No, we’re not talking about people losing their jobs here. We’re talking about automation being able to assist people to do a better job, and that’s an important distinction.
Previously, we blogged about some work we did for Independence Group NL around the work we did on developing systems and procedures around the management of their disturbance and rehabilitation reporting requirements to the environmental regulator. That project was all about Data Curation, probably what I’d nominate as the second step of the chain – curation of the data. In simple terms there is a three step process for a lot of data projects, namely:
This previous project was all about making sure the data was curated well. What about the other two steps?
In terms of delivering data, proponents are reporting based on either the need to do an Annual Environmental Report (AER), and/or for the Mining and Rehabilitation Fund (MRF). Both of these have specific reporting requirements, and the AERs under the Mining Act 1978 are now being submitted online through the EARS2 system (the same system for the MRF lodgement, too). For other AERs under the Environmental Protection Act 1986, there’s also the need to submit a data package and the transition period for that approach is ending soon, too – so that will shortly be mandatory. As we talked about back in 2017, these sorts of online systems are really starting to streamline the data submission process, and the assessment processes.
Recently, we worked with a mining client to focus on the Data Delivery end of the process, in terms of assessing what they needed to produce – creating some tables that outlined just what data they needed to use. This project had some challenges, mainly due to issues in the data curation step – their data needed quite a bit of work to make sure that the values made sense at the end. One of the things we strive for is accurate reporting on these sorts of datasets. It’s difficult to be accurate if there are issues in the underlying data, and this is often caused by operators not knowing about such seemingly unimportant things as topology (which is why Jake wrote the previous blog about that subject). There are some other things to consider as well in this curation step – for example, making sure that your disturbance polygons are tagged with the same categories that are used in the AER and MRF reporting templates is a must!
Once you have the data prepared you can then do all sorts of things to automate the reporting of the data. In one of our recent projects, we’ve been developing models in ArcGIS that can spit out the data in the right formats ready for compilation into the reporting forms. In historical projects, we’ve also prepared the reports ourselves (including a wide range of illustrative maps) and done a fair bit of work in that area manually. But as the systems like EARS2 evolve, it wouldn’t be a stretch to see a scenario where data is submitted through even more real-time methods, which will be a real challenge for all concerned!
The Data Collection piece remains one that we’re still working on pretty heavily, and that’s still a manual process for most groups. This might entail taking GPS readings, or digitising from imagery (including drone, aerial photography and satellite imagery). We’ve done a few projects attempting to automate the classification of disturbance from imagery over the years, in cooperation with Universities and other research groups, and there is still some way to go to get an all-singing, all-dancing solution here. With the development of other tools like Artificial Intelligence and Machine Learning, though, there are some really good ideas floating about that we’re working on to see what we can do in that space in the future.
So, disturbance mapping and reporting can be as manual as someone wandering around a site, taking some GPS co-ordinates and then working with spatial data and Excel reports to deliver on their regulatory requirements. But in the future, this automation will mean that a site enviro will be able to focus their time on reviewing the outputs and results, rather than doing the work themselves – freeing them up to deal with other site issues and to make sure other commitments are met.
In principle, that should mean a better outcome for the environment, which is why we do what we do.
P.S. If you are interested in how we can help you with this sort of automation, then feel free to get in touch directly via email, on (08) 92277309, or by starting a conversation on Facebook, Twitter or LinkedIn.