Monotonic trend computations for streamflow statistics within and near the Mobile Bay and Perdido Bay watersheds, United States, 1950–2022

This data release provides comprehensive results of monotonic trend assessment for long-term U.S. Geological Survey (USGS) streamgages in or proximal to the watersheds of Mobile and Perdido Bays, south-central United States (Tatum and others, 2024). Long-term is defined as streamgages having at least five complete decades of daily streamflow data since January 1, 1950, exclusive to those streamgages also having the entire 2010s decade represented. Input data for the trend assessment are daily streamflow data retrieved on March 8, 2024 (U.S. Geological Survey, 2024) and formatted using the fill_dvenv() function in akqdecay (Crowley-Ornelas and others, 2024). Monotonic trends were assessed for each of 69 streamgages using 26 Mann-Kendall hypothesis tests for 20 hydrologic metrics understood as particularly useful in ecological studies (Henriksen and others, 2006) with another 6 metrics measuring well-known streamflow properties, such as annual harmonic mean streamflow (Asquith and Heitmuller, 2008) and annual mean streamflow with decadal flow-duration curve quantiles (10th, 50th, and 90th percentiles) (Crowley-Ornelas and others, 2023). Helsel and others (2020) provide background and description of the Mann-Kendall hypothesis test. Some of the trend analyses are based on the annual values of a hydrologic metric (calendar year is the time interval test) whereas others are decadal (decade is the time interval for the test). The principal result output for this data release (**monotrnd_1hyp.txt**) clearly distinguishes the time interval for the respective tests. This data release includes the computational workflow to conduct the hypothesis testing and requisite data manipulations to do so. The workflow is comprised of the core computation script **monotrnd_script.R** and an auxiliary script containing functions for 20 ecological flow metrics. This means that script **monotrnd_script.R** requires additional functions to be loaded into the R workspace and sources the file **monotrnd_ecomets_include.R**. This design is useful as part of isolation of the 20 ecological-oriented hydrologic metrics (subroutines) (logic and nomenclature therein is informed by Henriksen and others, 2006) from the streamgage-looping workflow and other data manipulation features in **monotrnd_script.R**. The **monotrnd_script.R** is designed to use time series of daily mean streamflow stored in an R environment data object using the streamgage identification number as the key and a data frame (table) of the daily streamflows in the format defined by the **dvget()** and filled by the **filldv_env()** functions of the **akqdecay** R package (See supplemental information section; Crowley-Ornelas and others, 2024). Additionally, **monotrnd_script.R** tags a specific subset of streamgages within the workflow, identified by the authors as "major nodes," with a binary indicator (1 or 0) to support targeted analyses on these selected locations. The data in file **monotrnd_1hyp.txt** are comma-delimited results of Kendall tau or other test statistics and p-values of the Mann-Kendall hypothesis tests as part of monotonic trend assessment for 69 USGS streamgages using 26 Mann–Kendall hypothesis tests on a variety of streamflow metrics. The data include USGS streamgage identification numbers with prepended "S" character, decimal latitudes and longitudes for the streamgage locations, range of calendar year and decades of streamflow processed along with integer counts of number of calendar years and decades, Kendall tau (or other test statistic) and associated p-value of the test statistic for the 26 streamflow metrics considered. Broadly, the "left side of the table" presents the results for the tests on metrics using calendar year time steps, and the "right side of the table" presents the results for the tests on metrics using decade time steps. The content of the file does not assign or draw conclusions on statistical significance because the p-values are provided. The file **monotrnd_dictionary_1hyp.txt** is a simple plain-text, pipe-delimited file of directly human-readable short definitions for the columns in the **monotrnd_1hyp.txt**. (This dictionary and two others accompany this data release to facilitate potential reuse of information by some users.) The source of **monotrnd_1hyp.txt** stems from ending computational steps in script **monotrnd_script.R**. Short summaries synthesizing information in file **monotrnd_1hyp.txt** are available in files **monotrnd_3cnt.txt** and **monotrnd_2stn.txt** also accompanying this data release. The data in file **monotrnd_2stn.txt** are comma-delimited summaries by streamgage identification number of the monotonic trend assessments for 26 Mann-Kendall hypothesis tests on streamflow metrics as described elsewhere in this data release. The summary data herein are composed of records (rows) by streamgage that include columns of (1) streamgage identification numbers with a prepended "S" character, (2) decimal latitudes and longitudes for the streamgage locations, (3) the integer counts of the number of hypothesis tests, (4) the integer count of number of tests for which the computed hypothesis test p-values less than the 0.05 level of statistical significance (so-called alpha = 0.05), and (5) colon-delimited strings of alphanumeric characters identifying each of the statistically significant tests for the respective streamgage. The file **monotrnd_dictionary_2stn.txt** is a simple plain-text, pipe-delimited file of directly human-readable short definitions for the columns in **monotrnd_2stn.txt**. The source of **monotrnd_2stn.txt** stems from ending computational steps in script **monotrnd_script.R** described elsewhere in this data release from its production of the **monotrnd_1hyp.txt**; this later data file provides the values used to assemble **monotrnd_2stn.txt**. The information in file **monotrnd_3cnt.txt** are comma-delimited summaries of Kendall tau or other test statistic arithmetic means as well as integer counts of statistically significant trends as part of monotonic trend assessment using 26 Mann-Kendall hypothesis tests on a variety of streamflow metrics for 69 USGS streamgages as described elsewhere in this data release. The two-column summary data herein are composed of a first row indicating by character string of the integer number of streamgages (69) and then subsequent rows in pairs of three-decimal character-string representation of mean Kendall tau (or the test statistics of a seasonal Mann-Kendall test) followed by character string of the integer number of the counts of statistically significant tests for the respective test at it was applied to the 69 streamgages. Statistical significance is defined as p-values less than the 0.05 level of statistical significance (so-called alpha = 0.05). The file **monotrnd_dictionary_3cnt.txt** is a simple plain-text, pipe-delimited file of directly human-readable short definitions for the columns in the **monotrnd_3cnt.txt**. The source of **monotrnd_3cnt.txt** stems from ending computational steps in script **monotrnd_script.R** described elsewhere in this data release from its production of the **monotrnd_1hyp.txt**; this later data file provides the values used to assemble **monotrnd_3cnt.txt**.

[
  "010:12"
]

[
  "geospatial"
]