Use of lidar constant

[HolgerPollyNet]

Adi:

1. Is there any usage of the lidar calibration? I realize now that it is not used for aerosol retrieval in the Klett approach, or am I missing something?

2. Is there any process that extracts the calibration (LC) for the case of overlap correction? Is there a way to calculate and save it?

Besides the questions above, when using the overlap correction mode with ALiDAn's generated overlap function, I noticed some difficulties and fixed them:

When choosing the followings:
overlapCalMode = 0
overlapCorMode=1
then, normRange is an empty array (due to earlier initializations), so the correction is not applied. (This is for a case of loading a pre-calculated/known Overlap function from a file.)

I fixed this in picassoProcV3 in the section of overlap correction with the following code:

if PollyConfig.overlapCorMode==1 && PollyConfig.overlapCalMode==0 olAttri355.normRange = [0 PollyConfig.heightFullOverlap(flag355FR)]; olAttri532.normRange = [0 PollyConfig.heightFullOverlap(flag532FR)]; olAttri1064.normRange =[0 PollyConfig.heightFullOverlap(flag1064FR)]; end
Additionally, notice that the following line returns an error for overlapCorMode=1:
sigOLCor = olCor(sigFR, transpose(olSm), height, height(p.Results.normRange));
https://github.com/PollyNET/Pollynet_Processing_Chain/blob/b4f0cb38613f1059cf0e9cb2b60211e62e59bdb4/lib/qc/pollyOLCor.m#L107-#L108
I fixed this with the following line:
sigOLCor = olCor(sigFR, transpose(olSm), height, p.Results.normRange);
as OlCor expects that normRange is a two elements array.

[HolgerPollyNet]

@ZPYin:
1: This are many useages of the lidar calibration results. Lidar calibration is the core of quantitative results. It was used in the retrieval of high-resolution products, i.e., attenuated backscatter, quasi backscatter and target classification (

Pollynet_Processing_Chain/lib/interface/picassoProcV3.m

Lines 3871 to 3958 in b4f0cb3

	%% attenuated backscatter
	print_msg('Start calculating attenuated backscatter.\n', 'flagTimestamp', true);
	flag355 = data.flagFarRangeChannel & data.flag355nmChannel & data.flagTotalChannel;
	att_beta_355 = NaN(length(data.height), length(data.mTime));
	if (sum(flag355) == 1)
	att_beta_355 = squeeze(data.signal(flag355, :, :)) .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed355;
	att_beta_355(:, data.depCalMask) = NaN;
	end
	flag532 = data.flagFarRangeChannel & data.flag532nmChannel & data.flagTotalChannel;
	att_beta_532 = NaN(length(data.height), length(data.mTime));
	if (sum(flag532) == 1)
	att_beta_532 = squeeze(data.signal(flag532, :, :)) .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed532;
	att_beta_532(:, data.depCalMask) = NaN;
	end
	flag1064 = data.flagFarRangeChannel & data.flag1064nmChannel & data.flagTotalChannel;
	att_beta_1064 = NaN(length(data.height), length(data.mTime));
	if (sum(flag1064) == 1)
	att_beta_1064 = squeeze(data.signal(flag1064, :, :)) .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed1064;
	att_beta_1064(:, data.depCalMask) = NaN;
	end
	flag387 = data.flagFarRangeChannel & data.flag387nmChannel;
	att_beta_387 = NaN(length(data.height), length(data.mTime));
	if (sum(flag387) == 1)
	att_beta_387 = squeeze(data.signal(flag387, :, :)) .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed387;
	att_beta_387(:, data.depCalMask) = NaN;
	end
	flag607 = data.flagFarRangeChannel & data.flag607nmChannel;
	att_beta_607 = NaN(length(data.height), length(data.mTime));
	if (sum(flag607) == 1)
	att_beta_607 = squeeze(data.signal(flag607, :, :)) .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed607;
	att_beta_607(:, data.depCalMask) = NaN;
	end
	flag355 = data.flagFarRangeChannel & data.flag355nmChannel & data.flagTotalChannel;
	att_beta_OC_355 = NaN(length(data.height), length(data.mTime));
	if (sum(flag355) == 1)
	att_beta_OC_355 = sigOLCor355 .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed355;
	att_beta_OC_355(:, data.depCalMask) = NaN;
	end
	flag532 = data.flagFarRangeChannel & data.flag532nmChannel & data.flagTotalChannel;
	att_beta_OC_532 = NaN(length(data.height), length(data.mTime));
	if (sum(flag532) == 1)
	att_beta_OC_532 = sigOLCor532 .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed532;
	att_beta_OC_532(:, data.depCalMask) = NaN;
	end
	flag1064 = data.flagFarRangeChannel & data.flag1064nmChannel & flagTotalChannel;
	att_beta_OC_1064 = NaN(length(data.height), length(data.mTime));
	if (sum(flag1064) == 1)
	att_beta_OC_1064 = sigOLCor1064 .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed1064;
	att_beta_OC_1064(:, data.depCalMask) = NaN;
	end
	% flag387 = data.flagFarRangeChannel & data.flag387nmChannel;
	% att_beta_OC_387 = NaN(length(data.height), length(data.mTime));
	% if (sum(flag387) == 1)
	% att_beta_OC_387 = sigOLCor387 .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed387;
	% att_beta_OC_387(:, data.depCalMask) = NaN;
	% end
	% flag607 = data.flagFarRangeChannel & data.flag607nmChannel;
	% att_beta_OC_607 = NaN(length(data.height), length(data.mTime));
	% if (sum(flag607) == 1)
	% att_beta_OC_607 = sigOLCor607 .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed607;
	% att_beta_OC_607(:, data.depCalMask) = NaN;
	% end
	flag355NR = data.flagNearRangeChannel & data.flag355nmChannel & data.flagTotalChannel;
	att_beta_NR_355 = NaN(length(data.height), length(data.mTime));
	if (sum(flag355NR) == 1)
	att_beta_NR_355 = squeeze(data.signal(flag355NR, :, :)) .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed355NR;
	att_beta_NR_355(:, data.depCalMask) = NaN;
	end
	flag532NR = data.flagNearRangeChannel & data.flag532nmChannel & data.flagTotalChannel;
	att_beta_NR_532 = NaN(length(data.height), length(data.mTime));
	if (sum(flag532NR) == 1)
	att_beta_NR_532 = squeeze(data.signal(flag532NR, :, :)) .* repmat(transpose(data.height), 1, length(data.mTime)).^2 / LCUsed.LCUsed532NR;
	att_beta_NR_532(:, data.depCalMask) = NaN;
	end
	print_msg('Finish.\n', 'flagTimestamp', true);

).
2. As I checked, there is not such a process yet. You can of course implement it for your research by extenting the lidar calibration with overlap corrected extinction and backscatter.

[HolgerPollyNet]

Hey Adi, @cristoferjimenez has worked on the improvement of the overlap function and we made a recent update:
Overlap_Raman

Does the issue still exit?