found on our forecast catalog page."
+ },
+ {
+ "objectID": "instructions.html#questions",
+ "href": "instructions.html#questions",
+ "title": "How to forecast",
+ "section": "8 Questions?",
+ "text": "8 Questions?\nThanks for reading this document!\n\nIf you still have questions about how to submit your forecast to the NEON Ecological Forecasting Challenge, we encourage you to email Dr. Quinn Thomas (rqthomas{at}vt.edu)."
},
{
"objectID": "targets.html",
@@ -81,7 +88,7 @@
"href": "targets.html#sec-targets",
"title": "What to forecast",
"section": "Explore the targets and themes",
- "text": "Explore the targets and themes\nInformation on the targets files for the five “themes” is below. In the tables,\n\n“duration” is the time-step of the variable where PT30M is a 30-minute mean, P1D is a daily mean, and P1W is a weekly total.\nThe “forecast horizon” is the number of days-ahead that we want you to forecast.\nThe “latency” is the time between data collection and data availability in the targets file\n\n\nTerrestrial fluxesAquaticsPhenologyBeetle communitiesTick populations\n\n\n\nThe exchange of water and carbon dioxide between the atmosphere and the land is akin to earth’s terrestrial ecosystems breathing rate and lung capacity. \nThe terrestrial flux theme challenges you to forecast the gas exchange at up to 47 sites across the U.S.\nThere are two variables and two time-steps (or duration) that you can forecast.\n\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nforecast horizon\nLatency\n\n\n\n\nle\nP1D\ndaily mean latent heat flux (W/m2)\n30 days\n~ 5 days\n\n\nnee\nP1D\ndaily mean Net ecosystem exchange (gC/m2/day)\n30 days\n~ 5 days\n\n\nle\nPT30M\n30 minute mean latent heat flux (W/m2)\n10 days\n~ 5 days\n\n\nnee\nPT30M\n30 minute mean net ecosystem exchange (umol/m2/s)\n10 days\n~ 5 days\n\n\n\n\n\n\nDaily mean\nThe daily mean target file is located at the following URL.\n\nurl_P1D <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1D/terrestrial_daily-targets.csv.gz\"\n\nYou can directly load it into R using the following\n\nterrestrial_targets <- read_csv(url_P1D, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nBART\n2017-02-06\nP1D\nle\n11.5794551\n\n\nneon4cast\nBART\n2017-02-07\nP1D\nle\n4.8951620\n\n\nneon4cast\nBART\n2017-02-09\nP1D\nle\n7.5281656\n\n\nneon4cast\nBART\n2017-02-11\nP1D\nle\n1.1577581\n\n\nneon4cast\nBART\n2017-02-12\nP1D\nle\n0.1999174\n\n\nneon4cast\nBART\n2017-02-13\nP1D\nle\n10.9325370\n\n\n\n\n\nand the time series for the focal sites\n\nterrestrial_targets |> \n filter(site_id %in% terrestrial_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\n\n\n\n\n\n30 minute\nThe 30 minute duration targets are designed for forecasting sub-daily carbon and water dynamics. The URL is found at:\n\nurl_PT30M <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=PT30M/terrestrial_30min-targets.csv.gz\"\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Terrestrial.html\n\n\n\n\nFreshwater surface water temperature, dissolved oxygen, and chlorophyll-a all influence drinking water quality, are critical for life in aquatic environments, and can represent the health of the ecosystem.\nThe aquatics theme challenges you to forecast daily mean water quality variables at up-to 7 lakes and 27 river/stream NEON sites.\n\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nhorizon\nLatency\n\n\n\n\nchla\nP1D\ndaily mean Chlorophyll-a (ug/L)\n30 days\n~ 3 days\n\n\noxygen\nP1D\nSurface Mean Daily Dissolved Oxygen Concentration (mgL)\n30 days\n~ 3 days\n\n\ntemperature\nP1D\nSurface Mean Daily Water Temperature (Celsius)\n30 days\n~ 3 days\n\n\n\n\n\nThe daily mean target file is located at the following URL.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1D/aquatics-targets.csv.gz\"\n\nYou can directly load it into R using the following\n\naquatics_targets <- read_csv(url, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nARIK\n2016-08-12\nP1D\noxygen\n3.402153\n\n\nneon4cast\nARIK\n2016-08-13\nP1D\noxygen\n4.156236\n\n\nneon4cast\nARIK\n2016-08-14\nP1D\noxygen\n4.071263\n\n\nneon4cast\nARIK\n2016-08-15\nP1D\noxygen\n3.909114\n\n\nneon4cast\nARIK\n2016-08-16\nP1D\noxygen\n3.862653\n\n\nneon4cast\nARIK\n2016-08-17\nP1D\noxygen\n4.354618\n\n\n\n\n\nand the time series for the focal sites\n\naquatics_targets |> \n filter(site_id %in% aquatics_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\nWarning: Removed 827 rows containing missing values or values outside the scale range\n(`geom_point()`).\n\n\n\n\n\nWater temperature at multiple depths measured at the UTC 00 hour are available for the 7 NEON lake sites. These data can be used for model development but will not be used for forecast evaluation.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/supporting_data/project_id=neon4cast/aquatics-expanded-observations.csv.gz\"\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Aquatics.html\n\n\n\nPhenology (the changes in plant canopies over the year) has been identified as one of the primary ecological fingerprints of global climate change.\nThe greenness and redness, as measured by a camera looking down at the top of vegetation are a quantitative measure of phenology. The phenology theme challenges you to forecast daily mean greeness and/or redness at up-to 47 terrestrial NEON sites.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1D/phenology-targets.csv.gz\"\nread_csv(url, show_col_types = FALSE) |> \n distinct(variable, duration) |> \n left_join(target_metadata, by = c(\"variable\",\"duration\")) |> \n filter(variable %in% c(\"gcc_90\",\"rcc_90\")) |> \n select(-class) |> \n knitr::kable()\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nhorizon\nLatency\n\n\n\n\ngcc_90\nP1D\nGreen chromatic coordinate is the ratio of the green digital number to the sum of the red, green, blue digital numbers from a digital camera.\n30 days\n~ 2 days\n\n\nrcc_90\nP1D\nRed chromatic coordinate is the ratio of the Red digital number to the sum of the red, green, blue digital numbers from a digital camera.\n30 days\n~ 2 days\n\n\n\n\n\nThe daily mean target file is located at the following URL.