heat_index warnings and bug by anissa111 · Pull Request #807 · NCAR/geocat-comp

anissa111 · 2026-01-13T22:46:44Z

PR Summary

This PR rewrites heat_index to match the NWS's javascript implementation on https://www.wpc.ncep.noaa.gov/html/heatindex.shtml

function heat1_c(HI1_C){
    if(HI1_C.RHumidity.value > 100){
        alert("Relative humidity cannot exceed 100%.");
    }
    else if (HI1_C.RHumidity.value < 0) {
        alert("Relative humidity cannot be less than 0%.");
    }
    else if (parseFloat(HI1_C.tempair.value) <= 4.44) {
        var hi = HI1_C.tempair.value;
    }
    else{
        var tempair_in_fahrenheit = 1.80 * HI1_C.tempair.value + 32.0;
        var hitemp = 61.0+((tempair_in_fahrenheit-68.0)*1.2)+(HI1_C.RHumidity.value*0.094);
        var fptemp = parseFloat(HI1_C.tempair.value);
        var hifinal = 0.5*(fptemp+hitemp);

        if(hifinal > 79.0){
            var hi = -42.379 + 2.04901523 * tempair_in_fahrenheit + 10.14333127 * HI1_C.RHumidity.value - 0.22475541 * tempair_in_fahrenheit * HI1_C.RHumidity.value - 6.83783 * (Math.pow(10, -3)) * (Math.pow(tempair_in_fahrenheit, 2)) - 5.481717 * (Math.pow(10, -2)) * (Math.pow(HI1_C.RHumidity.value, 2)) + 1.22874 * (Math.pow(10, -3)) * (Math.pow(tempair_in_fahrenheit, 2)) * HI1_C.RHumidity.value + 8.5282 * (Math.pow(10, -4)) * tempair_in_fahrenheit * (Math.pow(HI1_C.RHumidity.value, 2)) - 1.99 * (Math.pow(10, -6)) * (Math.pow(tempair_in_fahrenheit, 2)) * (Math.pow(HI1_C.RHumidity.value,2));

            if((HI1_C.RHumidity.value <= 13) && (tempair_in_fahrenheit >= 80.0) && (tempair_in_fahrenheit <= 112.0)) {
                var adj1 = (13.0-HI1_C.RHumidity.value)/4.0;
                var adj2 = Math.sqrt((17.0-Math.abs(tempair_in_fahrenheit-95.0))/17.0);
                var adj = adj1 * adj2;
                var hi = hi - adj;
            }
            else if ((HI1_C.RHumidity.value > 85.0) && (tempair_in_fahrenheit >= 80.0) && (tempair_in_fahrenheit <= 87.0)) {
                var adj1 = (HI1_C.RHumidity.value-85.0)/10.0;
                var adj2 = (87.0-tempair_in_fahrenheit)/5.0;
                var adj = adj1 * adj2;
                var hi = hi + adj;
            }
        }
        else{
            var hi = hifinal;
        }
    }
    twoplaces_rh1 = hi.toFixed(1);
    celsius_temp_rh1 = (hi - 32) * .556;
    twoplaces_c_rh1 = celsius_temp_rh1.toFixed(1);
    HI1_C.heatindex.value = twoplaces_rh1 + " F" +  " / " + twoplaces_c_rh1 + " C";

Summary of changes:

updates steadman calculation to avoid duplication of the temperature average
changes some of the ranges very slightly where adjustments are applied to match javascript implementation
removes unnecessary internal functions _xheat_index and _heat_index and absorbed them into the main heat_index function
slightly alters the xarray metadata to reflect the calculation changes
fixes heat_index encounters sqrt of negatives #596 by adding an abs to the sqrt calculation
adds testing based on output from the NWS javascript

In case anybody needs to do this in the future, this is the script I ran on a chrome console on the webpage with the javascript:

const form = document.forms[4];
const t = form.elements[0];
const rh = form.elements[4];
const calc = form.elements[5];
const out = form.elements[7];
const rhs = [40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100];
const temps = [80, 82, 84, 86, 88, 90, 92, 94, 96, 98, 100, 102, 104, 106, 108, 110];

var nt = temps.length;
var nrh = rhs.length;

for (let i = 0; i < nt; i++) {
    for (let j = 0; j < nrh; j++) {
        t.value = temps[i];
        rh.value = rhs[j];
        calc.click();
        result[i][j] = out.value.split(' ')[0];
        console.log(temps[i] + " " + rhs[j] + " " + result[i][j])
    }
}

