------------------- GENERAL INFORMATION ------------------- 1. Title: R Code, Data, and Output Supporting: Nutrient Data from U.S. Manure Systems 2. Author Information: a) *Nancy Bohl Bormann, University of Minnesota, Soil, Water, and Climate Department. https://orcid.org/0009-0004-7214-991X b) Melissa Wilson, University of Minnesota, Soil, Water, and Climate Department. c) Erin Cortus, University of Minnesota, Bioproducts and Biosystems Engineering Department. d) Kevin Silverstein, University of Minnesota, Research Computing. e) Kevin Janni, University of Minnesota, Bioproducts and Biosystems Engineering Department. (retired) f) Larry Gunderson, Minnesota Department of Agriculture. *Corresponding author: nancy.bohl.bormann@gmail.com 3. Description: This repository contains R code, processed data, and associated outputs supporting the results reported in: Bohl Bormann, 2024. Manure Nutrient Data Compilation and Analysis for Agronomic and Environmental Applications. PhD Dissertation. -------------------------- Version History -------------------------- 06/XX/2024 -- First upload to DRUM. -------------------------- SHARING/ACCESS INFORMATION -------------------------- 1. Licenses/restrictions placed on the data and code: This work is licensed under a Creative Commons Attribution 4.0 International License. This data contains the manure and other organic amendment sample analyses shared by participating laboratories and partners in the ManureDB project. These results are made available to all interested parties for the purpose of aggregating manure analyses where data can be utilized in a standardized way. Neither participating laboratories nor the University of Minnesota are responsible for any errors or omissions, or for the results obtained from the use of the laboratory results or the metadata. All information in this Website is provided “as is,” with no guarantee of completeness, accuracy, timeliness or of the results obtained from the use of this information. In no event will the University or the participating laboratories be liable to you or anyone else for any decision made or action taken in reliance on the information in this Site or for any direct, consequential, special, or similar damages, including lost profit or loss of business opportunity. 2. Links to publications that cite or use the data: [Forthcoming] 3. Links to other publicly accessible locations of the data: [Forthcoming] 4. Recommended citation for this archive: Bohl Bormann, N.L., Wilson, M.L., Cortus, E.L, Silverstein, K., Janni, K., and Gunderson, L. (2024) R Code, Data, and Output Supporting: Nutrient Data from U.S. Manure Systems. Data Repository of the University of Minnesota. --------------------- FILE OVERVIEW --------------------- Please consult the files listed here for details. 1. 2024-06_ManureDB_Stats_NBB_Dissertation.qmd - This Quarto script runs the statistics, output tables, and data visualizations. 2. 2024-06_ManureDB_Stats_NBB_Dissertation.html - This is the rendered html file from the 2024-06_ManureDB_Stats_NBB_Dissertation.qmd Quarto file. 3. ManureDB-Archive-DevelopmentSnapshot-2024-02-19.csv - This CSV file includes manure and organic amendment nutrient data and sample metadata from ManureDB as of 2024-02-19. Number of variables: 58 Number of cases/rows: 497279 4. ManureDB_datadownload_header_explainer_2024-06-26.txt - This txt file explains the data variables listed in ManureDB-Archive-DevelopmentSnapshot-2024-02-19.csv -------------------------- METHODOLOGICAL INFORMATION -------------------------- 1. Methods: (taken from Bohl Bormann, 2024. Manure Nutrient Data Compilation and Analysis for Agronomic and