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1D/phenology-targets.csv.gz\"\n\nYou can directly load it into R using the following\n\nphenology_targets <- read_csv(url, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nABBY\n2017-05-30\nP1D\ngcc_90\n0.41659\n\n\nneon4cast\nABBY\n2017-05-31\nP1D\ngcc_90\n0.41570\n\n\nneon4cast\nABBY\n2017-06-01\nP1D\ngcc_90\n0.41780\n\n\nneon4cast\nABBY\n2017-06-02\nP1D\ngcc_90\n0.41539\n\n\nneon4cast\nABBY\n2017-06-03\nP1D\ngcc_90\n0.42216\n\n\nneon4cast\nABBY\n2017-06-04\nP1D\ngcc_90\n0.41659\n\n\n\n\n\nand the time series for the focal sites\n\nphenology_targets |> \n filter(site_id %in% phenology_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\nWarning: Removed 8600 rows containing missing values or values outside the scale range\n(`geom_point()`).\n\n\n\n\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Phenology.html\n\n\n\nSentinel species (such as beetles) can give forewarning of environmental risk to humans, so are particularly useful for such monitoring and forecasting efforts because they can provide surrogates for other co-located components of biodiversity.\nThe beetles theme challenges you to forecast weekly ground beetles (Family: Carabidae) abundance and richness (two measures of biodiversity) at up-to 47 terrestrial NEON sites.\n\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nhorizon\nLatency\n\n\n\n\nabundance\nP1W\nTotal number of carabid individuals per trap-night, estimated each week of the year at each NEON site\n1 year\n~ 6 months\n\n\nrichness\nP1W\nTotal number of unique ‘species’ in a sampling bout for each NEON site each week.\n1 year\n~ 6 months\n\n\n\n\n\nThe daily mean target file is located at the following URL.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1W/beetles-targets.csv.gz\"\n\nYou can directly load it into R using the following\n\nbeetles_targets <- read_csv(url, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nABBY\n2016-09-12\nP1W\nabundance\n1.0489796\n\n\nneon4cast\nABBY\n2016-09-12\nP1W\nrichness\n14.0000000\n\n\nneon4cast\nABBY\n2016-09-26\nP1W\nabundance\n4.4535714\n\n\nneon4cast\nABBY\n2016-09-26\nP1W\nrichness\n13.0000000\n\n\nneon4cast\nABBY\n2017-05-01\nP1W\nabundance\n0.0553571\n\n\nneon4cast\nABBY\n2017-05-01\nP1W\nrichness\n10.0000000\n\n\n\n\n\nand the time series for the focal sites\n\nbeetles_targets |> \n filter(site_id %in% beetles_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\n\n\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Beetles.html\n\n\n\nTarget species for the tick population forecasts are Amblyomma americanum nymphal ticks. A. americanum is a vector of ehrlichiosis, tularemia, and southern tick-associated rash illness. The species is present in the eastern United States, and their populations are expanding. There is a correlation between tick population abundance and disease incidence, meaning forecasts for tick abundance have the potential to aid in our understanding of disease risk through time and space.\nThe beetles theme challenges you to forecast weekly Amblyomma americanum nymphal tick abundance at up-to 9 terrestrial NEON sites.\n\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nhorizon\nLatency\n\n\n\n\namblyomma_americanum\nP1W\nThe density of Amblyomma americanum nymphs per week (ticks per 1600m^2)\n1 year\n~ 6 months\n\n\n\n\n\nThe weekly target file is located at the following URL.\n\n\"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1W/ticks-targets.csv.gz\"\n\n[1] \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1W/ticks-targets.csv.gz\"\n\n\nYou can directly load it into R using the following\n\nticks_targets <- read_csv(url, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\n\n\n\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nBLAN\n2015-04-20\nP1W\namblyomma_americanum\n0.000000\n\n\nneon4cast\nBLAN\n2015-05-11\nP1W\namblyomma_americanum\n9.815951\n\n\nneon4cast\nBLAN\n2015-06-01\nP1W\namblyomma_americanum\n10.000000\n\n\nneon4cast\nBLAN\n2015-06-08\nP1W\namblyomma_americanum\n19.393939\n\n\nneon4cast\nBLAN\n2015-06-22\nP1W\namblyomma_americanum\n3.137255\n\n\nneon4cast\nBLAN\n2015-07-13\nP1W\namblyomma_americanum\n3.661327\n\n\n\n\n\nand the time series for the focal sites\n\nticks_targets |> \n filter(site_id %in% ticks_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\n\n\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Ticks.html"
+ "text": "Explore the targets and themes\nInformation on the targets files for the five “themes” is below. In the tables,\n\n“duration” is the time-step of the variable where PT30M is a 30-minute mean, P1D is a daily mean, and P1W is a weekly total.\nThe “forecast horizon” is the number of days-ahead that we want you to forecast.\nThe “latency” is the time between data collection and data availability in the targets file\n\n\nTerrestrial fluxesAquaticsPhenologyBeetle communitiesTick populations\n\n\n\nThe exchange of water and carbon dioxide between the atmosphere and the land is akin to earth’s terrestrial ecosystems breathing rate and lung capacity. \nThe terrestrial flux theme challenges you to forecast the gas exchange at up to 47 sites across the U.S.\nThere are two variables and two time-steps (or duration) that you can forecast.\n\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nforecast horizon\nLatency\n\n\n\n\nle\nP1D\ndaily mean latent heat flux (W/m2)\n30 days\n~ 5 days\n\n\nnee\nP1D\ndaily mean Net ecosystem exchange (gC/m2/day)\n30 days\n~ 5 days\n\n\nle\nPT30M\n30 minute mean latent heat flux (W/m2)\n10 days\n~ 5 days\n\n\nnee\nPT30M\n30 minute mean net ecosystem exchange (umol/m2/s)\n10 days\n~ 5 days\n\n\n\n\n\n\nDaily mean\nThe daily mean target file is located at the following URL.\n\nurl_P1D <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1D/terrestrial_daily-targets.csv.gz\"\n\nYou can directly load it into R using the following\n\nterrestrial_targets <- read_csv(url_P1D, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nBART\n2017-02-06\nP1D\nle\n11.5794551\n\n\nneon4cast\nBART\n2017-02-07\nP1D\nle\n4.8951620\n\n\nneon4cast\nBART\n2017-02-09\nP1D\nle\n7.5281656\n\n\nneon4cast\nBART\n2017-02-11\nP1D\nle\n1.1577581\n\n\nneon4cast\nBART\n2017-02-12\nP1D\nle\n0.1999174\n\n\nneon4cast\nBART\n2017-02-13\nP1D\nle\n10.9325370\n\n\n\n\n\nand the time series for the focal sites\n\nterrestrial_targets |> \n filter(site_id %in% terrestrial_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\n\n\n\n\n\n30 minute\nThe 30 minute duration targets are designed for forecasting sub-daily carbon and water dynamics. The URL is found at:\n\nurl_PT30M <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=PT30M/terrestrial_30min-targets.csv.gz\"\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Terrestrial.html\n\n\n\n\nFreshwater surface water temperature, dissolved oxygen, and chlorophyll-a all influence drinking water quality, are critical for life in aquatic environments, and can represent the health of the ecosystem.