// make csv
let csvContent = ''

result.forEach(row => {
  csvContent += row.join(',') + '\n'
})

const blob = new Blob([csvContent], { type: 'text/csv;charset=utf-8,' })
const objUrl = URL.createObjectURL(blob)

const link = document.createElement('a')
link.setAttribute('href', objUrl)
link.setAttribute('download', 'File.csv')
link.textContent = 'Click to Download'

document.querySelector('body').append(link)

link //click on link displayed

Related Tickets & Documents

Closes #806 and #596

PR Checklist

General

PR includes a summary of changes
Link relevant issues, make one if none exist
Add a brief summary of changes to docs/release-notes.rst in a relevant section for the upcoming release.
Add appropriate labels to this PR
PR follows the Contributor's Guide

Functionality

New function(s) intended for public API added to geocat/comp/__init__.py file

Testing

Update or create tests in appropriate test file

…eat_index internal funcs

github-actions · 2026-01-13T23:10:46Z

Meowdy! Here's your benchmark comparison:
Baseline: 004f90d (NCAR:main)
Contender: dcd4cab (anissa111:negsqrt)

Benchmarks that have stayed the same:

Before [`004f90d`]	After [`dcd4cab`]	Ratio	Benchmark (Parameter)
426±3ms	430±2ms	1.01	bench_climatologies.Bench_calendar_average.time_calendar_average [runnervmi13qx/conda-py3.10]
427±10ms	428±2ms	1	bench_climatologies.Bench_calendar_average.time_calendar_average [runnervmi13qx/conda-py3.13]
37.6±0.4ms	39.2±0.2ms	1.04	bench_climatologies.Bench_climate_anomaly.time_climate_anomaly [runnervmi13qx/conda-py3.10]
37.5±0.6ms	38.6±0.3ms	1.03	bench_climatologies.Bench_climate_anomaly.time_climate_anomaly [runnervmi13qx/conda-py3.13]
44.9±0.7ms	45.0±1ms	1	bench_climatologies.Bench_climatology_average.time_climatology_average [runnervmi13qx/conda-py3.10]
45.2±0.4ms	44.3±0.3ms	0.98	bench_climatologies.Bench_climatology_average.time_climatology_average [runnervmi13qx/conda-py3.13]
7.94±0.06ms	7.92±0.06ms	1	bench_climatologies.Bench_month_to_season.time_month_to_season [runnervmi13qx/conda-py3.10]
8.33±0.2ms	8.01±0.03ms	0.96	bench_climatologies.Bench_month_to_season.time_month_to_season [runnervmi13qx/conda-py3.13]
510M	510M	1	bench_fourier.Bench_fourier.peakmem_band_block [runnervmi13qx/conda-py3.10]
510M	510M	1	bench_fourier.Bench_fourier.peakmem_band_block [runnervmi13qx/conda-py3.13]
145±1ms	143±1ms	0.99	bench_fourier.Bench_fourier.time_band_block [runnervmi13qx/conda-py3.10]
145±2ms	141±0.8ms	0.98	bench_fourier.Bench_fourier.time_band_block [runnervmi13qx/conda-py3.13]
262M	263M	1	bench_interpolation.Bench_interp_hybrid_to_pressure.peakmem_interp_hybrid_to_pressure [runnervmi13qx/conda-py3.10]
263M	263M	1	bench_interpolation.Bench_interp_hybrid_to_pressure.peakmem_interp_hybrid_to_pressure [runnervmi13qx/conda-py3.13]
39.4±0.4ms	39.7±0.3ms	1.01	bench_interpolation.Bench_interp_hybrid_to_pressure.time_interp_hybrid_to_pressure [runnervmi13qx/conda-py3.10]
39.9±0.5ms	39.5±0.2ms	0.99	bench_interpolation.Bench_interp_hybrid_to_pressure.time_interp_hybrid_to_pressure [runnervmi13qx/conda-py3.13]
