This study examines existing buildings in Haikou in China under tropical island climate conditions. It presents three retrofit design models for greenhouses roofs (GHR), green roofs (GR) and photovoltaic roofs (PVR). The carbon cost of each retrofit roof model is calculated in the production and transportation phases of building materials, construction, and demolition. The changes in electricity consumption, water consumption, and plant carbon reduction are coupled to calculate the carbon reduction generated by each phase of the use of the retrofitted roofs. The carbon reduction per unit area for GHR, GR and PVR over the life cycle (20 years) is then comprehensively calculated. The life cycle carbon reduction per unit area is 262.57 kg·m⁻² for GHR, 127.41 kg·m⁻² for GR and 2567.12 kg·m⁻² for PVR. Among the three retrofit methods, PVR has the greatest potential for reducing carbon emissions. The study can as a guide for implementing carbon reduction measures in tropical island areas. Domestic research on rooftop greenhouses also focuses on technology, yield, and energy consumption, mostly for northern regions with cold winters, and less research on rooftop greenhouses applied to regions with hot summers and warm winters. But domestic and foreign studies on the potential of rooftop greenhouses to reduce emissions have not yet been combined with plant cultivation of hydroelectric carbon emissions and plant carbon sequestration.

Freshwater resources are scarce in tropical island areas. Treating rainwater to produce drinking water through biological slow filtration (BSF) technology can significantly alleviate the problem of freshwater shortages. The characteristics of the filter material are the key factors determining the decontamination performance of BSF technology. However, most existing studies focus on a single filter material. This study was conducted using volcanic rock and coconut shell activated carbon to compare their pollutant removal characteristics in slightly polluted rainwater during the early stage of BSF operation (from the start of operation to day 6, with the first sampling time being 48 h after operation) and during the stable stage (26 days later) and further explore the influence of their mixing ratio. The results show that in the early stages of operation, the pollutant removal performance of volcanic rock and coconut shell activated carbon is better than that of quartz sand. Among them, coconut shell activated carbon showed average removal rates for NH₃-N, TOC, and Cr(VI) that were 6.72, 8.46, and 19.01 percentage points higher than those of volcanic rock, respectively, but its average turbidity removal rate decreased by 5.00%. The removal effect of the mixed filter material was enhanced through the synergistic adsorption mechanism, but most of the improvements were within the standard deviation range and did not exceed the removal range of the single filter material. When the mixing ratio was 1:3, the average total organic carbon removal rate of the filter material was 71.51 ± 0.64%, approximately 0.96 percentage points higher than that of coconut shell activated carbon (70.55 ± 0.42%). While coconut shell activated carbon showed the best removal effect among all single filter materials, this improvement was still within the standard deviation range.

The high summer temperatures of PV (photovoltaic) glass curtain walls lead to reduced power generation performance of PV modules and increased indoor temperatures. To address this issue, this study constructed a test platform for planted photovoltaic glass curtain walls to investigate the effect of plants on their power generation performance. The study’s results indicate the following: (1) reducing the average surface temperature of the surface temperature measurement instrument for the photovoltaic glass curtain wall by 13.6 °C can increase its average power generation capacity by 76 w, demonstrating its power generation performance; (2) plant cultivation influences the micro-environmental temperature on the surface temperature of the photovoltaic glass curtain wall, resulting in a decrease in average micro-environmental temperature by 3.2 °C and average surface temperature by 10.1 °C; (3) compared to traditional PV glass curtain walls, the planted PV glass curtain wall increases cumulative PV power generation output by 21.5 kWh over 15 days and average daily power generation output by 1.4 kWh. Furthermore, during sunny weather with high temperatures, the PV power generation output of the planted PV glass curtain wall is significantly enhanced.

As urbanization accelerates and the availability of arable land declines sharply, building-integrated agriculture (BIA) has emerged as a crucial strategy for enhancing urban food security and it also promotes the establishment of sustainable urban food production systems. This study focuses on indoor living walls (ILWs) and employs the analytic hierarchy process (AHP) and the entropy weighting method to construct a comprehensive suitability evaluation model. The model evaluates different spatial layouts across five key dimensions: indoor microenvironment, physiology of vegetable, morphology of vegetable, yield of vegetable and quality of vegetable. The results reveal that among the experimental groups, R2 was classified as suitable, with an average group score of 2.29. The remaining groups were classified as moderately suitable, with descending average scores of 1.64 for R3, 1.43 for R4, and 1.16 for R1. Based on the climatic characteristics of Hainan Province, the optimal configuration is recommended to include a north-facing room, a west-wall planting layout, and a “partial human–vegetable separation” spatial strategy, with an installation height exceeding 1.3 m.

Commercial plots in hot–humid tropical island cities require effective pedestrian-level ventilation; however, the extent to which different layout forms enhance wind-driven cooling potential under fixed development intensity remains insufficiently quantified. Taking Haikou as a representative tropical island case, this study examines how building count and spatial enclosure form affect the pedestrian-level wind environment of a commercial plot. A total of 63 layouts with one, two, and three building units were constructed under identical development constraints. ANSYS Fluent 2023 R1 (ANSYS, Inc., Canonsburg, PA, USA). was used to simulate pedestrian-level wind fields under representative summer southerly and east–northeasterly (ENE) wind conditions. Six ventilation-related indicators, including area-weighted mean wind speed, maximum wind speed, and the non-low-wind-speed area ratio in both seasons, were integrated into a Wind-Driven Cooling Potential Index (WDCPI). The weighting scheme combined climate-informed seasonal weights with entropy-based objective indicator weights. The results show that summer ventilation is more sensitive to layout form than winter ventilation. Although the average WDCPI decreases as building subdivision increases, layout B2 shows the highest WDCPI among the tested scenarios because its open inter-building space forms a continuous ventilation path aligned with the prevailing summer wind. The sensitivity analysis supports the relative stability of the ranking results. These findings highlight airflow connectivity and windward openness as important layout-screening principles for cooling-oriented commercial plot design in tropical island cities.

In tropical island cities, the combined pressures of rapid high-density urbanization and year-round hot-humid climates make the pedestrian-level wind environment a critical determinant of outdoor thermal comfort and cooling performance. Focusing on Haikou, a tropical island city, this study optimizes building layouts on commercial plots under constant development intensity. A Pedestrian-level Cooling Performance Index (PLCPI) was constructed, prioritizing summer cooling and winter wind control through an AHP-EWM combined weighting method. The index integrates maximum pedestrian-level wind speeds and amplification factors to evaluate 65 layout configurations, including detached, row, perimeter, and courtyard types. The results reveal a nonlinear relationship between building count and cooling performance. Single-building layouts achieve the highest mean PLCPI (2.367), three-building layouts the lowest (1.825), prone to ventilation stagnation, and four-building layouts show a performance rebound (2.271) with stable efficiency. Crucially, spatial enclosure form is the decisive determinant under a constant building count: the optimal two-building layout B-8 (PLCPI=2.456) surpasses the best single-building layout A-2 (2.419), demonstrating that well-designed dispersed layouts can outperform centralized ones. This study proposes a dual-season adaptive evaluation framework for tropical commercial plots and reveals the nonlinear mechanism between building quantity and cooling performance, providing a scientific basis for fine-grained urban design in tropical island climates.