Optimal sampling and assay for soil organic carbon estimation

01/19/2021
by   Jacob V. Spertus, et al.
0

The world needs around 150 Pg of negative carbon emissions to mitigate climate change. Global soils may provide a stable, sizeable reservoir to help achieve this goal by sequestering atmospheric carbon dioxide as soil organic carbon (SOC). In turn, SOC can support healthy soils and provide a multitude of ecosystem benefits. To support SOC sequestration, researchers and policy makers must be able to precisely measure the amount of SOC in a given plot of land. SOC measurement is typically accomplished by taking soil cores selected at random from the plot under study, mixing (compositing) some of them together, and analyzing (assaying) the composited samples in a laboratory. Compositing reduces assay costs, which can be substantial. Taking samples is also costly. Given uncertainties and costs in both sampling and assay along with a desired estimation precision, there is an optimal composite size that will minimize the budget required to achieve that precision. Conversely, given a fixed budget, there is a composite size that minimizes uncertainty. In this paper, we describe and formalize sampling and assay for SOC and derive the optima for three commonly used assay methods: dry combustion in an elemental analyzer, loss-on-ignition, and mid-infrared spectroscopy. We demonstrate the utility of this approach using data from a soil survey conducted in California. We give recommendations for practice and provide software to implement our framework.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
05/27/2021

Rethinking InfoNCE: How Many Negative Samples Do You Need?

InfoNCE loss is a widely used loss function for contrastive model traini...
research
01/18/2021

Buying Data Over Time: Approximately Optimal Strategies for Dynamic Data-Driven Decisions

We consider a model where an agent has a repeated decision to make and w...
research
11/11/2018

Subsampling to Enhance Efficiency in Input Uncertainty Quantification

In stochastic simulation, input uncertainty refers to the output variabi...
research
02/03/2023

Robust Budget Pacing with a Single Sample

Major Internet advertising platforms offer budget pacing tools as a stan...
research
05/20/2021

Multiple Support Recovery Using Very Few Measurements Per Sample

In the problem of multiple support recovery, we are given access to line...
research
06/20/2020

Towards Adaptive Benthic Habitat Mapping

Autonomous Underwater Vehicles (AUVs) are increasingly being used to sup...

Please sign up or login with your details

Forgot password? Click here to reset