\nThe aquatics theme challenges you to forecast daily mean water quality variables at up-to 7 lakes and 27 river/stream NEON sites.\n\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nhorizon\nLatency\n\n\n\n\nchla\nP1D\ndaily mean Chlorophyll-a (ug/L)\n30 days\n~ 3 days\n\n\noxygen\nP1D\nSurface Mean Daily Dissolved Oxygen Concentration (mgL)\n30 days\n~ 3 days\n\n\ntemperature\nP1D\nSurface Mean Daily Water Temperature (Celsius)\n30 days\n~ 3 days\n\n\n\n\n\nThe daily mean target file is located at the following URL.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1D/aquatics-targets.csv.gz\"\n\nYou can directly load it into R using the following\n\naquatics_targets <- read_csv(url, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nARIK\n2016-08-12\nP1D\noxygen\n3.402153\n\n\nneon4cast\nARIK\n2016-08-13\nP1D\noxygen\n4.156236\n\n\nneon4cast\nARIK\n2016-08-14\nP1D\noxygen\n4.071263\n\n\nneon4cast\nARIK\n2016-08-15\nP1D\noxygen\n3.909114\n\n\nneon4cast\nARIK\n2016-08-16\nP1D\noxygen\n3.862653\n\n\nneon4cast\nARIK\n2016-08-17\nP1D\noxygen\n4.354618\n\n\n\n\n\nand the time series for the focal sites\n\naquatics_targets |> \n filter(site_id %in% aquatics_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\nWarning: Removed 827 rows containing missing values or values outside the scale range\n(`geom_point()`).\n\n\n\n\n\nWater temperature at multiple depths measured at the UTC 00 hour are available for the 7 NEON lake sites. These data can be used for model development but will not be used for forecast evaluation.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/supporting_data/project_id=neon4cast/aquatics-expanded-observations.csv.gz\"\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Aquatics.html\n\n\n\nPhenology (the changes in plant canopies over the year) has been identified as one of the primary ecological fingerprints of global climate change.\nThe greenness and redness, as measured by a camera looking down at the top of vegetation are a quantitative measure of phenology. The phenology theme challenges you to forecast daily mean greeness and/or redness at up-to 47 terrestrial NEON sites.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1D/phenology-targets.csv.gz\"\nread_csv(url, show_col_types = FALSE) |> \n distinct(variable, duration) |> \n left_join(target_metadata, by = c(\"variable\",\"duration\")) |> \n filter(variable %in% c(\"gcc_90\",\"rcc_90\")) |> \n select(-class) |> \n knitr::kable()\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nhorizon\nLatency\n\n\n\n\ngcc_90\nP1D\nGreen chromatic coordinate is the ratio of the green digital number to the sum of the red, green, blue digital numbers from a digital camera.\n30 days\n~ 2 days\n\n\nrcc_90\nP1D\nRed chromatic coordinate is the ratio of the Red digital number to the sum of the red, green, blue digital numbers from a digital camera.\n30 days\n~ 2 days\n\n\n\n\n\nThe daily mean target file is located at the following URL.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1D/phenology-targets.csv.gz\"\n\nYou can directly load it into R using the following\n\nphenology_targets <- read_csv(url, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nABBY\n2017-05-30\nP1D\ngcc_90\n0.41659\n\n\nneon4cast\nABBY\n2017-05-31\nP1D\ngcc_90\n0.41570\n\n\nneon4cast\nABBY\n2017-06-01\nP1D\ngcc_90\n0.41780\n\n\nneon4cast\nABBY\n2017-06-02\nP1D\ngcc_90\n0.41539\n\n\nneon4cast\nABBY\n2017-06-03\nP1D\ngcc_90\n0.42216\n\n\nneon4cast\nABBY\n2017-06-04\nP1D\ngcc_90\n0.41659\n\n\n\n\n\nand the time series for the focal sites\n\nphenology_targets |> \n filter(site_id %in% phenology_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\nWarning: Removed 8602 rows containing missing values or values outside the scale range\n(`geom_point()`).\n\n\n\n\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Phenology.html\n\n\n\nSentinel species (such as beetles) can give forewarning of environmental risk to humans, so are particularly useful for such monitoring and forecasting efforts because they can provide surrogates for other co-located components of biodiversity.\nThe beetles theme challenges you to forecast weekly ground beetles (Family: Carabidae) abundance and richness (two measures of biodiversity) at up-to 47 terrestrial NEON sites.\n\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nhorizon\nLatency\n\n\n\n\nabundance\nP1W\nTotal number of carabid individuals per trap-night, estimated each week of the year at each NEON site\n1 year\n~ 6 months\n\n\nrichness\nP1W\nTotal number of unique ‘species’ in a sampling bout for each NEON site each week.\n1 year\n~ 6 months\n\n\n\n\n\nThe daily mean target file is located at the following URL.\n\nurl <- \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1W/beetles-targets.csv.gz\"\n\nYou can directly load it into R using the following\n\nbeetles_targets <- read_csv(url, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nABBY\n2016-09-12\nP1W\nabundance\n1.0489796\n\n\nneon4cast\nABBY\n2016-09-12\nP1W\nrichness\n14.0000000\n\n\nneon4cast\nABBY\n2016-09-26\nP1W\nabundance\n4.4535714\n\n\nneon4cast\nABBY\n2016-09-26\nP1W\nrichness\n13.0000000\n\n\nneon4cast\nABBY\n2017-05-01\nP1W\nabundance\n0.0553571\n\n\nneon4cast\nABBY\n2017-05-01\nP1W\nrichness\n10.0000000\n\n\n\n\n\nand the time series for the focal sites\n\nbeetles_targets |> \n filter(site_id %in% beetles_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\n\n\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Beetles.html\n\n\n\nTarget species for the tick population forecasts are Amblyomma americanum nymphal ticks. A. americanum is a vector of ehrlichiosis, tularemia, and southern tick-associated rash illness. The species is present in the eastern United States, and their populations are expanding. There is a correlation between tick population abundance and disease incidence, meaning forecasts for tick abundance have the potential to aid in our understanding of disease risk through time and space.