EVALUATING HURRICANE DAMAGE AND RESILIENCE IN COFFEE AGROECOSYSTEMS IN PUERTO RICO

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: AFRI COMPETITIVE GRANT
• Reporting Frequency: Annual
• Accession No.: 1015928
• Grant No.: 2018-67030-28239
• Cumulative Award Amt.: $100,000.00
• Proposal No.: 2017-09271
• Project Start Date: Jun 1, 2018
• Project End Date: May 31, 2022
• Grant Year: 2018
• Program Code: [A1801]- Exploratory: Exploratory Research

Recipient Organization
UNIVERSITY OF MICHIGAN
(N/A)
ANN ARBOR,MI 48109

Performing Department
(N/A)

Non Technical Summary
Puerto Rico is regularly battered by hurricanes. The recent impact of hurricane Maria, while devastating for farmers, offers an opportunity to study the impact of such strong climatic events on agriculture on the island. Coffee is one of the most important crops in Puerto Rico, and offers an ideal vehicle for the study of disturbance and resilience on agroecosystems and their biodiversity. We are proposing to take advantage of this devastating hurricane in Puerto Rico to study the resilience of coffee farms under various management and its ecological recuperation over the next coupler of years. We seek to evaluate the degree of damage done to the coffee sector in the central mountainous region by the hurricane and follow the initial stages of recovery of the coffee and its biodiversity. For this proposal, we will focus on the taxon for which we have some pre-hurricane data: birds, frogs, lizards, ants, parasitoid wasps and bees, all of which are important ecosystem services providers. We will also determine the impact of the hurricane of coffee yield and therefore farmer's livelihoods and income. In addition, we will quantify biomass loss due to the impact of the hurricane as well as number of landslides and will relate these to the type of management before the hurricane

Animal Health Component

50%

Research Effort Categories

Basic

50%

Applied

50%

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
125	2232	1070	100%

Knowledge Area
125 - Agroforestry;

Subject Of Investigation
2232 - Coffee;

Field Of Science
1070 - Ecology;

Keywords

(n/a)

Goals / Objectives
This project has three major goals related to the impact of hurricane Maria to the coffee sector in the central mountainous region of Puerto Rico:Evaluate the extent of damage done by the hurricane to coffee farms;Determine the resistance of coffee farms and how it relates to the management style;Determine the resilience of coffee farms and how it relates to the management style.

