Thomas Lees

Mr. Thomas Lees

orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 July 2021 in Hydrology and Earth System Sciences
Verified
Journal Article
Open access
PDF file
LSLouise Slater
Louise Slater
BABailey Anderson
Bailey Anderson
MBMarcus Buechel
Marcus Buechel
SDSimon Dadson
Simon Dadson
SHSha Han
Sha Han
RWRobert L. Wilby
Robert L. Wilby
Hydroclimatic extremes such as intense rainfall, floods, droughts, heatwaves, and wind or storms have devastating effects each year. One of the key challenges for society is understanding how these extremes are evolving and likely to unfold beyond their historical distributions under the influence of multiple drivers such as changes in climate, land cover, and other human factors. Methods for analysing hydroclimatic extremes have advanced considerably in recent decades. Here we provide a review of the drivers, metrics, and methods for the detection, attribution, management, and projection of nonstationary hydroclimatic extremes. We discuss issues and uncertainty associated with these approaches (e.g. arising from insufficient record length, spurious nonstationarities, or incomplete representation of nonstationary sources in modelling frameworks), examine empirical and simulation-based frameworks for analysis of nonstationary extremes, and identify gaps for future research.

21 October 2021 in Hydrology and Earth System Sciences
Verified
Journal Article
Open access
PDF file
TLThomas Lees
Thomas Lees
MBMarcus Buechel
Marcus Buechel
BABailey Anderson
Bailey Anderson
LSLouise Slater
Louise Slater
SRSteven Reece
Steven Reece
SDSimon J. Dadson
Simon J. Dadson
Long short-term memory (LSTM) models are recurrent neural networks from the field of deep learning (DL) which have shown promise for time series modelling, especially in conditions when data are abundant. Previous studies have demonstrated the applicability of LSTM-based models for rainfall–runoff modelling; however, LSTMs have not been tested on catchments in Great Britain (GB). Moreover, opportunities exist to use spatial and seasonal patterns in model performances to improve our understanding of hydrological processes and to examine the advantages and disadvantages of LSTM-based models for hydrological simulation. By training two LSTM architectures across a large sample of 669 catchments in GB, we demonstrate that the LSTM and the Entity Aware LSTM (EA LSTM) models simulate discharge with median Nash–Sutcliffe efficiency (NSE) scores of 0.88 and 0.86 respectively. We find that the LSTM-based models outperform a suite of benchmark conceptual models, suggesting an opportunity to use additional data to refine conceptual models. In summary, the LSTM-based models show the largest performance improvements in the north-east of Scotland and in south-east of England. The south-east of England remained difficult to model, however, in part due to the inability of the LSTMs configured in this study to learn groundwater processes, human abstractions and complex percolation properties from the hydro-meteorological variables typically employed for hydrological modelling.

20 June 2022 in Hydrology and Earth System Sciences
Verified
Journal Article
Open access
PDF file
TLThomas Lees
Thomas Lees
SRSteven Reece
Steven Reece
FKFrederik Kratzert
Frederik Kratzert
DKDaniel Klotz
Daniel Klotz
MGMartin Gauch
Martin Gauch
SDSimon J. Dadson
Simon J. Dadson
Neural networks have been shown to be extremely effective rainfall-runoff models, where the river discharge is predicted from meteorological inputs. However, the question remains: what have these models learned? Is it possible to extract information about the learned relationships that map inputs to outputs, and do these mappings represent known hydrological concepts? Small-scale experiments have demonstrated that the internal states of long short-term memory networks (LSTMs), a particular neural network architecture predisposed to hydrological modelling, can be interpreted. By extracting the tensors which represent the learned translation from inputs (precipitation, temperature, and potential evapotranspiration) to outputs (discharge), this research seeks to understand what information the LSTM captures about the hydrological system. We assess the hypothesis that the LSTM replicates real-world processes and that we can extract information about these processes from the internal states of the LSTM. We examine the cell-state vector, which represents the memory of the LSTM, and explore the ways in which the LSTM learns to reproduce stores of water, such as soil moisture and snow cover. We use a simple regression approach to map the LSTM state vector to our target stores (soil moisture and snow). Good correlations (R2>0.8) between the probe outputs and the target variables of interest provide evidence that the LSTM contains information that reflects known hydrological processes comparable with the concept of variable-capacity soil moisture stores. The implications of this study are threefold: (1) LSTMs reproduce known hydrological processes. (2) While conceptual models have theoretical assumptions embedded in the model a priori, the LSTM derives these from the data. These learned representations are interpretable by scientists. (3) LSTMs can be used to gain an estimate of intermediate stores of water such as soil moisture. While machine learning interpretability is still a nascent field and our approach reflects a simple technique for exploring what the model has learned, the results are robust to different initial conditions and to a variety of benchmarking experiments. We therefore argue that deep learning approaches can be used to advance our scientific goals as well as our predictive goals.

