With growing concerns surrounding global warming, both pollution and alternative sources of energy have become the focus of intense research. These research efforts have addressed the need to power pollution sensors by alternative means. Harvesting energy from trees is both minimally invasive to the environment and eliminates harmful waste produced by non-green energy solutions like batteries. The intent of this research is to verify the possibility of developing an unconventional thermoelectric generator (TEG) in order to increase the energy obtained from trees. We believe that a comprehensive understanding of the temperature characteristics of trees improves the current thermoelectric harvester design. A more robust TEG produces the voltages necessary to power these IoT devices. The primary concerns in selecting materials for a specialized TEG are the thermal impedance, the dimensions and arrangement of semiconducting material, and the overall geometric composition of the TEG itself. The selection of these materials and the overall physical characteristics of the TEG depend on the analysis of the internal temperature of trees. We have developed and will soon deploy a system that captures this valuable data over time. Furthermore, we have established a theoretical TEG design tailored specifically to harvest energy from trees more effectively than previous implementations. The new harvester will be utilized to power LoRa wireless sensor networks capable of monitoring pollution and other environmental hazards. If successful, future research should be devoted to optimizing the TEG to minimize thermal resistances and parasitic heat losses, as well as efforts to maximize temperature differentials with the use of smart heat exchangers on the TEG’s ambient side. Electrical matching of the TEG and sensor node could be desirable too. Finally, integration of the TEG into other sensor applications such as wildfire monitoring is more than reasonable.