About Landscape Context Index

villca01villca01 Member
edited September 2016 in General
Hello Guys,

I'm trying to understand the process to obtain the Landscape Context Index (LCI), the manual said: "The Landscape Context Index represents the level of connectivity
of a natural ecosystem patch to other patches. It is calculated as the fraction of
natural ecosystem within a 500 m buffer strip around a particular patch".

I understand that if I have a parch "B" I have to create a buffer of about 500m and then inside that buffer i have to find the same parch B so LCI will be the division of both (area of the parch founded inside the buffer with the total area of buffer).

I hope your help. Thank you.
Post edited by villca01 on

Comments

  • jdouglassjdouglass Administrator, NatCap Staff
    Hi @villca01,

    The LCI is only calculated for the buffered band around parcel B, which does not include the area of parcel B itself.  The LCI is the proportion of the area of this 500m band surrounding parcel B that overlaps with other natural parcels.

    In pseudocode, the LCI calculation would look like this:

    patch_B = some polygon representing a natural parcel
    remaining_LCI_buffer = buffer_by_500m(patch_B) - patch_B
    original_LCI_buffer_area = remaining_LCI_buffer.area
    for each parcel that intersects with remaining_LCI_buffer:
        remaining_LCI_buffer = remaining_LCI_buffer - parcel

    patch_B_LCI = 1 - (remaining_LCI_buffer / original_LCI_buffer_area)

  • Thank you @jdouglass,

    You give me
    lights to understand LCI, but to clarify the idea, if I have patch_C and D that overlaps with the buffer of parcel B, the LCI of patch_B is:

    patch_B_LCI = 1- [(remaining_LCI_buffer- (patch_C + patch_D))/original_LCI_buffer_area]

    I hope your help. 

    Thank you very much.

  • jdouglassjdouglass Administrator, NatCap Staff
    Almost ... the numerator (remaining_LCI_buffer - (patch_C + patch_D)) is the area remaining in remaining_LCI_buffer and not the geometry itself.

    I've attached an image that provides an example, just for the sake of clarity.  In the Patches chart, the grey parcel is the parcel for which we're trying to find the LCI and the blue parcels are nearby natural parcels.

    In the LCI chart, we can see the buffer around the grey parcel, and the area of the buffer that intersects the other two natural parcels has been removed, resulting in an LCI of about 0.083.


    656 x 385 - 23K
  • thank you very much, so when i have a case like the picture that i've attached the LCI is 1? and what is the bibliography that supports that.

    Thank you very much for your
     cooperation and I apologize for the incovenience.

    1012 x 600 - 16K
  • jdouglassjdouglass Administrator, NatCap Staff
    Sorry about the delay!  Assuming that Parcels A, C, and D are all natural parcels, then the LCI for parcel B should be 1.

    The Landscape Context Index calculations are taken from the Colombian national compensation policy.  I'd recommend taking a look at the paper that introduces OPAL (http://www.sciencedirect.com/science/article/pii/S1364815216302110) for a full citation.

    Hope this helps!
  • Thank very much, my doubts shall be clarified.

  • Hello @jdouglass,

    I need your help you, i have another question

    How work OPAL when has a case like the image attached, two parcels very close (both parcels B) even I have in others zones patch closer. 

    Does OPAL join when you have patches very close?

    Thank you.
    2094 x 1200 - 350K
  • jdouglassjdouglass Administrator, NatCap Staff
    Hi @villca01, In this case, Parcel B is divided into two separate polygons, so OPAL treats them as separate parcels (parcel B1 and parcel B2).  Each parcel will have its LCI calculated based on the natural and non-natural parcels nearby.

    In this case, parcel B1 will have an LCI of 1 because parcels A, C, D are all natural parcels.

    Parcel B2, however, will have a lower LCI because of the location of a non-natural parcel within its 500m buffer.
    2094 x 1200 - 506K
Sign In or Register to comment.