After applying RUSLE and doing all calculations for soil erosion assessment we get to know that there is no actual sediment data available, collected by authorities or any other researcher for the watershed. How can I validate my results ?
You are the first evaluator. you have to answer the question of whether the erosion model estimates are reasonable? Secondly, if the mechanical operation of soil protection is done like check dams, you can use the sediment data collected behind it. third, you can use the data of the nearest hydro metric and sediment station by expanding the relationships to the watershed you are studying.
To elaborate on what Reza Sokouti posted, if there is no sediment data that has already been collected, the validation process will take several years to complete because a single year's sediment data, no matter how you collect it, cannot be assumed to be representative of of a long-term average. I say this assuming that you used average climate/weather values in your erosion assessment. If you were to collect site specific data of rainfall etc. at the same time as collecting the site specific sediment production data, the results of your model using that input data could be used to begin to validate the model when run using average, or long-term, values.
At the risk of sounding pedantic, I would add that one should be careful about saying/writing "validate Soil erosion assessment data (tons/hact./yr) calculated by RUSLE." The results of a model should not be thought of as "data." Models produce estimates or projections. Data is measured/observed. This is a common trap for people to slip into.
Tenga en consideración que los modelos son hipótesis teóricas y TIENEN que validados en la Realidad;
Los modelos son para usarlos, no para creer en ellos.
I would also like to add that the application of the model in a specific place/site/ watershed should be done with the measured data in long term, here are 6 factors, and it should also be noted that the results of the RUSLE model are an estimate of the actual data based on the accuracy of the model , up to 75% can be different from real data. This means that in most cases you can not do any validation with statistical methods .
Article Mapping potential soil erosion using RUSLE, Remote Sensing, ...
Chapter Identification of the relationship between temporally varyin...
Article Assessment of Water-Induced Soil Erosion as a Threat to Natu...
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Programming Languages
- Python