JavaScript is disabled in your web browser or browser is too old to support JavaScript. Today almost all web pages contain JavaScript, a scripting programming language that runs on visitor's web browser. It makes web pages functional for specific purposes and if disabled for some reason, the content or the functionality of the web page can be limited or unavailable.

Thursday, July 3, 2025

Embracing data-driven models for

climate resilience in the Caribbean

by

Guardian Media Limited
11 days ago
20250622

As a uni­ver­si­ty fo­cused on us­ing re­search for the up­lift­ment of its com­mu­ni­ty, The Uni­ver­si­ty of the West In­dies, St Au­gus­tine (UWI STA), the pre­mier ter­tiary in­sti­tu­tion in the re­gion, pro­duc­ing world-class schol­ars, be­lieves sci­ence should be ac­ces­si­ble to the pub­lic. We are pleased to present our me­dia se­ries, “UWI Sci­en­tists Speak”. In this se­ries, our sci­en­tists—three of whom re­ceived the na­tion’s high­est ho­n­our, the Or­der of the Re­pub­lic of Trinidad and To­ba­go, in 2023 and one in 2024—will show­case some of their work.

This week, we hear from Dr Lete­tia Ad­di­son about her trans­for­ma­tive work us­ing da­ta-dri­ven in­sights to build mul­ti­dis­ci­pli­nary, sus­tain­able mod­els in cli­mate change.

Dr Ad­di­son is an ex­pe­ri­enced ed­u­ca­tor, sta­tis­ti­cian, and re­searcher at the UWI. She spe­cialis­es in math­e­mat­ics and sta­tis­tics ed­u­ca­tion and de­vel­ops pre­dic­tive, da­ta-dri­ven mod­els for cli­mate change sus­tain­abil­i­ty. 

–Prof Rose-Marie Belle An­toinePrin­ci­pal, UWI STA

Da­ta is un­de­ni­ably play­ing a piv­otal role in Caribbean de­vel­op­ment. My pas­sion for build­ing math­e­mat­i­cal and sta­tis­ti­cal mod­els has sparked my in­ter­est in con­tribut­ing da­ta-dri­ven in­sights to con­struct mul­ti­dis­ci­pli­nary, sus­tain­able ap­proach­es in var­i­ous projects in cli­mate re­silience.

As a re­searcher and ed­u­ca­tor, I have wit­nessed the re­mark­able in­flu­ence da­ta can wield in tack­ling in­tri­cate is­sues and fos­ter­ing tan­gi­ble progress.

At the core of my ap­proach is the be­lief that tra­di­tion­al meth­ods alone are in­suf­fi­cient to tack­le the in­tri­ca­cies of con­tem­po­rary prob­lems. I in­te­grate da­ta-dri­ven mod­el­ling, in­ter­dis­ci­pli­nary col­lab­o­ra­tion, and a deep com­mit­ment to sus­tain­abil­i­ty to de­vel­op pre­dic­tive mod­els that ad­dress the root caus­es of is­sues re­lat­ed to cli­mate change, for in­stance, flood risk pre­dic­tion and coastal re­silience. 

By bridg­ing the gap be­tween acad­e­mia and in­dus­try, we can lever­age the ex­per­tise, re­sources, and re­al-world in­sights of both sec­tors to de­vel­op in­no­v­a­tive so­lu­tions that ad­dress the unique chal­lenges fac­ing the Caribbean.

This col­lab­o­ra­tive ap­proach not on­ly en­rich­es our re­search en­deav­ours but en­sures our sus­tain­able mod­els are prac­ti­cal, scal­able, and di­rect­ly ap­plic­a­ble to the needs of com­mu­ni­ties and in­dus­tries across the re­gion. 

My work in cli­mate change be­gins with the in­ter­con­nect­ed­ness of en­vi­ron­men­tal, so­cial, and eco­nom­ic fac­tors. Through rig­or­ous da­ta analy­sis, I un­cov­er hid­den pat­terns and cor­re­la­tions that pro­vide in­valu­able in­sights in­to the unique dy­nam­ics of our Caribbean re­gion.

By in­te­grat­ing these in­sights in­to sus­tain­able da­ta-dri­ven mod­els, I aim to em­pow­er stake­hold­ers to make in­formed de­ci­sions that mit­i­gate en­vi­ron­men­tal im­pact and fos­ter re­silience. 

In 2022, I led a cross-cam­pus UWI team to first place in the Growth and Re­silience Di­a­logue Cli­mate Re­silience Da­ta Chal­lenge. Our pro­to­type pre­dict­ed flood risk us­ing tem­per­a­ture and rain­fall da­ta, of­fer­ing ac­tion­able in­sights for the OECS re­gion.