\nThe beetles theme challenges you to forecast weekly Amblyomma americanum nymphal tick abundance at up-to 9 terrestrial NEON sites.\n\n\n\n\n\n\n\n\n\n\n\n\nvariable\nduration\nDescription\nhorizon\nLatency\n\n\n\n\namblyomma_americanum\nP1W\nThe density of Amblyomma americanum nymphs per week (ticks per 1600m^2)\n1 year\n~ 6 months\n\n\n\n\n\nThe weekly target file is located at the following URL.\n\n\"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1W/ticks-targets.csv.gz\"\n\n[1] \"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/targets/project_id=neon4cast/duration=P1W/ticks-targets.csv.gz\"\n\n\nYou can directly load it into R using the following\n\nticks_targets <- read_csv(url, show_col_types = FALSE)\n\nThe file contains the following columns\n\n\n\n\n\n\n\n\n\n\n\n\n\nproject_id\nsite_id\ndatetime\nduration\nvariable\nobservation\n\n\n\n\nneon4cast\nBLAN\n2015-04-20\nP1W\namblyomma_americanum\n0.000000\n\n\nneon4cast\nBLAN\n2015-05-11\nP1W\namblyomma_americanum\n9.815951\n\n\nneon4cast\nBLAN\n2015-06-01\nP1W\namblyomma_americanum\n10.000000\n\n\nneon4cast\nBLAN\n2015-06-08\nP1W\namblyomma_americanum\n19.393939\n\n\nneon4cast\nBLAN\n2015-06-22\nP1W\namblyomma_americanum\n3.137255\n\n\nneon4cast\nBLAN\n2015-07-13\nP1W\namblyomma_americanum\n3.661327\n\n\n\n\n\nand the time series for the focal sites\n\nticks_targets |> \n filter(site_id %in% ticks_focal_sites) |> \n ggplot(aes(x = datetime, y = observation)) +\n geom_point() +\n facet_grid(variable~site_id, scales = \"free_y\") +\n theme_bw()\n\n\n\n\nLearn more at: https://projects.ecoforecast.org/neon4cast-docs/Ticks.html"
},
{
"objectID": "targets.html#explore-the-sites",
@@ -102,14 +109,14 @@
"href": "performance.html#sec-performance",
"title": "Forecast performance",
"section": "Most recent forecasts",
- "text": "Most recent forecasts\nOnly the top-performing models from the last 30 days are shown.\nForecasts submitted on 2024-09-13\n\nTerrestrial: net ecosystem exchangeTerrestrial: latent heat fluxPhenology: greennessPhenology: rednessAquatics: water temperatureAquatics: dissolved oxygenAquatics: chlorophyll-aBeetle community richnessBeetle community abundanceTicks: Amblyomma americanum\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here"
+ "text": "Most recent forecasts\nOnly the top-performing models from the last 30 days are shown.\nForecasts submitted on 2024-09-14\n\nTerrestrial: net ecosystem exchangeTerrestrial: latent heat fluxPhenology: greennessPhenology: rednessAquatics: water temperatureAquatics: dissolved oxygenAquatics: chlorophyll-aBeetle community richnessBeetle community abundanceTicks: Amblyomma americanum\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here\n\n\n\n\n\n\n\n\nForecast summaries are available here"
},
{
"objectID": "performance.html#forecast-analysis",
"href": "performance.html#forecast-analysis",
"title": "Forecast performance",
"section": "Forecast analysis",
- "text": "Forecast analysis\nBelow are forecasts submitted 30 days ago and include the observations used to evaluate them. Mouse over to see the team id, scroll to zoom. Only the top five performing models are shown. Information on how to access the scores can be found in our catalog\n\nTerrestrial: net ecosystem exchangeTerrestrial: latent heat fluxPhenology: greennessPhenology: rednessAquatics: water temperatureAquatics: dissolved oxygenAquatics: chlorophyll-aBeetle community richnessBeetle community abundanceTicks: Amblyomma americanum\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNULL\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNULL"
+ "text": "Forecast analysis\nBelow are forecasts submitted 30 days ago and include the observations used to evaluate them. Mouse over to see the team id, scroll to zoom. Only the top five performing models are shown. Information on how to access the scores can be found in our catalog\n\nTerrestrial: net ecosystem exchangeTerrestrial: latent heat fluxPhenology: greennessPhenology: rednessAquatics: water temperatureAquatics: dissolved oxygenAquatics: chlorophyll-aBeetle community richnessBeetle community abundanceTicks: Amblyomma americanum"
},
{
"objectID": "performance.html#aggregated-scores",
@@ -140,66 +147,59 @@
"text": "Catalog of forecast submissions and evaluations\nThe catalog of submitted forecasts and the evaluation of the forecasts (“scores”) is available through the SpatioTemporal Asset Catalogs browser (below). \nThe catalog provides the code that you can use to access forecasts and scores. \nA full page version can be found here"
},
{
- "objectID": "instructions.html",
- "href": "instructions.html",
- "title": "How to forecast",
- "section": "",
- "text": "We provide an overview of the steps for submitting with the details below:\n\nExplore the data (e.g., targets) and build your forecast model.\nRegister and describe your model at https://forms.gle/kg2Vkpho9BoMXSy57. You are not required to register if your forecast submission uses the word “example” in your model_id”. Any forecasts with “example” in the model_id will not be used in forecast evaluation analyses. Use neon4cast as the challenge you are registering for.\nGenerate a forecast!\nWrite the forecast output to a file that follows our standardized format (described below).\nSubmit your forecast using an R function (provided below).\nWatch your forecast be evaluated as new data are collected."
- },
- {
- "objectID": "instructions.html#tldr-how-to-submit-a-forecast",
- "href": "instructions.html#tldr-how-to-submit-a-forecast",
- "title": "How to forecast",
+ "objectID": "index.html",
+ "href": "index.html",
+ "title": "Forecasting Challenge",
"section": "",
- "text": "We provide an overview of the steps for submitting with the details below:\n\nExplore the data (e.g., targets) and build your forecast model.\nRegister and describe your model at https://forms.gle/kg2Vkpho9BoMXSy57. You are not required to register if your forecast submission uses the word “example” in your model_id”. Any forecasts with “example” in the model_id will not be used in forecast evaluation analyses. Use neon4cast as the challenge you are registering for.\nGenerate a forecast!\nWrite the forecast output to a file that follows our standardized format (described below).\nSubmit your forecast using an R function (provided below).\nWatch your forecast be evaluated as new data are collected."