31 March 2020 in Earth System Science Data
Verified
Journal Article
Open access
PDF file
Jian PengJian Peng
Jian Peng
SDSimon Dadson
Simon Dadson
FHFeyera A. Hirpa
Feyera A. Hirpa
EDEllen Dyer
Ellen Dyer
TLThomas Lees
Thomas Lees
CFChris Funk
Chris Funk
Droughts in Africa cause severe problems, such as crop failure, food shortages, famine, epidemics and even mass migration. To minimize the effects of drought on water and food security on Africa, a high-resolution drought dataset is essential to establish robust drought hazard probabilities and to assess drought vulnerability considering a multi- and cross-sectional perspective that includes crops, hydrological systems, rangeland and environmental systems. Such assessments are essential for policymakers, their advisors and other stakeholders to respond to the pressing humanitarian issues caused by these environmental hazards. In this study, a high spatial resolution Standardized Precipitation-Evapotranspiration Index (SPEI) drought dataset is presented to support these assessments. We compute historical SPEI data based on Climate Hazards group InfraRed Precipitation with Station data (CHIRPS) precipitation estimates and Global Land Evaporation Amsterdam Model (GLEAM) potential evaporation estimates. The high-resolution SPEI dataset (SPEI-HR) presented here spans from 1981 to 2016 (36 years) with 5 km spatial resolution over the whole of Africa. To facilitate the diagnosis of droughts of different durations, accumulation periods from 1 to 48 months are provided. The quality of the resulting dataset was compared with coarse-resolution SPEI based on Climatic Research Unit (CRU) Time Series (TS) datasets, Normalized Difference Vegetation Index (NDVI) calculated from the Global Inventory Monitoring and Modeling System (GIMMS) project and root zone soil moisture modelled by GLEAM. Agreement found between coarse-resolution SPEI from CRU TS (SPEI-CRU) and the developed SPEI-HR provides confidence in the estimation of temporal and spatial variability of droughts in Africa with SPEI-HR. In addition, agreement of SPEI-HR versus NDVI and root zone soil moisture – with an average correlation coefficient (R) of 0.54 and 0.77, respectively – further implies that SPEI-HR can provide valuable information for the study of drought-related processes and societal impacts at sub-basin and district scales in Africa. The dataset is archived in Centre for Environmental Data Analysis (CEDA) via the following link: https://doi.org/10.5285/bbdfd09a04304158b366777eba0d2aeb (Peng et al., 2019a).

2 February 2022 in Remote Sensing
Verified
Journal Article
Open access
PDF file
TLThomas Lees
Thomas Lees
GTGabriel Tseng
Gabriel Tseng
Clement ATZBERGERClement ATZBERGER
Clement ATZBERGER
SRSteven Reece
Steven Reece
SDSimon Dadson
Simon Dadson
East Africa has experienced a number of devastating droughts in recent decades, including the 2010/2011 drought. The National Drought Management Authority in Kenya relies on real-time information from MODIS satellites to monitor and respond to emerging drought conditions in the arid and semi-arid lands of Kenya. Providing accurate and timely information on vegetation conditions and health—and its probable near-term future evolution—is essential for minimising the risk of drought conditions evolving into disasters as the country’s herders directly rely on the conditions of grasslands. Methods from the field of machine learning are increasingly being used in hydrology, meteorology, and climatology. One particular method that has shown promise for rainfall-runoff modelling is the Long Short Term Memory (LSTM) network. In this study, we seek to test two LSTM architectures for vegetation health forecasting. We find that these models provide sufficiently accurate forecasts to be useful for drought monitoring and forecasting purposes, showing competitive performances with lower resolution ensemble methods and improved performances over a shallow neural network and a persistence baseline.

14 November 2020 in EGUsphere
Verified
Preprint
Open access
PDF file
LSLouise Slater
Louise Slater
BABailey Anderson
Bailey Anderson
MBMarcus Buechel
Marcus Buechel
SDSimon Dadson
Simon Dadson
SHSha Han
Sha Han
RWRobert L. Wilby
Robert L. Wilby
Hydroclimatic extremes such as intense rainfall, floods, droughts, heatwaves, and wind/storms have devastating effects each year. One of the key challenges for society is understanding how these extremes are evolving and likely to unfold beyond their historical distributions under the influence of multiple drivers such as changes in climate, land cover, and other human factors. Methods for analysing hydroclimatic extremes have advanced considerably in recent decades. Here we provide a review of the drivers, metrics and methods for the detection, attribution, prediction and projection of nonstationary hydroclimatic extremes. We discuss issues and uncertainty associated with these approaches (e.g arising from insufficient record length, spurious nonstationarities, or incomplete representation of nonstationary sources in modelling frameworks), examine empirical and simulation-based frameworks for analysis of nonstationary extremes, and identify gaps for future research.