258M	258M	1	bench_interpolation.Bench_interp_hybrid_to_pressure_extrap_temp.peakmem_interp_hybrid_to_pressure_extrap_temp [runnervmi13qx/conda-py3.10]
258M	258M	1	bench_interpolation.Bench_interp_hybrid_to_pressure_extrap_temp.peakmem_interp_hybrid_to_pressure_extrap_temp [runnervmi13qx/conda-py3.13]
18.8±0.2ms	18.6±0.2ms	0.99	bench_interpolation.Bench_interp_hybrid_to_pressure_extrap_temp.time_interp_hybrid_to_pressure_extrap_temp [runnervmi13qx/conda-py3.10]
18.5±0.2ms	18.4±0.1ms	1	bench_interpolation.Bench_interp_hybrid_to_pressure_extrap_temp.time_interp_hybrid_to_pressure_extrap_temp [runnervmi13qx/conda-py3.13]
719M	719M	1	bench_interpolation.Bench_interp_multidim.peakmem_interp_multidim_chunk [runnervmi13qx/conda-py3.10]
719M	719M	1	bench_interpolation.Bench_interp_multidim.peakmem_interp_multidim_chunk [runnervmi13qx/conda-py3.13]
5.18±0.06s	5.42±0.02s	1.05	bench_interpolation.Bench_interp_multidim.time_interp_multidim_chunk [runnervmi13qx/conda-py3.10]
5.39±0.01s	5.30±0.02s	0.98	bench_interpolation.Bench_interp_multidim.time_interp_multidim_chunk [runnervmi13qx/conda-py3.13]
3.38±0.03ms	3.42±0.04ms	1.01	bench_interpolation.Bench_interp_sigma_to_hybrid.time_interp_sigma_to_hybrid_1d [runnervmi13qx/conda-py3.10]
3.34±0.04ms	3.32±0.01ms	0.99	bench_interpolation.Bench_interp_sigma_to_hybrid.time_interp_sigma_to_hybrid_1d [runnervmi13qx/conda-py3.13]
14.7±0.2ms	14.4±0.3ms	0.98	bench_interpolation.Bench_interp_sigma_to_hybrid.time_interp_sigma_to_hybrid_3d [runnervmi13qx/conda-py3.10]
14.3±0.2ms	14.3±0.1ms	1	bench_interpolation.Bench_interp_sigma_to_hybrid.time_interp_sigma_to_hybrid_3d [runnervmi13qx/conda-py3.13]
8.92±0.02μs	8.74±0.1μs	0.98	bench_meteorology.Bench_actual_saturation_vapor_pressure.time_actual_saturation_vapor_pressure [runnervmi13qx/conda-py3.10]
8.79±0.1μs	8.60±0.03μs	0.98	bench_meteorology.Bench_actual_saturation_vapor_pressure.time_actual_saturation_vapor_pressure [runnervmi13qx/conda-py3.13]
18.6±0.2μs	18.4±0.5μs	0.99	bench_meteorology.Bench_delta_pressure.time_delta_pressure [runnervmi13qx/conda-py3.10]
18.4±0.07μs	18.0±0.06μs	0.98	bench_meteorology.Bench_delta_pressure.time_delta_pressure [runnervmi13qx/conda-py3.13]
17.1±0.2μs	17.3±0.03μs	1.01	bench_meteorology.Bench_dewtemp.time_dewtemp [runnervmi13qx/conda-py3.10]
17.2±0.1μs	17.3±0.1μs	1.01	bench_meteorology.Bench_dewtemp.time_dewtemp [runnervmi13qx/conda-py3.13]
183±0.8μs	179±2μs	0.98	bench_meteorology.Bench_heat_index.time_heat_index [runnervmi13qx/conda-py3.10]
181±1μs	176±0.7μs	0.97	bench_meteorology.Bench_heat_index.time_heat_index [runnervmi13qx/conda-py3.13]
318±1μs	320±3μs	1.01	bench_meteorology.Bench_max_daylight.time_max_daylight [runnervmi13qx/conda-py3.10]
316±0.6μs	317±3μs	1	bench_meteorology.Bench_max_daylight.time_max_daylight [runnervmi13qx/conda-py3.13]
1.30±0.02μs	1.27±0.02μs	0.98	bench_meteorology.Bench_psychrometric_constant.time_psychrometric_constant [runnervmi13qx/conda-py3.10]
1.29±0.04μs	1.26±0μs	0.98	bench_meteorology.Bench_psychrometric_constant.time_psychrometric_constant [runnervmi13qx/conda-py3.13]