+ "text": "We invite you to submit forecasts!\nThe NEON Ecological Forecasting Challenge is an open platform for the ecological and data science communities to forecast data from the National Ecological Observatory Network before they are collected.\nThe Challenge is hosted by the Ecological Forecasting Initiative Research Coordination Network and sponsored by the U.S. National Science Foundation."
},
{
- "objectID": "instructions.html#generating-a-forecast",
- "href": "instructions.html#generating-a-forecast",
- "title": "How to forecast",
- "section": "2 Generating a forecast",
- "text": "2 Generating a forecast\n\n2.1 All forecasting approaches are welcome\nWe encourage you to use any modeling approach to make a prediction about the future conditions at any of the NEON sites and variables.\n\n\n2.2 Must include uncertainty\nForecasts require you to make an assessment of the confidence in your prediction of the future. You can represent your confidence (i.e., uncertainty in the forecast) using a distribution or numerically using an ensemble (or sample) of predictions.\n\n\n2.3 Any model drivers/covariates/features are welcome\nYou can use any data as model inputs (including all of the forecast target data available to date). All sensor-based target data are available in with a 1 to 7 day delay (latency) from time of collection. You may want to use the updated target data to re-train a model or for use in data assimilation.\nAs a genuine forecasting challenge, you will need forecasted drivers if your model uses drivers as inputs. If you are interested in using forecasted meteorology, we are downloading and processing NOAA Global Ensemble Forecasting System (GEFS) weather forecasts for each NEON site. The NOAA GEFS forecasts extend 35-days ahead. More information about accessing the weather forecasts can be found here\n\n\n2.4 Forecasts can be for a range of horizons\nForecasts can be submitted for 1 day to 1 year-ahead, depending on the variable. See the variable tables for the horizon that is associated with each variable.\n\n\n2.5 Forecasts can be submitted everyday\nSince forecasts can be submitted everyday, automation is important. We provide an example GitHub repository that can be used to automate your forecast with GitHub Actions. It also includes the use of a custom Docker Container eco4cast/rocker-neon4cast:latest that has many of the packages and functions needed to generate and submit forecasts.\nWe only evaluate forecasts for any weekly variables (e.g., beetles and ticks) that were submitted on the Sunday of each week. Therefore we recommend only submitting forecasts of the weekly variables on Sundays."
+ "objectID": "index.html#why-a-forecasting-challenge",
+ "href": "index.html#why-a-forecasting-challenge",
+ "title": "Forecasting Challenge",
+ "section": "Why a forecasting challenge?",
+ "text": "Why a forecasting challenge?\nOur vision is to use forecasts to advance theory and to support natural resource management. We can begin to realize this vision by creating and analyzing a catalog of forecasts from a range of ecological systems, spatiotemporal scales, and environmental gradients.\nOur forecasting challenge is platform for the ecological and data science communities to advance skills in forecasting ecological systems and for generating forecasts that contribute to a synthetic understanding of patterns of predictability in ecology. Rewards for contributing are skill advancement, joy, and potential involved in manuscripts. We do not currently crown winner nor offer financial awards.\nThe Challenge is an excellent focal project in university courses.\n \n\n\n\n\n\n\n\n\n\nTotal forecasts submitted to the NEON Challenge\n76393\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nMost recent data for model training\n2024-09-15\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNumber of years of data for model training\n11.22\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nNumber of variables being forecasted\n10"
},
{
- "objectID": "instructions.html#you-can-forecast-at-any-of-the-neon-sites",
- "href": "instructions.html#you-can-forecast-at-any-of-the-neon-sites",
- "title": "How to forecast",
- "section": "3 You can forecast at any of the NEON sites",
- "text": "3 You can forecast at any of the NEON sites\nIf are you are getting started, we recommend a set of focal sites for each of the five “themes”. You are also welcome to submit forecasts to all or a subset of NEON sites . More information about NEON sites can be found in the site metadata and on NEON’s website"
+ "objectID": "index.html#the-challenge-is-a-platform",
+ "href": "index.html#the-challenge-is-a-platform",
+ "title": "Forecasting Challenge",
+ "section": "The Challenge is a platform",
+ "text": "The Challenge is a platform\nOur platform is designed to empower you to contribute by providing target data, numerical weather forecasts, and tutorials. We automatically score your forecasts using the latest NEON data. All forecasts and scores are publicly available through cloud storage and discoverable through our catalog.\n \nFigure from Thomas et al. 2023"
},
{
- "objectID": "instructions.html#forecast-file-format",
- "href": "instructions.html#forecast-file-format",
- "title": "How to forecast",
- "section": "4 Forecast file format",
- "text": "4 Forecast file format\nThe file is a csv format with the following columns:\n\nproject_id: use neon4cast\nmodel_id: the short name of the model defined as the model_id in your registration. The model_id should have no spaces. model_id should reflect a method to forecast one or a set of target variables and must be unique to the neon4cast challenge.\ndatetime: forecast timestamp. Format %Y-%m-%d %H:%M:%S with UTC as the time zone. Forecasts submitted with a %Y-%m-%d format will be converted to a full datetime assuming UTC mid-night.\nreference_datetime: The start of the forecast; this should be 0 times steps in the future. There should only be one value of reference_datetime in the file. Format is %Y-%m-%d %H:%M:%S with UTC as the time zone. Forecasts submitted with a %Y-%m-%d format will be converted to a full datetime assuming UTC mid-night.\nduration: the time-step of the forecast. Use the value of P1D for a daily forecast, P1W for a weekly forecast, and PT30M for 30 minute forecast. This value should match the duration of the target variable that you are forecasting. Formatted as ISO 8601 duration\nsite_id: code for NEON site.\nfamily name of the probability distribution that is described by the parameter values in the parameter column (see list below for accepted distribution). An ensemble forecast as a family of ensemble. See note below about family\nparameter the parameters for the distribution (see note below about the parameter column) or the number of the ensemble members. For example, the parameters for a normal distribution are called mu and sigma.\nvariable: standardized variable name. It must match the variable name in the target file.\nprediction: forecasted value for the parameter in the parameter column"
