Відео

You saved me tons of work! thank you!

Godlike work

Excellent, thanks Hector. Good work to you! Always looking to improve.

I think UA-cam removed your link. But I will post it in the video description.

Hey thanks Julian. Appreciate the feedback. I would need to see a screenshot or more info to help. Have you correctly entetered your username and password for the Sentinel imagery? It’s the same credentials as if you registered on the sentinel hub. But I’ll do one better. The guy who made SCP is awesome. He has a Facebook page where people can get help. Just post a comment with a screenshot and they can start to help. facebook.com/SemiAutomaticClassificationPlugin

Thank you for the very shift response! I'm new to QGIS and remote sensing and people like you really make the learning process enjoyable. It was the login for the Sentinel imagery that did the trick for me. I'm considering a fire in New South Wales. If I have questions along the way, can I return to this thread? :-)

@@JulianRoinSkovhus absolutely mate. Best of luck

@@spatial-exploration1478 Hi again. Using your video I have managed to get some pretty nice results. Thanks again! However, the fire scar I found is not as 'clear' as the one in Broome. Even with a shape file of the fire scars in NSW I find it difficult to locate coherent fire scars that make for good burn severity maps and calculation of burn area. I was wondering if you have any tips for choosing the fires? Or maybe you have analyzed one in NSW/Victoria during the Black Summer that could make for a good beginner-level analysis without domain knowledge on terrain and biomass?

@@JulianRoinSkovhus could be a couple of things. Different vegetation might have different intensities, but if you’re looking at those fires from last season I’d imagine you’d be seeing something. Could be the ranges are significant but your an order of magnitude out due to a small type-o in the ratio.

Hi, great question. No I did not address atmospheric correction in this video as it was more about using the basic raster tools in QGIS. The Sentinel product used in the example was L1C (which is what you get from EarthExplorer). L1C is top of atmosphere (TOA) reflectance with radiometric and geometric corrections. If I was using this product I would instead use the SCP plugin in QGIS to apply the DOS corrections. Alternatively, if I had used SCP in QGIS to download the imagery, which is my normal process, I could access the L2A product which is bottom of atmosphere (BOA) reflectance.

@@spatial-exploration1478 Thank you for the reply! Does the L2A product not need to be corrected? I am new to QGIS and quite like the options for pre-processing but I am still learning about correcting images. I use Sentinel 2 imagery almost exclusively.

Thanks Leslie. Glad it helped. This is just the tip of the iceberg. There is so many multispectral analyses you can easily implement in QGIS. cheers

Thanks Sanchi. I appreciate the positive feedback. It makes the time I put in worthwhile. 👌

Hi Diana, cool. So do are you hoping to make an interactive map that the user can switch between layer themselves, or just setting up a QGIS project yourself for your own analysis? If it’s the first one, I reckon the plug-in QGIS2Web will give you that interaction. If it’s just for yourself, it should be fairly straightforward. There are a few ways you can do it using your original .xlsx file. You can import this using the New Virtual Layer tool in QGIS3. Layer > Create Layer > New Virtual Layer Then it’s a case of working through the later and identifying the different data you want to present for each time period. One tip - if you have a grid of temperature data points, it might be more intuitive for the end user to view this as an interpolated raster. The brain is much better at interpreting patterns and changes in this. Sounds like a cool project. Feel free to ask more questions and I’m sure either myself or any of the other QGISers can help.

@@spatial-exploration1478 Hi, thank you so much. I'm a beginner in QGis. I would like to make an interactive map and to be able to filter by layer or more combinations between them. Although I have a limited time to make it. I have a temperature dataset over the next 4 hours(every 10 minutes) these are the result of the temperature prediction. So I need to show the combinations of all layers and filter by layer between them but I just need to show two or three possible combinations. I can show the results about the combination layers like an interactive map or setting up a QGis project to save the results and show them in my project, no problem. I have the location of some weather stations each one with a temperature dataset. I read about raster interpolation to create a surface of points that are very close. is this the first step to make a layer? Thanks in advance for your help!

@@dianaramirez3508 if your data is collected at discrete locations (e.g. weather stations dotted here and there across a wide area), it would not be suitable for raster interpolation. Actually it makes your objective a bit easier. So you have a map overlaid with points, each point is the location of a weather station. And you want to be able to change the date of the observations at all weather stations in one go. And all of this nicely packaged into a web map. QGIS2Web can definitely do this for you. I can try and put up a video over the weekend with a really basic version of what you want.

@@spatial-exploration1478 Thanks. I have a file with temperatures and coordinates from weather stations at 10 minutes intervals time over the next 4 hours. I need to make a layer from every time interval from set weather stations and to filter by layer and also select one or more combinations to get a map. I made a layer selecting features first group 10 minutes interval time. But I'm not sure about raster interpolation how to make it and show the map. Thanks in advance for your help! I share the file drive.google.com/file/d/1GvgZxEYAhwH--gYfrPzuXRNcZkSynX3s/view?usp=sharing

@@spatial-exploration1478 I have a Land Use and Cover Data, Slope and Distance to Road Data Layer. I am trying to sample random points with these spatial variables. I tried the point sampling tool but that didn't work. How can i carry that out please.

Thanks Cronje. Appreciate it. Good luck with your analysis.

Rohit Dhankar thanks, appreciate it

Thanks Gordon, I really appreciate the feedback. Certainly makes the effort that bit more worthwhile.

only Kosmos Terima kasih. Are you having problems with anything in particular? Happy to help. Cheers

Thanks James. I guess it depends on your research question, and at what scale (spatial and/or temporal) you want to detect any changes? If you had access to basic biodiversity indices (diversity, richness, evenness etc), you could use these as a dependent variable in a model with parameters derived from remote sensing. If you get a consistently good fit across tour study areas you could use this information to infer biodiversity change from remotely sensed data elsewhere (but with heaps of caveats for unexplained variation). I’m Interested to hear more about your work.

christopher guth thanks, appreciate the positive feedback mate.

Jamie Waller sorry I meant to provide this ages ago but got distracted. I coded it in R, so I’ll see if I can share my script. When I do I’ll put it in the description. Thanks

Anass AINOUCHE your welcome, thanks for the comment.

Felipe Bastos thanks for the comment 👍

Thanks Emma. Couple of possible issues - The elevation model you are using (i.e. the DEM); is the coordinate reference system (CRS) unit in degrees or metres? It needs to be the latter (check by right clicking on layer then Properties>Information). If its in degrees you simply save a new copy of the DEM and pick a CRS that has meters as its unit (e.g WGS84 - Pseudo Mercator). Then continue your analysis with this new layer. There are probably more appropriate CRS for your area of interest so either google it or keep an eye on the little red map in QGIS which shows you the geographical area for which a CRS is valid. - Are you converting your hillshade to slope or your original DEM/elevation model to slope? It should be the latter. -Otherwise maybe check the properties of the slope layer you created (Layer properties>Information) and check the minimum and maximum values here (should be between 0 - 90 degrees or 0 - 100%). If the min-max looks OK then just readjust the values to match in the symbology. I don't think this is your case but worth a look anyways. Let me know how you go, Cheers, James

@@spatial-exploration1478 Thanks for your response! It was just the trick. I was converting directing from hillshade to slope (not from the original DEM). Do you teach this material, or just work with the software? Looking forward to seeing more of your videos on QGIS!

Great to hear that @@emmaposluns1238. Will have a few new videos coming soon.