Novel satellite based analytical model developed to predict microalgae yields in open pond raceway systems and applied to Canadian sites

Interest in microalgae cultivation continues to increase based on its potential commercial value. Algae converts CO2, nitrates, phosphates and other nutrients into a biomass that can be processed into biofuels, pharmaceuticals, nutraceuticals, food, fertilizers, and other active compounds. Solar irradiance and media temperatures are key parameters in determining microalgae cultivation yield and hence these parameters are fundamental in existing models that have been constructed to predict yields in different locations for open pond raceway (OPR) cultivation systems. The challenge in estimating OPR yields in higher and lower latitudes (colder climates) is that there are no known attempts to cultivate algae at any scale in these regions, nor are there data sets that include shallow pond site-specific daily water temperature measurements, from which to construct algae cultivation models. To address these challenges, our research introduces a new data-intensive analytical SATOPR (SATellite Open Pond Raceway) model, relying on ubiquitous historical satellite data. Local solar irradiance and ambient temperature values which are used to predict microalgae production yields at any geographic location including the colder latitudes of central Alberta and the Northwest Territories in Canada. The model predicts that annual open pond algae cultivation to produce 1000 T biomass would require 17-20 ha at Mesa, AZ, 45-56 ha at Medicine Hat, AB, 57-68 ha at Fort Saskatchewan, AB and 71-80 ha at Great Slave Lake, NWT. The Mesa, AZ results are more conservative than forecast by a NREL model predicting 12 ha to produce 1000 T biomass. Modeled land area information provides the basis for life cycle assessments (LCAs), techno-economic analyses (TEAs), and photobioreactor (PBR) versus OPR performance studies, yields simulations based on various parameters, and can assist with algae production platform optimization.