+ "objectID": "index.html#contact",
+ "href": "index.html#contact",
+ "title": "Forecasting Challenge",
+ "section": "Contact",
+ "text": "Contact\neco4cast.initiative@gmail.com"
},
{
- "objectID": "instructions.html#representing-uncertainity",
- "href": "instructions.html#representing-uncertainity",
- "title": "How to forecast",
- "section": "5 Representing uncertainity",
- "text": "5 Representing uncertainity\nUncertainty is represented through the family - parameter columns in the file that you submit.\n\n5.0.1 Parameteric forecast\nFor a parametric forecast with a normal distribution, the family column would have the word normal to designate a normal distribution and the parameter column must have values of mu and sigma for each forecasted variable, site_id, depth, and time combination.\nParameteric forecasts for binary variables should use bernoulli as the family and prob as the parameter.\nThe following names and parameterization of the distribution are currently supported (family: parameters):\n\nlognormal: mu, sigma\nnormal: mu,sigma\nbernoulli: prob\nbeta: shape1, shape2\nuniform: min, max\ngamma: shape, rate\nlogistic: location, scale\nexponential: rate\npoisson: lambda\n\nIf you are submitting a forecast that is not in the supported list, we recommend using the ensemble format and sampling from your distribution to generate a set of ensemble members that represents your forecast distribution.\n\n\n5.0.2 Ensemble (or sample) forecast\nEnsemble (or sample) forecasts use the family value of ensemble and the parameter values are the ensemble index.\nWhen forecasts using the ensemble family are scored, we assume that the ensemble is a set of equally likely realizations that are sampled from a distribution that is comparable to that of the observations (i.e., no broad adjustments are required to make the ensemble more consistent with observations). This is referred to as a “perfect ensemble” by Bröcker and Smith (2007). Ensemble (or sample) forecasts are transformed to a probability distribution function (e.g., dressed) using the default methods in the scoringRules R package (empirical version of the quantile decomposition for the crps).\n\n\n5.1 Example forecasts\nHere is an example of a forecast that uses a normal distribution:\n\ndf <- readr::read_csv(\"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/forecasts/raw/T20231102190926_aquatics-2023-10-19-climatology.csv.gz\", show_col_types = FALSE)\n\n\ndf |> \n head() |> \n knitr::kable()\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nmodel_id\ndatetime\nreference_datetime\nsite_id\nfamily\nparameter\nvariable\nprediction\n\n\n\n\nclimatology\n2023-10-20\n2023-10-19\nARIK\nnormal\nmu\noxygen\n4.542862\n\n\nclimatology\n2023-10-20\n2023-10-19\nARIK\nnormal\nsigma\noxygen\n1.448393\n\n\nclimatology\n2023-10-20\n2023-10-19\nARIK\nnormal\nmu\ntemperature\n8.070854\n\n\nclimatology\n2023-10-20\n2023-10-19\nARIK\nnormal\nsigma\ntemperature\n1.330059\n\n\nclimatology\n2023-10-21\n2023-10-19\nARIK\nnormal\nmu\noxygen\n4.194895\n\n\nclimatology\n2023-10-21\n2023-10-19\nARIK\nnormal\nsigma\noxygen\n1.448393\n\n\n\n\n\nFor an ensemble (or sample) forecast, the family column uses the word ensemble to designate that it is a ensemble forecast and the parameter column is the ensemble member number (1, 2, 3 …)\n\ndf <- readr::read_csv(\"https://sdsc.osn.xsede.org/bio230014-bucket01/challenges/forecasts/raw/T20231102190926_aquatics-2023-10-19-persistenceRW.csv.gz\", show_col_types = FALSE)\n\n\ndf |> \n dplyr::arrange(variable, site_id, datetime, parameter) |> \n head() |> \n knitr::kable()\n\n\n\n\n\n\n\n\n\n\n\n\n\n\nmodel_id\ndatetime\nreference_datetime\nsite_id\nfamily\nparameter\nvariable\nprediction\n\n\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n1\nchla\n3.795652\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n2\nchla\n3.963322\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n3\nchla\n2.053637\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n4\nchla\n3.294723\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n5\nchla\n1.847344\n\n\npersistenceRW\n2023-10-20\n2023-10-19\nBARC\nensemble\n6\nchla\n7.829229"
+ "objectID": "index.html#acknowledgements",
+ "href": "index.html#acknowledgements",
+ "title": "Forecasting Challenge",
+ "section": "Acknowledgements",
+ "text": "Acknowledgements\nThomas, R. Q., Boettiger, C., Carey, C. C., Dietze, M. C., Johnson, L. R., Kenney, M. A., et al. (2023). The NEON Ecological Forecasting Challenge. Frontiers in Ecology and the Environment, 21(3), 112–113. https://doi.org/10.1002/fee.2616 We thank NEON for providing the freely available data and the EFI community for feedback on the design of the Challenge. This material is based upon work supported by the National Science Foundation under Grant DEB-1926388. Page last updated on 2024-09-17"
},
{
- "objectID": "instructions.html#submission-process",
- "href": "instructions.html#submission-process",
- "title": "How to forecast",
- "section": "6 Submission process",
- "text": "6 Submission process\n\n6.1 File name\nSave your forecast as a csv file with the following naming convention:\ntheme_name-year-month-day-model_id.csv. Compressed csv files with the csv.gz extension are also accepted.\nThe theme_name options are: terrestrial_daily, terrestrial_30min, aquatics, beetles, ticks, or phenology.\nThe year, month, and day are the year, month, and day the reference_datetime (horizon = 0). For example, if a forecast starts today and tomorrow is the first forecasted day, horizon = 0 would be today, and used in the file name. model_id is the id for the model name that you specified in the model metadata Google Form (model_id has no spaces in it).\n\n\n6.2 Uploading forecast\nIndividual forecast files can be uploaded any time.\nTeams will submit their forecast csv files through an R function. The csv file can only contain one unique model_id and one unique project_id.\nThe function is available using the following code\n\nremotes::install_github(\"eco4cast/neon4cast\")\n\nThe submit function is\n\nlibrary(neon4cast)\nneon4cast::submit(forecast_file = \"theme_name-year-month-day-model_id.csv\")"
