handprint : drying shifts

watercolor drying shifts

It's well known that watercolor paints change color appearance as they dry, but no art reference has published reliable information about these changes — until now.

Why Do Shifts Occur? A detailed explanation of paint color shifts is given in the page on the secret of glowing color. Briefly, color is created because a pigment absorbs some light wavelengths and reflects others. However, some light at all wavelengths (white light) is always scattered at any surface boundary between two substances that have different refractive indices or light bending abilities. Many substances, including paint pigments, have a similar refractive index to water, so wet paint appears to have a rich color because scattered white light is minimized. When watercolor paint dries, however, the water evaporates and leaves the paint in contact with air, which has a much lower refractive index than pigment. This causes more white light scattering at the pigment surface, which mixes with the reflected pigment color to make the apparent color appear lighter and duller.

Paint Measurements. The table below summarizes the drying shift in mass tone color appearance for 75 of the most commonly used watercolor pigments or convenience mixtures. The shifts were measured as the difference in masstone color appearance as measured with a GretagMacbeth Spectrolino spectrophotometer when the paints were at a satin wetness and again after they had dried for one hour; the tabulated results are the average of measurements on two separate samples of paint. The color shift is shown as separate changes in lightness, chroma, hue angle, and as the total color shift:

• L or lightness is measured from 0 (pure black) to 100 (pure white); a shift of 10 equals one step on a 11-step value scale; a minus sign means the paint became darker. The average shift in L is only 2.3.

• C or chroma is measured from 0 (pure gray) to 100 or more (high chroma); a minus sign means the paint became less saturated (duller). The average shift in C is -10.7; this loss of chroma is the main cause of the drying shift in watercolors.

• h or hue is measured in degrees of the CIELAB a*b* plane, with 0 at magenta and 90 at yellow. The average shift in h is -2.9: as they dry, most colors shift clockwise on the CIELAB a*b* plane or counterclockwise on the artist's color wheel (yellow toward red, blue toward green).

• Total Shift is the combined (three dimensional) change across L, C and h; this is the total apparent effect of the drying shift. The average across all paints is 12.9.

Color shifts are highly variable — they depend on paint vehicle, dilution and application method — so use this table as a rough guide. Even so, it's clear that drying shifts are not limited to a lightening of the paint, as usually claimed: over one third of the pigments listed here actually darkened! The consistent change is that paints lose chroma as they dry (see diagram at right); sometimes large hue shifts occur as well. All of these shifts combine to make up the total change in color appearance.

In the guide to watercolor pigments, the total drying shift is reported in terms of the percentage change in lightness, chroma or hue angle. An equal lightness or chroma change will appear larger in darker or duller paints, because they have so little lightness or chroma to begin with.

This table lists pigments by color index name. The links below allow you to see the data sorted by increasing changes in L, C, h, total shift, or percentage total shift.