25.1±0.2μs	25.6±0.1μs	1.02	bench_meteorology.Bench_relhum.time_relhum [runnervmi13qx/conda-py3.10]
25.1±0.2μs	24.7±0.1μs	0.98	bench_meteorology.Bench_relhum.time_relhum [runnervmi13qx/conda-py3.13]
8.62±0.1μs	8.61±0.1μs	1	bench_meteorology.Bench_saturation_vapor_pressure.time_saturation_vapor_pressure [runnervmi13qx/conda-py3.10]
8.50±0.1μs	8.56±0.1μs	1.01	bench_meteorology.Bench_saturation_vapor_pressure.time_saturation_vapor_pressure [runnervmi13qx/conda-py3.13]
11.0±0.2μs	11.1±0.2μs	1.01	bench_meteorology.Bench_saturation_vapor_pressure_slope.time_saturation_vapor_pressure_slope [runnervmi13qx/conda-py3.10]
10.9±0.1μs	10.8±0.2μs	0.99	bench_meteorology.Bench_saturation_vapor_pressure_slope.time_saturation_vapor_pressure_slope [runnervmi13qx/conda-py3.13]
344M	344M	1	bench_spherical.Bench_spherical.peakmem_decomposition [runnervmi13qx/conda-py3.10]
344M	344M	1	bench_spherical.Bench_spherical.peakmem_decomposition [runnervmi13qx/conda-py3.13]
305M	305M	1	bench_spherical.Bench_spherical.peakmem_recomposition [runnervmi13qx/conda-py3.10]
305M	305M	1	bench_spherical.Bench_spherical.peakmem_recomposition [runnervmi13qx/conda-py3.13]
229M	229M	1	bench_spherical.Bench_spherical.peakmem_scale_voronoi [runnervmi13qx/conda-py3.10]
229M	229M	1	bench_spherical.Bench_spherical.peakmem_scale_voronoi [runnervmi13qx/conda-py3.13]
2.06±0s	2.06±0s	1	bench_spherical.Bench_spherical.time_decomposition [runnervmi13qx/conda-py3.10]
2.07±0s	2.06±0.01s	1	bench_spherical.Bench_spherical.time_decomposition [runnervmi13qx/conda-py3.13]
2.07±0s	2.07±0.01s	1	bench_spherical.Bench_spherical.time_recomposition [runnervmi13qx/conda-py3.10]
2.07±0.01s	2.07±0.01s	1	bench_spherical.Bench_spherical.time_recomposition [runnervmi13qx/conda-py3.13]
250±6ms	239±5ms	0.96	bench_spherical.Bench_spherical.time_scale_voronoi [runnervmi13qx/conda-py3.10]
253±4ms	245±10ms	0.97	bench_spherical.Bench_spherical.time_scale_voronoi [runnervmi13qx/conda-py3.13]
210M	210M	1	bench_stats.Bench_eofunc_eofs.peakmem_eofunc [runnervmi13qx/conda-py3.10]
210M	210M	1	bench_stats.Bench_eofunc_eofs.peakmem_eofunc [runnervmi13qx/conda-py3.13]
9.58±0.2ms	9.72±0.09ms	1.01	bench_stats.Bench_eofunc_eofs.time_eofunc [runnervmi13qx/conda-py3.10]
9.80±0.2ms	9.49±0.2ms	0.97	bench_stats.Bench_eofunc_eofs.time_eofunc [runnervmi13qx/conda-py3.13]
415M	416M	1	bench_stats.Bench_eofunc_ps.peakmem_eofunc_ps [runnervmi13qx/conda-py3.10]
416M	416M	1	bench_stats.Bench_eofunc_ps.peakmem_eofunc_ps [runnervmi13qx/conda-py3.13]
426±5ms	423±2ms	0.99	bench_stats.Bench_eofunc_ps.time_eofunc_ps [runnervmi13qx/conda-py3.10]
426±3ms	419±3ms	0.98	bench_stats.Bench_eofunc_ps.time_eofunc_ps [runnervmi13qx/conda-py3.13]
48.8±0.9μs	49.8±0.1μs	1.02	bench_stats.Bench_pearson_r.time_pearson_r [runnervmi13qx/conda-py3.10]
48.0±0.1μs	49.3±0.3μs	1.03	bench_stats.Bench_pearson_r.time_pearson_r [runnervmi13qx/conda-py3.13]
1.74±0.01s	1.75±0.02s	1	import.Import.timeraw_import_geocat_comp [runnervmi13qx/conda-py3.10]
1.75±0.01s	1.74±0.01s	0.99	import.Import.timeraw_import_geocat_comp [runnervmi13qx/conda-py3.13]

