MIRANDA: MId-feature RANk-adversarial Domain Adaptation toward climate change-robust ecological forecasting with deep learning

Abstract

Plant phenology modelling aims to predict the timing of seasonal phases, such as leaf-out or flowering, from meteorological time series. Reliable predictions are crucial for anticipating ecosystem responses to climate change. While phenology modelling has traditionally relied on mechanistic approaches, deep learning methods have recently been proposed as flexible, data-driven alternatives with often superior performance. However, mechanistic models tend to outperform deep networks when data distribution shifts are induced by climate change. Domain Adaptation (DA) techniques could help address this limitation. Yet, unlike standard DA settings, climate change induces a temporal continuum of domains and involves both a covariate and label shift, with warmer records and earlier start of spring. To tackle this challenge, we introduce Mid-feature Rank-adversarial Domain Adaptation (MIRANDA). Whereas conventional adversarial methods enforce domain invariance on final latent representations, an approach that does not explicitly address label shift, we apply adversarial regularization to intermediate features. Moreover, instead of a binary domain-classification objective, we employ a rank-based objective that enforces year-invariance in the learned meteorological representations. On a country-scale dataset spanning 70 years and comprising 67,800 phenological observations of 5 tree species, we demonstrate that, unlike conventional DA approaches, MIRANDA improves robustness to climatic distribution shifts and narrows the performance gap with mechanistic models.

Figures (11)

• Figure 1: Climate change-robust machine learning. Deep learning models trained on historical climatic conditions face a distribution shift when applied to future climate projections. In this paper we consider models that predict plant phenological dates from meteorological time series and devise a domain adaptation method to enhance their robustness to such shifts. (Temperature data: CMIP6 CNRM-CM6-1 SSP2-4.5 eyring2016cmip6voldoire2019CNRMCM6, phenological data: Swiss Phenology Network SPN)
• Figure 1: Scatter plot of predicted (y-axis) vs true dates (x-axis) before (left, with $R^2$ = 24%) and after (right, with $R^2$ = 26%) adaptation for phenology task leaf unholding of horse chestnut in elevation data split.
• Figure 2: Phenology modeling (conceptual illustration). A phenology model takes a set of climatic and environmental covariates and predicts phenological dates. In this paper we consider models operating on daily meteorological time series to predict the date of $50\%$ leaf development of $5$ different tree species.
• Figure 2: Scatter plot of predicted (y-axis) vs true dates (x-axis) before (left, with $R^2$ = 12%) and after (right, with $R^2$ = 23%) adaptation for phenology task needle emergence of European larch in elevation data split.
• Figure 3: Method overview. Our framework addresses domain shifts in phenology modelling via two key components: rank-based adversarial training on intermediate features and hybrid layer normalization. Purple elements correspond to the main phenology prediction pathway: the learnable tokens $\mathbf{L}$ are concatenated with the input time series $E_i$ and processed by two transformer encoder layers ($\mathbf{t_1}$ and $\mathbf{t_2}$). The resulting global embeddings of the learnable tokens, denoted as $G_i$, are then fed into the decoder $\mathbf{d}$ to predict the target date using the regression loss $\mathcal{L}_{\mathrm{MSE}}$. We apply rank-based adversarial learning on the mid-level features $Z_i$ (green), where a discriminator $\mathbf{p}$ is trained with a ranking loss $\mathcal{L} _{\mathrm{rank}}$ through a gradient reversal layer to encourage domain-invariant mid-level features. Meanwhile, we replace the standard layer normalization in $t_2$ with our hybrid layer normalization (blue), which preserves domain-dependent variations in high-level representations.