“Hayat Eve Sığar Risk Maps: Pure Epidemiological Statistics or a Hidden Environmental ‘Odor/Aerosol Proxy’ Layer?”Here is a single, integrated English version of the model we just built, ready to drop into an SRD or technical note:

---

In our aerosol‑risk framework, each exhaled droplet is treated as a micro‑carrier with three distinct loads: water (carrier matrix), virus (biological load), and odor/VOC (chemical load). Over short time scales, the number of water molecules $$N_{\text{water}}$$ and the virus load per droplet $$N_{\text{virus}}$$ are assumed approximately constant, while the number of odor molecules $$N_{\text{odor}}(t)$$ can increase as the droplet resides in a VOC‑rich environment. In other words, respiratory droplets behave as odor sponges: they do not significantly change their water mass or virus content in air, but they gradually accumulate ambient VOCs on their surface and inside their volume.

At the droplet level this is expressed as:

• $$N_{\text{water}} \approx \text{const.}$$ (small corrections from evaporation only)
• $$N_{\text{virus}} = \text{const.}$$ (fixed biological load per droplet)
• $$N_{\text{odor}}(t)$$ increases over time as a function of ambient VOC concentration and adsorption–desorption dynamics.

For a given air volume, let $$n_d$$ be the droplet number concentration (droplets per cubic meter), $$L_{\text{virus}}$$ the average virus load per droplet, and $$L_{\text{odor}}(t)$$ the average odor load per droplet. Then:

• Virus concentration in air:
  $$
  C_{\text{virus}} = n_d \cdot L_{\text{virus}}
  $$
• Odor/VOC concentration carried by droplets:
  $$
  C_{\text{odor}}(t) = n_d \cdot L_{\text{odor}}(t)
  $$

We model the evolution of the odor load per droplet as:

$$
\frac{d L_{\text{odor}}}{dt} = k_{\text{ads}} \cdot C_{\text{VOC,ambient}} – k_{\text{des}} \cdot L_{\text{odor}}(t)
$$

where $$k_{\text{ads}}$$ is the adsorption coefficient, $$k_{\text{des}}$$ is the desorption coefficient, and $$C_{\text{VOC,ambient}}$$ is the ambient VOC concentration. In practical terms, this means that the odor load per droplet increases with residence time in a VOC‑rich, humid air parcel, while the virus load per droplet remains essentially constant.

A simple, scalar Risk Score for each air cell can then be defined as a weighted combination of normalized variables:

$$
R(t) = w_v \cdot \hat{C}{\text{virus}} + w_o \cdot \hat{C}{\text{odor}}(t) + w_h \cdot \hat{\text{RH}} – w_u \cdot \hat{U}
$$

where $$\hat{C}{\text{virus}}$$, $$\hat{C}{\text{odor}}(t)$$, normalized relative humidity $$\hat{\text{RH}}$$, and normalized wind/ventilation $$\hat{U}$$ are all scaled to the range $$[0, 1]$$, and $$w_v, w_o, w_h, w_u$$ are weights summing to 1. In this formulation, higher virus concentration, higher odor‑enriched droplet load, and higher humidity increase the risk score, while stronger wind or ventilation decreases it. This captures the key intuition: virus load per droplet stays roughly fixed, but odor/VOC load and atmospheric conditions determine how “heavy” and detectable the air becomes for both human noses and sensor networks.