Temporal data processing/seasonal aggregation - Revision history

Veroandreo at 22:24, 16 November 2015

2015-11-16T22:24:39Z

← Older revision		Revision as of 22:24, 16 November 2015
Line 87:		Line 87:
	==== Preparation of seasonal granularity dataset ====		==== Preparation of seasonal granularity dataset ====

	Now, we'll create a second time series (it might be raster, vector or raster3d) of seasonal granularity. We'll then take this granularity to aggregate our daily strds. Just as an example of how we would create a vector time series and to save some space, we'll create a spatio-temporal vector data set (stvds). But again, we could also create an ''ad-hoc'' raster time series and it would be equally useful for our objective. So,		Now, we'll create a second time series (it might be raster, vector or raster3d) of seasonal granularity. We'll then take this granularity to aggregate our daily strds. Just as an example of how we would create a vector time series and to save some space, we'll create a spatio-temporal vector data set (stvds). But again, we could also create an ''ad-hoc'' raster time series and it would be equally useful for our objective, even a one-pixel region setting would be fine, since we only need to copy the temporal aggregation/granularity. So,

	<source lang="bash">		<source lang="bash">

Veroandreo: answer to TODO

2015-09-22T12:26:47Z

answer to TODO

← Older revision		Revision as of 12:26, 22 September 2015
Line 87:		Line 87:
	==== Preparation of seasonal granularity dataset ====		==== Preparation of seasonal granularity dataset ====

	Now, we'll create a second time series (might be raster, vector or raster3d) of seasonal granularity. We'll then take this granularity to aggregate our daily strds. Just as an example, we'll create a spatio-temporal vector data set (stvds).		Now, we'll create a second time series (it might be raster, vector or raster3d) of seasonal granularity. We'll then take this granularity to aggregate our daily strds. Just as an example of how we would create a vector time series and to save some space, we'll create a spatio-temporal vector data set (stvds). But again, we could also create an ''ad-hoc'' raster time series and it would be equally useful for our objective. So,

	<!-~~- TODO: why vector dataset?? pls explain - is~~ it ~~to save space? -->~~

	<source lang="bash">		<source lang="bash">

Neteler: subtitles added + TODO

2015-09-22T08:24:36Z

subtitles added + TODO

← Older revision		Revision as of 08:24, 22 September 2015
Line 1:		Line 1:

	=== Seasonal aggregation example ===		=== Seasonal aggregation example ===

			==== Introduction ====

	As "seasons" or "astronomical seasons" are not yet implemented as a predefined granularity for aggregation, the workaround to get seasonal aggregation of a certain time series, is to create another time series with the desired granularity (i.e.: astronomical seasons) and apply or copy this granularity to aggregate data in our time series of interest.		As "seasons" or "astronomical seasons" are not yet implemented as a predefined granularity for aggregation, the workaround to get seasonal aggregation of a certain time series, is to create another time series with the desired granularity (i.e.: astronomical seasons) and apply or copy this granularity to aggregate data in our time series of interest.

			==== Preparation of daily dataset ====

	This is an example showing the use of {{cmd\|t.rast.aggregate.ds}} to aggregate a daily time series using a source time series with seasonal granularity. We first create daily maps for more than one year (400 maps). We then create the corresponding spatio-temporal raster data set (strds) and register the maps into it.		This is an example showing the use of {{cmd\|t.rast.aggregate.ds}} to aggregate a daily time series using a source time series with seasonal granularity. We first create daily maps for more than one year (400 maps). We then create the corresponding spatio-temporal raster data set (strds) and register the maps into it.
Line 80:		Line 84:
	\|		\|
	+----------------------------------------------------------------------------+		+----------------------------------------------------------------------------+

			==== Preparation of seasonal granularity dataset ====

	Now, we'll create a second time series (might be raster, vector or raster3d) of seasonal granularity. We'll then take this granularity to aggregate our daily strds. Just as an example, we'll create a spatio-temporal vector data set (stvds).		Now, we'll create a second time series (might be raster, vector or raster3d) of seasonal granularity. We'll then take this granularity to aggregate our daily strds. Just as an example, we'll create a spatio-temporal vector data set (stvds).

			<!-- TODO: why vector dataset?? pls explain - is it to save space? -->

	<source lang="bash">		<source lang="bash">
Line 160:		Line 168:
	\|		\|
	+----------------------------------------------------------------------------+		+----------------------------------------------------------------------------+

			==== Aggregation of daily to seasonal maps ====

	Finally, we aggregate our daily time series "precipitation daily" into seasonally accumulated precipitation with {{cmd\|t.rast.aggregate.ds}} using method=sum.		Finally, we aggregate our daily time series "precipitation daily" into seasonally accumulated precipitation with {{cmd\|t.rast.aggregate.ds}} using method=sum.
Line 232:		Line 242:
	+----------------------------------------------------------------------------+		+----------------------------------------------------------------------------+

	At the beginning we created 400 maps, remember??? We'll check now that only the maps "contained" in the period start_date-end_date of the stvds are used for the aggregation (the rest are left aside). That's why we use sampling=contains. Check {{cmd\|t.rast.aggregate.ds}} for other method and sampling options.		==== Verification ====

			At the beginning we created 400 maps, remember??? We'll check now that only the maps "contained" in the period start_date-end_date of the stvds are used for the aggregation (the rest are left aside). That's why we use sampling=contains. Check {{cmd\|t.rast.aggregate.ds}} for other method and sampling options. We use {{cmd\|t.rast.series}} for this verification:

	<source lang="bash">		<source lang="bash">

Veroandreo at 13:32, 27 August 2015

2015-08-27T13:32:54Z

← Older revision		Revision as of 13:32, 27 August 2015
Line 2:		Line 2:
	=== Seasonal aggregation example ===		=== Seasonal aggregation example ===

	This is an example showing the use of {{cmd\|t.rast.aggregate.ds}} to aggregate a daily time series using a source time series with seasonal granularity. We create ~~the~~ daily maps for more than one year (400 maps). We then create the corresponding spatio-temporal raster data set (strds) and register maps in it.		As "seasons" or "astronomical seasons" are not yet implemented as a predefined granularity for aggregation, the workaround to get seasonal aggregation of a certain time series, is to create another time series with the desired granularity (i.e.: astronomical seasons) and apply or copy this granularity to aggregate data in our time series of interest.

			This is an example showing the use of {{cmd\|t.rast.aggregate.ds}} to aggregate a daily time series using a source time series with seasonal granularity. We first create daily maps for more than one year (400 maps). We then create the corresponding spatio-temporal raster data set (strds) and register the maps into it.

	<source lang="bash">		<source lang="bash">

Veroandreo: add seasonal aggregation example

2015-08-27T02:33:44Z

add seasonal aggregation example

Show changes

Neteler: Created page with " TODO seasonal aggregation Category: Documentation Category:Temporal"

2015-08-25T21:54:30Z

Created page with " TODO seasonal aggregation Category: Documentation Category:Temporal"

New page

TODO seasonal aggregation

[[Category: Documentation]]
[[Category:Temporal]]