Sort by: Lightness ShiftChroma ShiftHue ShiftTotal ShiftTotal Shift % • Color Index Name
Watercolor Paint Drying Shifts
CI NamePigment/Paint NameBrandShiftTotal
PB15:1phthalo blue RSWinsor & Newton9.2-14.7-5.917.6 (71%)
PB15:3phthalo blue GSWinsor & Newton6.9-10.4-5.312.7 (46%)
PB16phthalo turquoiseMaimeriBlu7.4-14.04.615.9 (48%)
PB17phthalo cyanHolbein7.3-8.9-2.711.6 (35%)
PB27prussian blueWinsor & Newton11.4-9.3-11.315.2 (93%)
PB28cobalt blueWinsor & Newton6.5-10.7-2.912.7 (33%)
PB29french ultramarine blueWinsor & Newton7.3-15.8-4.117.7 (47%)
PB36cerulean blueWinsor & Newton0.0- (5%)
PB36cobalt turquoiseWinsor & Newton0.1-8.9-2.99.5 (22%)
PB60indanthrone blueWinsor & Newton8.7-4.7-7.010.2 (64%)
PB72cobalt blue deepWinsor & Newton3.5-7.3-1.98.2 (21%)
PBk7lamp blackWinsor & Newton16.21.532.116.2 (990%)
PBk9ivory blackWinsor & Newton13.31.57.513.5 (150%)
.sepiaWinsor & Newton10. (46%)
PBr7burnt sienna [dark]Daniel Smith3.7-12.7-4.213.4 (32%)
PBr7burnt sienna [light]Winsor & Newton0.0-14.5-2.914.6 (21%)
PBr7burnt umberWinsor & Newton4.5-9.3-2.510.5 (39%)
PBr7raw umber [dark]Daniel Smith7.7-3.7-2.78.5 (48%)
PBr7raw umber [light]Winsor & Newton1.1-4.2-0.84.6 (10%)
PBr7venetian redWinsor & Newton0.3-11.9-5.312.1 (23%)
.naples yellow Winsor & Newton-2.98.8-2.39.4 (21%)
PBr24naples yellow deepWinsor & Newton-4.4-2.1-4.95.8 (10%)
PG7phthalo green BSWinsor & Newton6.4-10.1-2.112.8 (33%)
.sap greenWinsor & Newton9.3-11.8-0.216.2 (57%)
.hooker's greenDaniel Smith7.1-14.2-3.015.9 (54%)
.permanent greenDaniel Smith3.0-18.2-3.818.6 (29%)
.permanent green lightMaimeriBlu0.8-10.9-1.411.0 (14%)
PG17chromium oxide greenWinsor & Newton-6.2-12.14.913.6 (49%)
PG36phthalo green YSWinsor & Newton6.3-12.8-2.514.6 (35%)
PG50cobalt green BSWinsor & Newton1.2-4.9-2.15.2 (14%)
PG50cobalt green YSWinsor & Newton-2.9-4.3-1.25.3 (14%)
PG50cobalt teal blueWinsor & Newton1.4-5.1-0.55.4 (12%)
PO20cadmium orange [pure]Daniel Smith-2.0-19.3-8.120.7 (23%)
PO20cadmium orange [mixed]Winsor & Newton-4.1-10.7-8.513.5 (19%)
PO43perinone orangeDaniel Smith-1.8-18.2-7.019.2 (22%)
PO48quinacridone burnt orangeDaniel Smith4.5-22.2-4.622.8 (38%)
PO49quinacridone goldWinsor & Newton-1.5-18.1-4.718.5 (24%)
PO62benzimidazolone orangeWinsor & Newton-1.0-16.6-5.217.3 (18%)
PR N/Aquinacridone carmineWinsor & Newton0.9-10.5-3.510.8 (17%)
PR88thioindigo violetWinsor & Newton4.2-10.5-0.211.5 (38%)
PR108cadmium red deepUtrecht-0.5-21.8-9.422.5 (31%)
PR108cadmium redDaniel Smith1.2-16.9-7.017.7 (23%)
PR108cadmium scarletWinsor & Newton-2.0-13.0-6.914.2 (19%)
PR112naphthol redHolbein-0.8-9.4-2.69.8 (13%)
PR122quinacridone magentaWinsor & Newton4.6-12.0-5.113.2 (29%)
PR149perylene scarletDaniel Smith5.4-21.6-7.422.7 (43%)
PR170naphthol redDaniel Smith0.9-16.5-6.317.2 (21%)
PR171benzimidazolone maroonDaniel Smith10.0- (67%)
PR176benzimidazolone carmineDaniel Smith3.8-18.0-6.418.8 (35%)
PR177anthraquinoid redDaniel Smith4.5-18.1-6.619.1 (37%)
PR178perylene redDaniel Smith2.4-19.3-6.519.9 (30%)
PR179perylene maroonWinsor & Newton5.1-20.4-3.121.0 (47%)
PR188naphthol scarletWinsor & Newton0.3-14.5-5.315.2 (16%)
PR206quinacridone maroonWinsor & Newton2.3-16.7-5.017.2 (29%)
PR209quinacridone redWinsor & Newton0.6-11.0-3.911.4 (14%)
PR254pyrrole redWinsor & Newton-0.1-13.1-4.613.6 (16%)
PR255pyrrole scarletDaniel Smith-0.3-12.6-5.013.3 (15%)
PV14cobalt violetHolbein8.4-2.32.310.7 (22%)
PV15ultramarine violetWinsor & Newton1.7-5.1-0.95.8 (14%)
PV16manganese violetWinsor & Newton3.8- (27%)
PV19quinacridone redDaniel Smith1.2-9.6-3.49.9 (16%)
PV19quinacridone roseWinsor & Newton0.8-9.7-1.89.9 (15%)
PV19quinacridone violetWinsor & Newton5.3-11.9-4.113.2 (35%)
PV23dioxazine violetWinsor & Newton7.7-9.8-1.414.0 (61%)
PY3 hansa yellow lightDaniel Smith-1.8-3.5-1.44.1 (6%)
PY35cadmium lemonWinsor & Newton-1.9-2.3-2.74.0 (5%)
PY35cadmium yellowWinsor & Newton-4.5-17.3-7.519.2 (23%)
PY35cadmium yellow deepWinsor & Newton-4.7-16.2-10.719.2 (25%)
PY42gold ochreWinsor & Newton0.7-7.6-0.38.7 (11%)
PY43yellow ochreWinsor & Newton-0.3-5.6-1.66.0 (9%)
PY53nickel titanate yellowWinsor & Newton-2.9-0.9-2.85.9 (6%)
PY65hansa yellow deepDaniel Smith-1.6-15.9-4.716.7 (17%)
PY97hansa yellowDaniel Smith-2.0-12.2-3.912.9 (14%)
PY129green goldWinsor & Newton-0.4-21.6-3.421.7 (30%)
PY150nickel azomethine yellowDaniel Smith-0.9-14.3-2.314.4 (18%)
PY153nickel dioxine yellowWinsor & Newton-0.7-5.9-1.86.2 (8%)
PY154benzimidazolone yellowWinsor & Newton-0.8-2.3-1.33.6 (4%)
PY184bismuth yellowWinsor & Newton-1.8-2.0-2.63.5 (5%)
Average2.3-10.7-2.912.9 (27%)
Sort by: Lightness ShiftChroma ShiftHue ShiftTotal ShiftTotal Shift % • Color Index Name


Last revised 08.01.2005 • © 2005 Bruce MacEvoy