Why the Lsum,avail,i of some pixels are negative?

Hello, I’m using the seasonal water yield model, and the Lsum,avail,i of some pixels are negative. What this means? I think the Lsum,avail,i should be positive because a pixel couldn’t recharge the upslope pixels.

Hi @Annabelle , your question makes a lot of sense to me, I cannot easily explain the negative values.

If I use the sample data we provide with this model, I see that L_sum_avail.tif has some values that are technically negative, but they are effectively equal to zero (for example, -0.000015) and are likely only negative due to the imprecision of floating point arithmetic.

Can you verify that the values in your raster are significantly large negative numbers? Sometimes GIS tools choose arbitrary cutoffs when displaying a legend like in your screenshot. Can you report some actual pixel values, such as using the “identify” tool in GIS? Thank you,

And if the values are negative, would you mind sharing your entire output workspace so we can look into this further? You could zip it and share a link via google drive, dropbox, etc (feel free to share in a private message to me if you prefer not to post your data here publicly).

1 Like