+ "objectID": "learn-more.html",
+ "href": "learn-more.html",
+ "title": "Learn more",
+ "section": "",
+ "text": "Introductory tutorial for submitting to Challenge focused on aquatics theme: https://github.com/OlssonF/NEON-forecast-challenge-workshop. A webinar version of tutorial\nMore advanced tutorial for submitting to Challenge focused on terrestrial theme: https://github.com/rqthomas/FluxCourseForecast\nOther tutorial materials about ecological forecasting\n\n\n\n\n\nLewis, A., W. Woelmer, H. Wander, D. Howard, J. Smith, R. McClure, M. Lofton, N. Hammond, R. Corrigan, R.Q. Thomas, C.C. Carey. 2022. Increased adoption of best practices in ecological forecasting enables comparisons of forecastability across systems. Ecological Applications 32: e02500 https://doi.org/10.1002/eap.2500\nLewis, A. S. L., Rollinson, C. R., Allyn, A. J., Ashander, J., Brodie, S., Brookson, C. B., et al. (2023). The power of forecasts to advance ecological theory. Methods in Ecology and Evolution, 14(3), 746–756. https://doi.org/10.1111/2041-210X.13955\n\n\n\nThomas, R. Q., Boettiger, C., Carey, C. C., Dietze, M. C., Johnson, L. R., Kenney, M. A., et al. (2023). The NEON Ecological Forecasting Challenge. Frontiers in Ecology and the Environment, 21(3), 112–113. https://doi.org/10.1002/fee.2616\nThomas, R.Q, R.P. McClure, T.N. Moore, W.M. Woelmer, C. Boettiger, R.J. Figueiredo, R.T. Hensley, C.C. Carey. Near-term forecasts of NEON lakes reveal gradients of environmental predictability across the U.S. Frontiers in Ecology and Environment 21: 220–226. https://doi.org/10.1002/fee.2623\nWheeler, K., M. Dietze, D. LeBauer, J. Peters, A.D. Richardson, R.Q. Thomas, K. Zhu, U. Bhat, S. Munch, R.F Buzbee, M. Chen, B. Goldstein, J.S. Guo, D. Hao, C. Jones, M. Kelly-Fair, H. Liu, C. Malmborg, N. Neupane. D. Pal, A. Ross, V. Shirey, Y. Song, M. Steen, E.A. Vance, W.M. Woelmer, J. Wynne and L. Zachmann. Predicting Spring Phenology in Deciduous Broadleaf Forests: An Open Community Forecast Challenge.\n\n\n\nDietze, M., R.Q. Thomas, J. Peters, C. Boettiger, A. Shiklomanov, and J. Ashander. 2023. A community convention for ecological forecasting: output files and metadata v1.0. Ecosphere 14: e4686 https://doi.org/10.1002/ecs2.4686\n\n\n\nMoore, T.N., R.Q. Thomas, W.M. Woelmer, C.C Carey. 2022. Integrating ecological forecasting into undergraduate ecology curricula with an R Shiny application-based teaching module. Forecasting 4:604-633. https://doi.org/10.3390/forecast4030033\nPeters, J. and R.Q. Thomas. 2021. Going Virtual: What We Learned from the Ecological Forecasting Initiative Research Coordination Network Virtual Workshop. Bulletin of the Ecological Society of America 102: e01828 https://doi.org/10.1002/bes2.1828\nWillson, A.M., H. Gallo, J.A. Peters, A. Abeyta, N. Bueno Watts, C.C. Carey, T.N. Moore, G. Smies, R.Q. Thomas, W.M. Woelmer, and J.S. McLachlan. 2023. Assessing opportunities and inequities in undergraduate ecological forecasting education. Ecology and Evolution 13: e10001. https://doi.org/10.1002/ece3.10001\nWoelmer, W. M., Bradley, L. M., Haber, L. T., Klinges, D. H., Lewis, A. S. L., Mohr, E. J., et al. (2021). Ten simple rules for training yourself in an emerging field. PLOS Computational Biology, 17(10), e1009440. https://doi.org/10.1371/journal.pcbi.1009440\nWoelmer, W.M., T.N. Moore, M.E. Lofton, R.Q. Thomas, and C.C. Carey. 2023. Embedding communication concepts in forecasting training increases students’ understanding of ecological uncertainty Ecosphere 14: e4628 https://doi.org/10.1002/ecs2.4628"
},
{
- "objectID": "instructions.html#post-submission",
- "href": "instructions.html#post-submission",
- "title": "How to forecast",
- "section": "7 Post-submission",
- "text": "7 Post-submission\n\n7.1 Processing\nAfter submission, our servers will process uploaded files by converting them to a parquet format on our public s3 storage. A pub_datetime column will be added that denotes when a forecast was submitted. Summaries are generated of the forecasts provide descriptive statistics of the forecast.\n\n\n7.2 Evaluation\nAll forecasts are scored daily using new data until the full horizon of the forecast has been scored. Forecasts are scored using the crps function in the scoringRules R package. More information about the scoring metric can be found at here\n\n\n7.3 Comparison\nForecast performance can be compared to the performance of baseline models. We are automatically submitting the following baseline models:\n\nclimatology: the normal distribution (mean and standard deviation) of that day-of-year in the historical observations\npersistenceRW: a random walk model that assumes no change in the mean behavior. The random walk is initialized using the most resent observation.\nmean: the historical mean of the data is submitted for the beetles theme.\n\nOur forecast performance page includes evaluations of all submitted models.\n\n\n7.4 Catalog\nInformation and code for accessing the forecasts and scores can be found on our forecast catalog page."
+ "objectID": "learn-more.html#tutorials",
+ "href": "learn-more.html#tutorials",
+ "title": "Learn more",
+ "section": "",
+ "text": "Introductory tutorial for submitting to Challenge focused on aquatics theme: https://github.com/OlssonF/NEON-forecast-challenge-workshop. A webinar version of tutorial\nMore advanced tutorial for submitting to Challenge focused on terrestrial theme: https://github.com/rqthomas/FluxCourseForecast\nOther tutorial materials about ecological forecasting"
},
{
- "objectID": "instructions.html#questions",
- "href": "instructions.html#questions",
- "title": "How to forecast",
- "section": "8 Questions?",
- "text": "8 Questions?\nThanks for reading this document!\n\nIf you still have questions about how to submit your forecast to the NEON Ecological Forecasting Challenge, we encourage you to email Dr. Quinn Thomas (rqthomas{at}vt.edu)."