Environmental Applications. PhD Dissertation.) The February 2024 ManureDB dataset was used for this summary, which included 489,034 samples with 55 animal or other amendment types, from 15 laboratories resulting in 5.2 million data points from sample metadata and results. ManureDB’s database display function rule required at least five samples of an animal type per year from a location (state, region, or nation) to be included in the dataset (i.e., data not meeting this requirement was not included in the dataset). ManureDB had analyte units and as-received (wet) or dry basis standardized in the summary using equations from the Recommended Methods of Manure Analysis second edition (Wilson et al., 2022). For each analyte, different analytical methods may have been used and were treated as equivalent (i.e., total Kjeldahl N and combustion methods for determining total N were considered to be equivalent, so both are reported as “total N”). To categorize samples by moisture content, ManureDB applied moisture designations determined by the MidWest Plan Service definitions with samples of 0-4% Total Solids="liquid", >4-10% Total Solids="slurry", >10-20% Total Solids="semi-solid", >20% Total Solids="solid" (Lorimor et al., 2004). If a sample had no Total Solids or Moisture level, ManureDB applied the laboratory’s Manure Type designation of liquid, solid, slurry, or semi-solid if available. If the sample had a Manure Type selection of litter, ManureDB assumed its moisture designation was a solid. If a sample had no Total Solids, Moisture, or Manure Type, ManureDB left this category blank. For this summary, liquid manure includes the moisture designations of “liquid” and “slurry,” and solid manure includes the moisture designations of “solid” and “semi-solid.” Animal or other amendment type was determined from common livestock and poultry species and then more specific life stages or genders were separated out if it could be determined from the laboratory submittal sheets. Data Selection We focused on livestock and poultry manure because they are the highest value U.S. animal commodities (USDA-APHIS, 2016), and ManureDB included thousands of samples in those categories. The eight “animal manure categories” analyzed were beef solid, chicken-broiler solid, chicken – layer solid, dairy solid, turkey solid, beef liquid, dairy liquid, and swine liquid manure. The beef, dairy, and swine categories include all life stages. Because of varying degrees of sample descriptions across all the sources, no differentiation was shown in this paper between manure storage systems, bedding, agitation, treatment, or storage length. While the database contains data from 1998-2023, the period from 2012-2022 was selected for its relevance to current conditions and density of data. Since manure, particularly storage and handling, is affected by climate, we grouped the data according to USDA Climate Hub regions (USDA, 2020): Midwest (MW), Northeast (NE), Northern Plains (NP), Pacific Northwest (PN), Southeast (SE), Southern Plains (SP), and Southwest (SW) (Table 2.1). Total N, NH4-N, P2O5, and K2O, were selected for data analysis, as they are the nutrients of concern for crop producers utilizing manure. Manure nutrients were reported on an as-received (wet) basis, as those values reflect how manure was handled and how application rates are calculated. Median, median absolute deviation (MAD), median coefficient of variation (MCV), 25th percentile, 75th percentile, and count for total N, NH4-N, P2O5, K2O for each of the eight animal manure categories by region and overall were calculated. The base R package was utilized for these analyses (R Core