# Scaling the variance plane and accounting for covariance

**URL:** <https://www.rubin.community/t/scaling-the-variance-plane-and-accounting-for-covariance/5752>\
**Category:** Data Management\
**Created:** [August 3, 2021, 1:36pm UTC](https://www.rubin.community/t/scaling-the-variance-plane-and-accounting-for-covariance/5752 "2021-08-03T13:36:39Z")\
**Posts on this page:** 1\
**Showing post:** 3

<div class="post-metadata">

**Author:** ![raphaelshirley](https://sea2.discourse-cdn.com/flex002/user_avatar/www.rubin.community/raphaelshirley/32/1230_2.png) [@raphaelshirley](https://www.rubin.community/u/raphaelshirley)\
**Post date:** [August 3, 2021, 4:31pm UTC](https://www.rubin.community/t/scaling-the-variance-plane-and-accounting-for-covariance/5752/3 "2021-08-03T16:31:33Z")

</div>

Here are the detected pixels in blue over the image for various detectCoaddSources config values

 ![Screenshot 2021-08-03 at 17.37.38](https://us1.discourse-cdn.com/flex002/uploads/rubin/original/2X/3/3b965153eef61b06453bec9d2d686723e2400dde.jpeg)  
 ![Screenshot 2021-08-03 at 17.37.45](https://us1.discourse-cdn.com/flex002/uploads/rubin/original/2X/a/afe8156f703c0fe2d5bb8d19c1b08ec49e772814.jpeg)  
 ![Screenshot 2021-08-03 at 17.37.53](https://us1.discourse-cdn.com/flex002/uploads/rubin/original/2X/4/49799ce02421a06a38086c0d6477bf810d2c6b03.jpeg)  
 ![Screenshot 2021-08-03 at 17.38.01](https://us1.discourse-cdn.com/flex002/uploads/rubin/original/2X/1/1ade1de8a980d65d96be08d0b73b48614c6d9792.jpeg)

For this reason we settled on stdev and 10 sigma based on the highly unscientific criteria of the detection planes ‘looking’ sensible. We are also considering simply including a threshold multiplier based on our measurements of the autocorrelation matrix sum.

---

_[View the full topic](https://www.rubin.community/t/scaling-the-variance-plane-and-accounting-for-covariance/5752)._