Large State Project in Redwood Forest, Expert Advice needed

PIX4Dmapper
1

Hello all and thanks for your time. I need to have some expert advice to properly set up a large project I have beginning this week. Our organization has done 100 or so projects per year for the last 3 years, using Pix4d aerial data and Emesent’s Hovermap system, as a handheld unit. This methodology works quite well for us to gather complex data sets for topographic survey purposes. The workstations handling the point clouds do well with clouds of up to about 150,000,000 points.

These projects are typically fairly small, averaging an acre or 2, and we have had several 40 Acre parcels, and a few at 50 and 60 Acres. We have accepted a project from the California Department of Conservation mapping about 350 Acres of Redwood forest that has been affected by the 2020 wildfires. Complicating matters is the fact that the centerline of our survey swaths is a creek, down in a gully, surrounded by trees of up to 250 feet in height, some of the largest trees in the world.

This project contains a highly specific group of requirements for gathering, processing, and reporting, which takes us well outside of our comfort zone. These requirements, though expansive as they are, represent standard working parameters for many. For those people, I have a number of questions to help with the planning of this work.

The attached map shows a mock-up of the site taken from google imagery, with the proposed tentative statitioning of the three flight lines. My main issue in this intial stage is to figure out the area of each flight for both P4D and Lidar.

With P4D, I am unsure of how many linear feet of the alignment centerline can be gathered per flight, and also how to gather that with those trees in the area.

The camera to be used is the Zenmuse P1, on an M-300 RTK Drone. The flight planning is happening inside DJI’s pilot 1 app.

The state requires aerial imagery to have a Ground Pixel Resolution and Ground Sampling Distance of .08’. Given the inordinate constraints outlined here. What sort of flight area, speed, and duration would be appropriate? How many flights might we expect, and how do you best gather orthophoto data from a redwood forest? Does this preclude any automated flying?

Thank you all very much

COVER - Forum.pdf (225.0 KB)

The stations are on 500 foot major and 100 foot minor stations

2

Hi candarco,
I can’t speak necessarily to what flight apps you should use but I can speak to the issues we routinely see from a processing point of view. First, keeping your image overlap consistent will be key. Projects that include heavily vegetated areas that also have a significant amount of elevation changes can be challenging. I would highly recommend using an automated flight app as freeflying this will almost certainly degrade your overlap. In fact I would say that this project would preclude freeflying. I would also use some type of terrain awareness in the flight app. If the terrain abruptly changes and the image overlap decreases then you will have problems. I would also err on the side of more overlap than less as most people underestimate the amount of required overlap. It’s always hard to say exactly how much you need but I would keep it between 80-90%, not to exceed 90. I hope this helps.