Team, 2023). For comparison to the previously published book values, we compared ManureDB analyte medians to the similar species manure type for the ASABE and MWPS book values. Often MWPS and ASABE had several values for a species to account for different life stages or manure storages. In those cases, the range of the highest and lowest analyte values for a species was compared to the ManureDB median and a percent difference was calculated by subtracting the ManureDB median from the closest book value number divided by the closest book value number, then multiplied by 100. Analysis by Region and Animal Type We compared medians and MCVs across animal manure categories, along with median comparisons between regions for animal manure categories. With eight animal manure categories and four analytes, 32 separate datasets. For the regional comparisons, we used a cutoff of 500 samples per dataset within the 2012-2022 timeframe, targeting approximately 40-50 samples per year across regions. Regions without that sample quantity were shown in the summary tables for each animal manure category. We observed positive right-skewed data in 31 of 32 datasets. The ManureDB means were larger than the medians 85% of the time, indicating some larger outliers pulling up those means. Multiple data transformations were attempted, and the data could not be normalized through those transformations. The Jarque-Bera test for normality confirmed the data sets (each analytes for a specific animal manure category) were not normally distributed by using jarque.bera.test function in the ‘tseries’ package (Jarque & Bera, 1987; Trapletti et al., 2023). The wilcox_test function was used to complete the Mann-Whitney U test along with the p.adjust function applied to the p-value for the Benjamini-Hochberg (false discovery rate) correction with both functions from the R package ‘coin’ (Benjamini & Hochberg, 1995; Hothorn et al., 2023; Mann & Whitney, 1947). Analyte Trends over Time Separate animal manure category datasets for each region with at least 500 samples including each year 2012-2022 were analyzed for trends across that time period. The SE, MW, and NE regions had respectively seven, six, and five animal manure categories that met those parameters. The non-parametric Mann-Kendall trend test was used for those regions with sufficient samples for each of the eight animal manure categories using the ‘MannKendall’ function in R (Kendall, 1955; Mann, 1945; McLeod, 2022). A test statistic and 2-sided p-values were calculated for each animal manure category, region, and analyte combination indicating no significant trend, a significant increasing trend, or a significant decreasing trend. Line graphs for medians and interquartile ranges (IQR) of each analyte per year and region for each animal manure category with sufficient samples were made utilizing the ‘ggplot2’ package (Wickham et al., 2024). 2. Instrument- or software-specific information needed to interpret the data: Programs were written for Program R (R Core Team 2023) version 4.3.2, and all packages used are noted in each of the R files included towards the top of the script file. Versions of relevant packages are included in the citations. During curation, all code ran without error with the following package versions: attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] Kendall_2.2.1 patchwork_1.2.0 extrafont_0.19 tseries_0.10-56 [5] MuMIn_1.48.4 lme4_1.1-35.3 Matrix_1.7-0 tinytex_0.51 [9] RColorBrewer_1.1-3 plotly_4.10.4 psych_2.4.3 gt_0.10.1 [13] writexl_1.5.0 coin_1.4-3 survival_3.6-4 car_3.1-2 [17] carData_3.0-5 lubridate_1.9.3 forcats_1.0.0 stringr_1.5.1 [21] dplyr_1.1.4 