+ "objectID": "learn-more.html#research-from-the-ecological-forecasting-initiative-research-coordination-network.",
+ "href": "learn-more.html#research-from-the-ecological-forecasting-initiative-research-coordination-network.",
+ "title": "Learn more",
+ "section": "",
+ "text": "Lewis, A., W. Woelmer, H. Wander, D. Howard, J. Smith, R. McClure, M. Lofton, N. Hammond, R. Corrigan, R.Q. Thomas, C.C. Carey. 2022. Increased adoption of best practices in ecological forecasting enables comparisons of forecastability across systems. Ecological Applications 32: e02500 https://doi.org/10.1002/eap.2500\nLewis, A. S. L., Rollinson, C. R., Allyn, A. J., Ashander, J., Brodie, S., Brookson, C. B., et al. (2023). The power of forecasts to advance ecological theory. Methods in Ecology and Evolution, 14(3), 746–756. https://doi.org/10.1111/2041-210X.13955\n\n\n\nThomas, R. Q., Boettiger, C., Carey, C. C., Dietze, M. C., Johnson, L. R., Kenney, M. A., et al. (2023). The NEON Ecological Forecasting Challenge. Frontiers in Ecology and the Environment, 21(3), 112–113. https://doi.org/10.1002/fee.2616\nThomas, R.Q, R.P. McClure, T.N. Moore, W.M. Woelmer, C. Boettiger, R.J. Figueiredo, R.T. Hensley, C.C. Carey. Near-term forecasts of NEON lakes reveal gradients of environmental predictability across the U.S. Frontiers in Ecology and Environment 21: 220–226. https://doi.org/10.1002/fee.2623\nWheeler, K., M. Dietze, D. LeBauer, J. Peters, A.D. Richardson, R.Q. Thomas, K. Zhu, U. Bhat, S. Munch, R.F Buzbee, M. Chen, B. Goldstein, J.S. Guo, D. Hao, C. Jones, M. Kelly-Fair, H. Liu, C. Malmborg, N. Neupane. D. Pal, A. Ross, V. Shirey, Y. Song, M. Steen, E.A. Vance, W.M. Woelmer, J. Wynne and L. Zachmann. Predicting Spring Phenology in Deciduous Broadleaf Forests: An Open Community Forecast Challenge.\n\n\n\nDietze, M., R.Q. Thomas, J. Peters, C. Boettiger, A. Shiklomanov, and J. Ashander. 2023. A community convention for ecological forecasting: output files and metadata v1.0. Ecosphere 14: e4686 https://doi.org/10.1002/ecs2.4686\n\n\n\nMoore, T.N., R.Q. Thomas, W.M. Woelmer, C.C Carey. 2022. Integrating ecological forecasting into undergraduate ecology curricula with an R Shiny application-based teaching module. Forecasting 4:604-633. https://doi.org/10.3390/forecast4030033\nPeters, J. and R.Q. Thomas. 2021. Going Virtual: What We Learned from the Ecological Forecasting Initiative Research Coordination Network Virtual Workshop. Bulletin of the Ecological Society of America 102: e01828 https://doi.org/10.1002/bes2.1828\nWillson, A.M., H. Gallo, J.A. Peters, A. Abeyta, N. Bueno Watts, C.C. Carey, T.N. Moore, G. Smies, R.Q. Thomas, W.M. Woelmer, and J.S. McLachlan. 2023. Assessing opportunities and inequities in undergraduate ecological forecasting education. Ecology and Evolution 13: e10001. https://doi.org/10.1002/ece3.10001\nWoelmer, W. M., Bradley, L. M., Haber, L. T., Klinges, D. H., Lewis, A. S. L., Mohr, E. J., et al. (2021). Ten simple rules for training yourself in an emerging field. PLOS Computational Biology, 17(10), e1009440. https://doi.org/10.1371/journal.pcbi.1009440\nWoelmer, W.M., T.N. Moore, M.E. Lofton, R.Q. Thomas, and C.C. Carey. 2023. Embedding communication concepts in forecasting training increases students’ understanding of ecological uncertainty Ecosphere 14: e4628 https://doi.org/10.1002/ecs2.4628"
}
]
\ No newline at end of file
diff --git a/sitemap.xml b/sitemap.xml
index cd7c8e626f..77602a8364 100644
--- a/sitemap.xml
+++ b/sitemap.xml
@@ -2,23 +2,23 @@
https://projects.ecoforecast.org/neon4cast-ci/catalog.html
- 2024-09-15
+ 2024-09-16https://projects.ecoforecast.org/neon4cast-ci/targets.html
- 2024-09-15
+ 2024-09-16https://projects.ecoforecast.org/neon4cast-ci/instructions.html
- 2024-09-15
+ 2024-09-16https://projects.ecoforecast.org/neon4cast-ci/performance.html
- 2024-09-15
+ 2024-09-16https://projects.ecoforecast.org/neon4cast-ci/index.html
- 2024-09-15
+ 2024-09-16https://radiantearth.github.io/stac-browser/#/external/raw.githubusercontent.com/eco4cast/neon4cast-ci/main/catalog/catalog.json
diff --git a/targets.html b/targets.html
index 5e16b39c59..53ebae50df 100644
--- a/targets.html
+++ b/targets.html
@@ -690,7 +690,7 @@
Warning: Removed 8600 rows containing missing values or values outside the scale range
+
Warning: Removed 8602 rows containing missing values or values outside the scale range
(`geom_point()`).
@@ -972,8 +972,8 @@
30 minute
Explore the sites
-
-
+
+
The following table lists all the sites in the NEON Ecological Forecasting Challenge. The columns with “theme” names incidate whether that site is included in that theme’s target file.
diff --git a/targets_files/figure-html/unnamed-chunk-21-1.png b/targets_files/figure-html/unnamed-chunk-21-1.png
index 6fe4bbf015..ee1edf7907 100644
Binary files a/targets_files/figure-html/unnamed-chunk-21-1.png and b/targets_files/figure-html/unnamed-chunk-21-1.png differ
diff --git a/targets_files/figure-html/unnamed-chunk-26-1.png b/targets_files/figure-html/unnamed-chunk-26-1.png
index a712ef6650..8d17e31a4a 100644
Binary files a/targets_files/figure-html/unnamed-chunk-26-1.png and b/targets_files/figure-html/unnamed-chunk-26-1.png differ
diff --git a/targets_files/figure-html/unnamed-chunk-31-1.png b/targets_files/figure-html/unnamed-chunk-31-1.png
index ad495d13be..9fc8d19a59 100644
Binary files a/targets_files/figure-html/unnamed-chunk-31-1.png and b/targets_files/figure-html/unnamed-chunk-31-1.png differ
diff --git a/targets_files/figure-html/unnamed-chunk-9-1.png b/targets_files/figure-html/unnamed-chunk-9-1.png
index 13c8df0f6a..8feea4480c 100644
Binary files a/targets_files/figure-html/unnamed-chunk-9-1.png and b/targets_files/figure-html/unnamed-chunk-9-1.png differ