In­spired by this, I have con­tin­ued to pur­sue AI-pow­ered, da­ta-dri­ven mod­els to solve com­plex cli­mate prob­lems. One of my cur­rent projects, AI4SIDS, is a cut­ting-edge cli­mate re­silience plat­form pro­pos­al for SIDS.

I led this ef­fort along with a mul­ti­dis­ci­pli­nary team of tal­ent­ed fel­low UWI grad­u­ate stu­dents and col­leagues, in­clud­ing Mr Ak­il Hosang, Mr Trevon Tewari, Ms Tamia-Ash­ley Tuitt, Dr Kris Manohar, and Prof Patrick Ho­sein. We re­ceived first place among 114 glob­al en­tries at the 2024 COP29 Glob­al AI In­no­va­tion Chal­lenge in Baku, Azer­bai­jan.

The proof of con­cept for the AI4SIDS plat­form us­es com­mu­ni­ty-lev­el da­ta, satel­lite im­agery, riv­er gauge and weath­er in­puts—and even so­cial me­dia—to pre­dict flood risk in re­al-time.

We are al­so ex­per­i­ment­ing with em­bed­ding a gen­er­a­tive AI com­po­nent in­to the sys­tem to sim­u­late stake­hold­er sce­nar­ios and sup­port com­mu­ni­ty ed­u­ca­tion. With na­tion­al part­ners such as the Of­fice of Dis­as­ter Pre­pared­ness and Man­age­ment (ODPM), we are look­ing for­ward to test­ing this pi­lot plat­form to eval­u­ate its scal­a­bil­i­ty and pol­i­cy im­pact as a dig­i­tal pub­lic good.

I am al­so part of AI4COASTS, an ex­cit­ing 18-month re­search project col­lab­o­ra­tion be­tween UWI sci­en­tists and in­ter­na­tion­al part­ners from the Uni­ver­si­ty of Dublin, fund­ed by the Our Shared Ocean Pro­gramme sup­port­ed by Irish Aid. Our goal is to im­prove coastal re­silience in da­ta-scarce re­gions like the Caribbean.

The project us­es ma­chine learn­ing to pre­dict flood risk, iden­ti­fy al­ter­na­tive da­ta sources, and en­hance out­reach in vul­ner­a­ble com­mu­ni­ties. This col­lab­o­ra­tive ef­fort is dri­ven by the ex­per­tise of Dr Deb­o­rah Vil­laroel-Lamb (lec­tur­er in Coastal En­gi­neer­ing and project orig­i­na­tor), Pro­fes­sor Patrick Ho­sein (Pro­fes­sor of Elec­tri­cal and Com­put­er En­gi­neer­ing), and Dr Md Salaud­din (As­sis­tant Pro­fes­sor of Civ­il En­gi­neer­ing, Uni­ver­si­ty of Dublin), com­bin­ing strengths in coastal en­gi­neer­ing, civ­il in­fra­struc­ture, and AI in­no­va­tion.

Across these projects, my goal is to guide the de­ci­sion-mak­ing process, par­tic­u­lar­ly in da­ta-sparse lo­ca­tions such as the Caribbean. Stake­hold­er en­gage­ment is cen­tral to these projects, in­clud­ing part­ner­ships with gov­ern­ment agen­cies and com­mu­ni­ty or­gan­i­sa­tions, with a fo­cus on ac­cess­ing and ex­plor­ing datasets. 

Of course, we face re­al bar­ri­ers—lim­it­ed ac­cess to da­ta, un­equal ac­cess to tech­nol­o­gy, and deep-root­ed sys­temic chal­lenges. How­ev­er, through in­clu­sive, in­ter­dis­ci­pli­nary col­lab­o­ra­tion, we can over­come these hur­dles. My vi­sion is clear: to use da­ta to em­pow­er de­ci­sion-mak­ers, sup­port lo­cal com­mu­ni­ties, and build re­silient fu­tures for gen­er­a­tions to come.

As the cli­mate cri­sis ac­cel­er­ates, we must move from re­ac­tive to proac­tive plan­ning. The tools are avail­able. The time is now. I hope this work in­spires oth­ers—es­pe­cial­ly young Caribbean sci­en­tists—to join in build­ing sus­tain­able, da­ta-dri­ven so­lu­tions that sup­port our re­gion and be­yond.

For more in­for­ma­tion on my on­go­ing re­search and col­lab­o­ra­tive in­no­va­tion projects in cli­mate re­silience, such as AI4SIDS and AI4COASTS, vis­it https://cli­mate.lab.tt/.


Related articles

Sponsored

Weather

PORT OF SPAIN WEATHER

Sponsored