purrr_1.0.2 readr_2.1.5 tidyr_1.3.1 [25] tibble_3.2.1 ggplot2_3.5.1 tidyverse_2.0.0 clipr_0.8.0 loaded via a namespace (and not attached): [1] tidyselect_1.2.1 viridisLite_0.4.2 farver_2.1.1 libcoin_1.0-10 [5] fastmap_1.2.0 lazyeval_0.2.2 TH.data_1.1-2 digest_0.6.35 [9] timechange_0.3.0 lifecycle_1.0.4 magrittr_2.0.3 compiler_4.4.1 [13] sass_0.4.9 rlang_1.1.4 tools_4.4.1 utf8_1.2.4 [17] yaml_2.3.8 data.table_1.15.4 knitr_1.47 labeling_0.4.3 [21] htmlwidgets_1.6.4 curl_5.2.1 mnormt_2.1.1 TTR_0.24.4 [25] xml2_1.3.6 multcomp_1.4-25 abind_1.4-5 withr_3.0.0 [29] grid_4.4.1 stats4_4.4.1 fansi_1.0.6 xts_0.14.0 [33] colorspace_2.1-0 extrafontdb_1.0 scales_1.3.0 MASS_7.3-60.2 [37] cli_3.6.2 mvtnorm_1.2-4 rmarkdown_2.27 generics_0.1.3 [41] rstudioapi_0.16.0 httr_1.4.7 tzdb_0.4.0 minqa_1.2.6 [45] modeltools_0.2-23 splines_4.4.1 parallel_4.4.1 matrixStats_1.3.0 [49] vctrs_0.6.5 boot_1.3-30 sandwich_3.1-0 jsonlite_1.8.8 [53] hms_1.1.3 quantmod_0.4.26 glue_1.7.0 nloptr_2.0.3 [57] codetools_0.2-20 stringi_1.8.4 gtable_0.3.5 quadprog_1.5-8 [61] munsell_0.5.1 pillar_1.9.0 htmltools_0.5.8.1 R6_2.5.1 [65] evaluate_0.23 lattice_0.22-6 Rcpp_1.0.12 Rttf2pt1_1.3.12 [69] nlme_3.1-164 xfun_0.44 zoo_1.8-12 pkgconfig_2.0.3 ------------------------------------------------------------------------------ Variable information for: ManureDB-Archive-DevelopmentSnapshot-2024-02-19.csv ------------------------------------------------------------------------------ Rows: 497,278 Columns: 57 Each column header is listed below with any accompanying explanations and/or options. METADATA 1. ManureDB Sample ID -Unique sample identifier randomly assigned to each sample by database 2. Animal or Other Amendment Type [required] -Animal or other amendment type as determined from the lab sumittal sheet -Options: Alpaca, Bat, Beef, Beef - Breeding, Beef - Calf, Beef - Replacement Heifers, Beef - Finisher, Beef - Stocker, Biosolids, Bison, Cattle, Chicken, Chicken - Breeder, Chicken - Broiler, Chicken - Layer, Chicken - Pullet, Cricket, Dairy, Dairy - Calf, Dairy - Calf and Heifer, Dairy - Heifers, Dairy - Lactating Cow, Dairy - Dry Cow, Dairy - Steer, Deer, Dog, Donkey, Duck, Egret, Elephant, Elk, Emu, Ferret, Fish, Geese, Goat, Guinea Pig, Horse, Llama, Mink, Mouse, Partridge, Pigeon, Poultry, Poultry - Mortality, Rabbit, Rat, Quail, Sheep, Swine - Farrow to Wean, Swine - Farrow to Feeder, Swine - Farrow to Finish, Swine - Feeder to Finish, Swine - Finisher, Swine - Gestation, Swine - Wean to Finish, Swine - Nursery, Swine - Boar Stud, Swine - Mortality, Swine, Turkey, Turkey - Meat, Turkey - Hen, Veal, Worm, Other - Brewery Waste, Other - Biochar, Other - Cardboard Waste, Other - Egg Shells, Other - Egg Wash Water, Other - Fish Pond Sludge, Other - Feed Leachate, Other - Food Waste, Other - Grape Pomace, Other - Milkhouse Rinse Water, Other - Mushroom, Other - Paper Waste, Other - Paunch, Other - Sewage Sludge, Other - Spoiled Feed, Other - Whey, Other - Truck Wash, Other - Yard Waste, Mixed, Other Animal, Other Organic Amendment 3. Animal Type Combined Category [required] -Common animal groupings based on Animal or Other Amendment Type label -Options: Beef, Dairy, Other Animal, Other Organic Amendment, Poultry, Swine 4. Year Analyzed [required] -The year a sample was analyzed. 