Hi
When I am running the change allocation module in land change modeler I receive this error message:Number of band images must be 3. Could you help me please? Thanks
Hi
When I am running the change allocation module in land change modeler I receive this error message:Number of band images must be 3. Could you help me please? Thanks
We do not fully understand the issue you are experiencing. Could you please provide more details about the steps you took leading up to encountering this error message and they type of data you were using.
Thank you.
Laura, Clark Labs Technical Support
Hello, I just attempted to run the Land Change Modeler (prediction tab) and I got a similar message: "Error in MLP: number of band images must be 5". i'M not sure either of what this means, but i am trying to predict two transitions:
Secondary forest to Cropland AND secondary rainforest to grassland. I want to identify possible future areas of agricultural expansion.
In order to do this i chose to create 4 different transition potential maps (in the transition potential tab), because I identified 4 different systematic transitions thanks to Pontius et al., 2004 formula. I understood i had to create a transition map for each systematic transition. Anyway, when used the Prediction tab, I ended up choosing only the two transitions of the 4 I first identified, and when i ran the model, then I got the message. I don't know if this explanation makes the issue any clearer.
Cheers
On a separate comment, when I tried to model only one transition (secondary forest to cropland), then it worked...I haven't validated this result but it seems strange that I could do it with one transition instead of the four I wanted to model. I have a feeling it has to do with the fact that I created the transition potential maps individually, that is, one by one (I changed slightly the explanatory variables for each case), that way my accuracy rates went up, above 70% instead of 30% that was the value I always got when I tried to create them in a single step.
Maybe the answer is somewhere hidden in these previous steps....
Cheers again
Hi Gabriela and Oumar,
I suspect that in the process of testing many different submodels LCM has become confused. Please note the appropriate driver variables and submodels that you wish to run and create a new project. In this new project use the driver variables and submodels that you had settled on without testing other combinations and let us know how it goes!
Best,
Andrew Hostetler
Clark Labs Technical Support