test/test_dask.py

docs/release-notes.rst

kafitzgerald

This all looks good to me with perhaps the exception of listing it as a bug fix or something more user facing so that folks can easily find something to understand what happened should they notice a diff in the results.

cyschneck

Looks good! I wonder if this will solve the issues I ran into when attempting to adding heat_index into applications (old PR that should be closed/fixed: NCAR/geocat-applications#124)

anissa111 added 9 commits January 7, 2026 17:45

remade heat_index, not cleaned up, but seems to be largely working

6cd8447

log comparison notebook and function state, remove notebook later

54c2013

mostly finished, better performance than old version

28903aa

begin switch over, the nws table does not match their calculator

381db90

finished transfer from old functions, removed old _xheat_index and _h…

9462c5e

…eat_index internal funcs

remove deleted internal heat index functions from internal API

852e401

Merge remote-tracking branch 'upstream' into negsqrt

b55deb2

add to release notes

8c0b9b9

remove testing notebook

e4bf555

anissa111 added bug Something isn't working testing Issue related to testing labels Jan 13, 2026

anissa111 added 2 commits January 13, 2026 15:49

put function back in original order to help git diff not be so gnarly

8cbc190

add new testing data

08e225e

anissa111 added the run-benchmark Add tag to a PR to run ASV comparison on new commits label Jan 13, 2026

anissa111 added 2 commits January 13, 2026 16:12

add try/catch for testing file path like others

b44c8f2

also change testing arrays on dask test

410e7f3

anissa111 marked this pull request as ready for review January 13, 2026 23:55

anissa111 requested review from cyschneck, jukent and kafitzgerald January 13, 2026 23:56

remove other references to removed internal funcs, typo

eb853d3

kafitzgerald reviewed Jan 14, 2026

View reviewed changes

test/test_dask.py Show resolved Hide resolved

kafitzgerald reviewed Jan 14, 2026

View reviewed changes

docs/release-notes.rst Outdated Show resolved Hide resolved

kafitzgerald approved these changes Jan 14, 2026

View reviewed changes

internal change -> bug fix

caecde5

cyschneck approved these changes Jan 15, 2026

View reviewed changes

anissa111 merged commit ac74483 into NCAR:main Jan 15, 2026
17 checks passed

anissa111 deleted the negsqrt branch January 15, 2026 23:52

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

heat_index warnings and bug#807

heat_index warnings and bug#807
anissa111 merged 15 commits intoNCAR:mainfrom
anissa111:negsqrt

anissa111 commented Jan 13, 2026 •

edited

Loading

Uh oh!

github-actions bot commented Jan 13, 2026 •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

kafitzgerald left a comment

Uh oh!

cyschneck left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

anissa111 commented Jan 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

PR Summary

Related Tickets & Documents

PR Checklist

Uh oh!

github-actions bot commented Jan 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kafitzgerald left a comment

Choose a reason for hiding this comment

Uh oh!

cyschneck left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

anissa111 commented Jan 13, 2026 •

edited

Loading

github-actions bot commented Jan 13, 2026 •

edited

Loading