I am currently undertaking a research on the Lake Bosomtwe in Ghana using the recreation model. However, all results return the error message. I kindly request some assistance on input parameters for the excel file and any help that would enable me achieve the desired outcome. Kind regards, Blankson.
Hi @Bobbi -
What is the error message? Please post the whole log file (.txt) that the model produces in your Workspace folder, so we have some information to work with. Thanks.
~ Stacie
1 Like
Pls find attached a copy of the uploaded shapefile for your consideration.
Robert E. Blankson
Lands Commission
P. O. BOX CT 5008
Cantonments
Accra
Ghana
Tel : .+233 209 242444/ +233 277 171541
E-mail : reblankson@landscom.gov.gh
P Kindly, consider the Environment before printing this e-mail.
Save Paper, Save Trees… Save our Planet…!
Lake_Bosomtwe.zip (1.57 KB)
STUDY AREA.zip (4.02 KB)
Hi Robert -
Those seem to be shapefiles of your study area. But in order to help explain the error message, we need to see the log file created during the model run, which will list your inputs, model version and all of the messages produced while the model is running, including the specific error message. It should be in your Workspace, and named something like InVEST-Habitat-Quality-log-2019-08-07–15_29_58.txt.
~ Stacie
I was able to run the software later on after sending the email. However, there are images taken that don’t show up in the results on return I need some assistance with that now.
Robert E. Blankson
Lands Commission
P. O. BOX CT 5008
Cantonments
Accra
Ghana
Tel : .+233 209 242444/ +233 277 171541
E-mail : reblankson@landscom.gov.gh
P Kindly, consider the Environment before printing this e-mail.
Save Paper, Save Trees… Save our Planet…!
Please find attached results in the zipped folder for your kind consideration. The analysis was able to complete itself however, regarding the area in question, there is no data on images in the results table generated especially when the shapefiles are imported to ArcMap. What could be the cause?
pud_bosomtwe (2).zip (6.2 KB)
monthly_table_lake.zip (898 KB)
Thanks for sharing your data. I downloaded your results and I don’t see anything unexpected. There are some small PUD values in the tables, indicating that some images were found within the polygons. The density of images varies widely across the globe so there are no guarantees about the volume of data you will find in a given place.
Ok sir. Thanks and gratitude in abundance for the feedback. I’m truly grateful. Kind regards, Blankson.
Robert E. Blankson
Lands Commission
P. O. BOX CT 5008
Cantonments
Accra
Ghana
Tel : .+233 209 242444/ +233 277 171541
E-mail : reblankson@landscom.gov.gh
P Kindly, consider the Environment before printing this e-mail.
Save Paper, Save Trees… Save our Planet…!
Hi please I humbly request further assistance with the parameters to enter into the excel file if I need to compute regression for the same area under consideration. Kind regards, Blankson.
In addition, how do I interpret the results obtained? I am working on the impact on tourism on lulc within the study area as well as an assessment of recreational ecosystem value. Any assistance would be appreciated.
Robert E. Blankson
Lands Commission
P. O. BOX CT 5008
Cantonments
Accra
Ghana
Tel : .+233 209 242444/ +233 277 171541
E-mail : reblankson@landscom.gov.gh
P Kindly, consider the Environment before printing this e-mail.
Save Paper, Save Trees… Save our Planet…!
Hi there,
Are you asking about interpreting the results in regression_coefficients.txt, which is an output you get when using the compute_regression option?
The recreation model is designed to answer these basic questions:
- where are people visiting? This is captured in the
pud_results.shpoutput. - which features of the natural/built environment are influencing the spatial patterns seen in
pud_results.shp?
The answer to question 2 is in the interpretation of regression_coefficients.txt. The model is a linear regression model, so I would encourage you to seek out a good source for understanding linear regression estimates (coefficients) and errors. Very briefly, the estimate describes the direction (positive/negative) and magnitude of the effect that predictor has on the response variable, which in this case is (log-transformed) average PUD.
Here’s also an excerpt from a related post:
“The coefficient values in regression_coefficients.txt should change with different sets of predictors. These coefficients are a way to measure the effect of a predictor on the existing visitation patterns. If you wish to see how visitation patterns might change under a scenario, first you’ll need to compute a regression that models the existing relationship between the predictors and the PUD patterns, and then you could use that regression model to estimate changes in the PUD patterns given changes in the predictor datasets. That would require evaluating the regression results in regression_coefficients.txt and then once happy with them applying the Scenario Predictor Table option.”
Thank you for the feedback. I read the data requirements for input in excel to obtain the regressional output. My assistance also bothers on the parameters and how they are entered in excel to obtain the desired results. The sample that comes with the installation is not self explanatory. Kind regards, Blankson.
Okay here is the relevant section of the User’s Guide for the Predictor Table input (http://releases.naturalcapitalproject.org/invest-userguide/latest/recreation.html#data-needs – see item #6). Please ask specific questions if this information is not clear.
Are you getting some errors when you run the model? If so please upload the log (e.g. InVEST-Recreation-Model-log-2019-01-28--14_56_42.txt) from your output workspace so we can see exactly what the issue is. You could also upload the CSV file itself.
Kindly find attached log file and default csv file for your reference.
InVEST-Recreation-Model-log-2019-09-19–02_14_48.txt (7.25 KB)
predictors.csv (273 Bytes)
Following up on the earlier mail sent with the log and csv file, I also attach a copy of predictors and request assistance as to how they should be entered in the predictor file for upload and running.
(Attachment Recreational model predictors.docx is missing)
Following up on the earlier mail sent with the log and csv file, I also provide some indicators/predictors and request assistance as to how they should be entered in the predictor file for upload and running on the software. They are: recreational activities; fees; area; boating. Kind regards, Blankson.
Hi there,
It appears that you are using the sample data for the predictors and the predictor table. That sample data is only suitable for one very specific area in The Bahamas. You may wish to use it along with the sample data AOI. If you wish to run the model in a new location, you will need to collect your own GIS data to serve as predictors in your study area and then construct a new predictors.csv table.
Here is an excerpt from the User’s Guide that may help:
" We find that it often helps to consider at least one predictor variable from several main categories: natural capital (eg habitats, lakes), built capital (eg roads, hotels), industrial activities, and access or cost (eg distance to major airport). Often, single variables representing each of these categories can explain the majority of variation in photo-user-days. This tool allows users to provide predictor variables in GIS formats and to process those layers in a variety of ways (described in the Data Needs section)."
Since you previously posted your pud_results.shp, I should also mention that regression results will only be meaningful if there are a relatively large number of polygons with PUD values relative to the number of predictors.
1 Like
Thank you very much for the feedback. Truly grateful.
Robert E. Blankson
Lands Commission
P. O. BOX CT 5008
Cantonments
Accra
Ghana
Tel : .+233 209 242444/ +233 277 171541
E-mail : reblankson@landscom.gov.gh
P Kindly, consider the Environment before printing this e-mail.
Save Paper, Save Trees… Save our Planet…!