5. State -State where the sample came from or was shipped from. 6. Region -U.S. divided into region based on State label -Options: a. Midwest - MI, OH, WI, MN, IA, MO, IN and IL b. Northeast - ME, VT, NH, CT, MA, NY, PA, RI, DE, MD, NJ, WV c. Southeast - VA, KY, TN, AR, LA, MS, AL, GA, FL, SC, NC, DC, PR d. Southern Plains - TX, OK, KS e. Northern Plains - ND, SD, NE, CO, WY, MT f. Southwest - CA, NV, UT, AZ, NM, HI g. Pacific Northwest - AK, WA, OR, ID 7. Manure Type -Determined off of lab submittal sheet -Options: Liquid, Slurry, Semi-solid, Solid, Separated Liquid, Separated Solid, Digestate, Digested Solid, Runoff, Sludge, As-excreted (urine), As-excreted (feces), Litter, Compost, Unknown 8. Manure Treatment -Determined off of lab submittal sheet -Options: Dried, Aerobic, Alum, KLASP, Phytase, PLT, Poultry Guard, Other Treatment, No Treatment, Unknown 9. Agitated -Determined off of lab submittal sheet -Options: Yes, No, Unknown 10. Bedding Type -Determined off of lab submittal sheet -Options: Hardwood Sawdust, Hardwood Shavings, Paper, Peanut Hulls, Pine Sawdust, Pine Shavings, Rice Hulls, Sand, Straw, Other/Unspecified Bedding, No Bedding, Unknown 11. Storage Type -Options were based on MWPS-18 Section 2 categories -Determined off of lab submittal sheet -Options: Lagoon, Uncovered Pit or Tank, Covered Pit or Tank, Earthen Basin, Runoff Holding Pond, Underfloor Solid Storage, Stockpiled Under Cover, Stockpiled Outdoors, Cage-free Poultry, Unknown 12. Length of Storage -Determined off of lab submittal sheet -Options: Daily haul, 0 to <3 months, 3 to <6 months, 6 to <12 months, 12+ months, Unknown 13. Application Method -Determined off of lab submittal sheet -Options: Irrigation, Broadcast, Broadcast - Incorporated within 24 hours, Broadcast - Incorporated after 24 hours, Injection, Unknown 14. Moisture Designation -The moisture designation assigned based off a sample’s total solids or moisture content using the Midwest Plan Service definition if 0-4% total solids="liquid", >4-10% total solids="slurry", >10-20% total solids="semi-solid", >20% total solids="solid" (Lorimor et al., 2004). If a sample has no total solids or moisture level, use lab-designated Manure Type label to assign the manure designation categories of liquid, solid, slurry, semi-solid, or litter as a solid. If a sample has no total solids, moisture, or manure type, this category was left blank. -Options: Liquid, Slurry, Semi-solid, Solid ANALYTES Each item below is listed by: a. Analyte name b."Combined" means all analytical methods for this particular analyte are combined for this column c. Units d. Basis 15. Aluminum|Combined|%|As Received (Wet) Basis 16. Ammonium-N|Combined|%|As Received (Wet) Basis 17. Arsenic|Combined|%|As Received (Wet) Basis 18. Ash|Combined|%|As Received (Wet) Basis 19. Barium|Combined|%|As Received (Wet) Basis 20. Boron|Combined|%|As Received (Wet) Basis 21. Cadmium|Combined|%|As Received (Wet) Basis 22. Calcium|Combined|%|As Received (Wet) Basis 23. Calcium Carbonate Equivalent (CCE)|Combined|%|As Received (Wet) Basis 24. Carbon:Nitrogen Ratio|Combined|C:N|As Received (Wet) Basis 25. Cation Ratio of Soil Structural Stability (CROSS)|Combined|Ratio|As Received (Wet) Basis 26. Chloride|Combined|%|As Received (Wet) Basis 27. Chromium|Combined|%|As Received (Wet) Basis 28. Cobalt|Combined|%|As Received (Wet) Basis 29. Copper|Combined|%|As Received (Wet) Basis 30. Electrical Conductivity|Combined|dS/m|As Received (Wet) Basis 31. Iron|Combined|%|As Received (Wet) Basis 32. Lead|Combined|%|As Received (Wet) Basis 33. Magnesium|Combined|%|As Received (Wet) Basis 34. Manganese|Combined|%|As Received (Wet) Basis 35. Moisture|Combined|%|As Received (Wet) Basis 36. Molybdenum|Combined|%|As Received (Wet) Basis 37. Nickel|Combined|%|As Received (Wet) Basis 38. Nitrate-N|Combined|%|As Received (Wet) Basis 39. Organic Matter|Combined|%|As Received (Wet) Basis 40. pH|Combined|none|As Received (Wet) Basis 41. Phosphorus|Combined|%|As Received (Wet) Basis 42. Potassium|Combined|%|As Received (Wet) Basis 