Welcome to the Earthdata User Forum! Here, subject matter experts from several NASA Distributed Active Archive Centers (DAAC) can discuss general questions, research needs and data applications. Users can query how to access, view and interpret the data.
by aklaboss » Sun Aug 01, 2021 2:29 pm America/New_York
I have been working with monthly POC data (2003-2015) from MODIS Aqua at 4km resolution (an example: A20030322003059.L3m_MO_POC_poc_4km.nc) and have been noticing a lot of negative values, particularly in coastal values which are of interest to me. I was reading a previous forum post (https://oceancolor.gsfc.nasa.gov/forum/ ... c%20values) regarding a similar issue, but I was unsure how to solve the issue from this information. The issue can't be solved with by just applying a scaling factor because there is a mix of positive and negative values in the same monthly file. Does the scaling factor only need to be applied to the negative values? Or are the negative values just bad data points?
The minimum POC value in that file is 8.3999. The data are stored as a scaled integer and must have the scaling applied to get the geophysical value. For this file the scale_factor is 0.2 and the add_offset is 6400. Most netCDF readers should apply this for you, but if yours does not, you would need to multiply the scaled integers by 0.2 and add 6400 to get the geophysical value.
The _fillValue is -32767 and should not be converted to a geophysical value, as it is just identifying pixels with no data...converted it would come out to be -153.4 - which is a non-physical value
by aklaboss » Mon Aug 02, 2021 4:14 pm America/New_York
Thank you for the quick response! My netCDF reader does appear to be scaling it in automatically, but I am still getting negative values that are not no data points. I think the previous file I was referring to may not have any issues with it, but this data file does. I am looking at A20031522003181.L3m_MO_POC_poc_4km.nc and getting a minimum value for the file, after scaling, to be -4119.80029296875. There is a patch of nonnull negative values in that location (around 74.1875,68.8125). Looking at satellite images from of that location in that time frame, it is over an ocean, but may have ice coverage. I am wondering if the ice could be causing those nonnull negative values.
Something isn't right with the way you are reading the data.
The minimum unscaled (non-fill) value in that file (A20031522003181.L3m_MO_POC_poc_4km.nc) is -31943, which when converted to the geophysical value is 11.4 mg/m^3...which is the minimum POC value I found in the file. Even if you were to scale the fill value (-32767), it would only be -153.4, which is nowhere near the -4119.80029296875 you report. The values around 74.1875N,68.8125E in that file are all fill values.