Extreme heat in the United States is a leading cause of weather-related deaths, disproportionately affecting low-income communities of color who tend to live in substandard housing with limited indoor cooling and fewer trees. Trees in cities have been documented to improve public health in many ways and provide climate regulating ecosystem services via shading, absorbing, and transpiring heat, measurably reducing heat-related illnesses and deaths. Advancing “urban forest equity” by planting trees in marginalized neighborhoods is acknowledged as a climate health equity strategy. But information is lacking about the efficacy of tree planting programs in advancing urban forest equity and public wellbeing. There is a need for frameworks to address the mismatch between policy goals, governance, resources, and community desires on how to green marginalized neighborhoods for public health improvement—especially in water-scarce environments. Prior studies have used environmental management-based approaches to evaluate planting programs, but few have focused on equity and health outcomes. We adapted a theory-based, multi-dimensional socio-ecological systems (SES) framework regularly used in the public health field to evaluate the Tree Ambassador, orPromotor Forestal, program in Los Angeles, US. The program is modeled after the community health worker model—where frontline health workers are trusted community members. It aims to address urban forest equity and wellbeing by training, supporting, and compensating residents to organize their communities. We use focus groups, surveys, and ethnographic methods to develop our SES model of community-based tree stewardship. The model elucidates how interacting dimensions—from individual to society level—drive urban forest equity and related public health outcomes. We then present an alternative framework, adding temporal and spatial factors to these dimensions. Evaluation results and our SES model highlight drivers aiding or hindering program trainees in organizing communities, including access to properties, perceptions about irrigation responsibilities, and lack of trust in local government. We also find that as trainee experience increases, measures including self- and collective efficacy and trust in their neighbors increase. Findings can inform urban forestry policy, planning, and management actions at the government and non-profit levels that aim to increase tree cover and reduce heat exposure in marginalized communities.