43. Selenium|Combined|%|As Received (Wet) Basis 44. Silicon|Combined|%|As Received (Wet) Basis 45. Sodium|Combined|%|As Received (Wet) Basis 46. Sodium Absorption Ratio (SAR)|Combined|Ratio|As Received (Wet) Basis 47. Strontium|Combined|%|As Received (Wet) Basis 48. Sulfate-S|Combined|%|As Received (Wet) Basis 49. Sulfur|Combined|%|As Received (Wet) Basis 50. Total Carbon|Combined|%|As Received (Wet) Basis 51. Total Dissolved Solids|Combined|%|As Received (Wet) Basis 52. Total Nitrogen|Combined|%|As Received (Wet) Basis 53. Total Solids|Combined|%|As Received (Wet) Basis 54. Total Suspended Solids|Combined|mg/L|As Received (Wet) Basis 55. Volatile Solids|Combined|%|As Received (Wet) Basis 56. Water Extractable Phosphorus|Combined|%|As Received (Wet) Basis 57. Zinc|Combined|%|As Received (Wet) Basis ----------------------------------------- REFERENCES ----------------------------------------- 1. R Core Team (2023). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. 1. Benjamini, Y., & Hochberg, Y. (1995). Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society: Series B (Methodological), 57(1), 289–300. https://doi.org/10.1111/j.2517-6161.1995.tb02031.x 2. Bohl Bormann, N. L., Wilson, M. L., Cortus, E. L., Silverstein, K. A. T., Janni, K. A., & Gunderson, L. M. (2024). ManureDB. http://manuredb.umn.edu/ 3. Hothorn, T., Winell, H., Hornik, K., van de Wiel, M. A., & Zeileis, A. (2023). coin: Conditional Inference Procedures in a Permutation Test Framework (1.4-3) [Computer software]. https://CRAN.R-project.org/package=coin 4. Jarque, C. M., & Bera, A. K. (1987). A Test for Normality of Observations and Regression Residuals. International Statistical Review / Revue Internationale de Statistique, 55(2), 163–172. https://doi.org/10.2307/1403192 5. Kendall, M. G. (1955). Rank Correlation Methods. Griffin. https://www.cambridge.org/core/journals/journal-of-the-institute-of-actuaries/article/abs/ rank-correlation-methods-by-maurice-g-kendall-ma-pp-vii-160-london-charles-griffin-and-co- ltd-42-drury-lane-1948-18s/335FD9F4DDCBDF8A7167B4B66EE4DFB9 6. Mann, H. B. (1945). Nonparametric Tests Against Trend. Econometrica, 13(3), 245. 7. Mann, H. B., & Whitney, D. R. (1947). On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other. The Annals of Mathematical Statistics, 18(1), 50–60. https://doi.org/10.1214/aoms/1177730491 8. McLeod, A. I. (2022). Kendall: Kendall Rank Correlation and Mann-Kendall Trend Test (2.2.1) [Computer software]. https://cran.r-project.org/web/packages/Kendall/index.html 9. R Core Team. (2023). R: A language and environment for statistical computing (4.3.2) [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org/ 10. Sainani, K. L. (2012). Dealing With Non-normal Data. PM&R, 4(12), 1001–1005. https://doi.org/10.1016/j.pmrj.2012.10.013 11. Trapletti, A., Hornik, K., & LeBaron, B. (2023). tseries: Time Series Analysis and Computational Finance (0.10-55) [Computer software]. https://cran.r-project.org/package=tseries 12. Wickham, H., Chang, W., Henry, L., Pedersen, T. L., Takahashi, K., Wilke, C., Woo, K., Yutani, H., Dunnington, D., Brand, T. van den, Posit, & PBC. (2024). ggplot2: Create Elegant Data Visualisations Using the Grammar of Graphics (3.5.0) [Computer software]. https://cran.r-project.org/web/packages/ggplot2/index.html 13. Wilson, M. L., Cortus, S., Brimmer, R., Floren, J., Gunderson, L., Hicks, K., Hoerner, T., Lessl, J., Meinen, R. J., Miller, R. O., Mowrer, J., Porter, J., Spargo, J. T., Thayer, B., & Vocasek, F. (2022). Recommended Methods of Manure Analysis, Second Edition. University of Minnesota Libraries Publishing. http://conservancy.umn.edu/handle/11299/227650