Project Methods
To accomplish the stated objectives, we will select a subset of the 178 farms that we have already identified and from which we have relevant pre-hurricane data, and gather data on 1) vegetation structure, above ground biomass and landslides, 2) below-ground carbon, 3) biodiversity, 4) yields.1) Vegetation structure, above-ground biomass and landslides: Vegetation structure: We have vegetation structure data for 72 farms. Tree density (trees/ 0.04 ha) and shrub density (shrubs/ 0.008 ha) were estimated for each farm at various points depending on the size of the farm. All trees and shrubs were identified to species. We used ArcMap 10.4.1 (ESRI 2011) to estimate the percent vegetation cover within 50 and 200 meter radii of each sampling point using images from Google earth V 7.1.8.3036 (32-bit; September 13, 2016), Puerto Rico DigitalGlobe, www.digitalglobe.com; Google Earth, www.google.com/earth;CNES, cnes.fr/en). We already have polygons for all 72 farms and will compare pre- and post-hurricane vegetation cover in these polygons. In addition to landscape quantification from imagery, we will visit all 72 farms and count the number of trees and shrubs that are snapped, truncated, uprooted or completely down, as well as the percent of the coffee plants that are snapped, uprooted, or dead. Also, we will do trail survey within the farms and quantify the number of plantain/banana plants and tuber pants that are uprooted or dead. We also have pre-hurricane data on percent canopy cover for one 25X25 M plot on 36 of the farms. All these data will be gathered again as soon as possible but no longer than 4 months after the hurricane, and then again, one year and two years after the hurricane. These data will not only allow us to assess the immediate impact of the hurricane on the coffee farms, but also will allow us to measure the level of resistance and resilience of the vegetation and relate that to pre-hurricane management style and vegetation structure. Above-ground biomass: We have above-ground biomass for individual 25X25 M plots in 36 farms. Using the same methodology, but modifying to account for the trees that are snapped or truncated, we will estimate above ground biomass and compare it with the pre-hurricane data. Above ground biomass (AGB) for trees will be estimated by applying the allometric equations. Specific wood density will be compiled from various sources. For AGB stored in coffee we will use another allometric equation developed specifically for coffee. AGB for trees and leaf litter will be converted to above ground carbon (AGC) by using a conversion factor of 0.47 and 0.45 respectively. AGC for coffee plants will be obtained by using the conversion factor of 0.5. Landslides: We will estimate the number of landslides for a minimum of 72 farms. These data will be analyzed against a number of vegetation variables (pre-hurricane tree and shrub density and diversity and percent canopy cover, and post-hurricane percent of uprooted and truncated trees, etc.), elevation, aspect, and slope, to determine what factors explain the highest percent of variability in landslides.2) Below-ground carbon:Although we don't have any pre-hurricane below-ground carbon estimates, we nevertheless think that it would be valuable to estimate below-ground carbon at various intervals after the hurricane to estimate the resilience of soils under different management systems. Below ground soil carbon will be estimated by taking three triplets of soil cores to a depth of 25 cm from the 25X25 M plot (each sample close to each of the nine sampling points per plot - see biodiversity methods, below). The soil cores triplet will be pooled into a single sample, resulting in three samples per plot and analyzed for SOC. Additionally, three soil core samples of 510 cm3 will be taken at random locations in each plot and dried at 105oC for 48 h to determine bulk density. Root biomass will be estimated from total AGB by using an allometric equation developed for tropical forests, and assuming that the carbon content in roots is 50%. Finally, soil organic carbon content (percent) will be determined directly by dry combustion and the total estimated SOC (Mg ha-1) will be projected by multiplying the average SOC% per plot times the sample depth (25 cm), times the areas (1 ha) times the average bulk density per plot. Soil carbon data will be taken in the same 36 farms where AGB is estimated.3) Biodiversity: Birds: We have pre-hurricane avian surveys for 72 farms and will select a subset of 40 of these representing the gradient of intensification for conducting post-hurricane bird surveys. Following the exact procedure for the pre-hurricane survey, one survey station will be established in each of the 40 farms. Two observers will record all bird species, detected visually or by sound, within 50 meters of the center of the station for 10 minutes. One survey will be conducted as soon as possible (or a maximum of 4 months after the hurricane) and two additional surveys will be conducted one and two years after the hurricane. Coqui frogs: We have estimates of coqui frog species richness from a study conducted last summer in 20 coffee farms using acoustic monitoring. We will use the same sites and method to estimate species richness and coqui activity in coffee farms after the hurricane. Recordings will be made as soon as possible (or a maximum of 4 months after the hurricane) and two additional surveys will be conducted in the summer of 2018 and the summer of 2019. Data will be analyzed by means of spectrogram in an ARBIMON-acoustic visualizer. Anole lizards: We have survey data for anole lizards for 36 farms. To sample anole lizards (Anolis spp.), we will walk two transects measuring 25 m x 4 m along a row of coffee bushes in each plot. Within each transect, we will visually search the ground and all vegetation for anoles and will identify them to species. Ants, parasitic wasps, and bees: We have data on these arthropods for a 25X25 M plot in each of 36 farms. For comparative purposes, the methods used for sampling these arthropods will be the same as those used by Iverson. Parasitoid wasps and bees will be sampled placing yellow, white and blue pan traps (three of each) in a 25X25 M plot within each farm. Pan traps will be placed on the ground and filled with soapy water in the early morning hours and left open for 5 consecutive hours. Ants will be sample by placing 1 gram of tuna fish in oil in 25 stations on coffee plants and 25 stations on the ground within the 25X25 M plot in each farm. After 20 minutes the bait will be checked and the ants on the baits will be counted and identified to species on site or collected for further identification in the laboratory.4) Yields:Estimates of pre- and post-hurricane yields for coffee, fruit trees, tubers and plantains/bananas will be obtained through short interviews with the farmers for the 72 farms where the bird surveys will be conducted. We have pre-hurricane yield data for 36 of these farms for 2015 that will be used to validate data from the interviews. With these data we will estimate economic losses due to crop losses from the impact of the hurricane.

Progress 06/01/18 to 04/01/22

Outputs
Target Audience:The academic community and coffee farmers in Puerto Rico. Changes/Problems:The COVID-19 pandemic delayed our reserch activities significantly since we were not able to travel to Puerto Rico during the entire year in 2020. Nonetheless, we were able to achieve all the major goals of the project. What opportunities for training and professional development has the project provided?Throughout the project we provided research opportunity and training to two UM postdoctoral fellows (Kevin Li and Niel McCune, funded by project grants), five UM PhD students (Nicholas Medina, Chatura Vaidya;, Alexa White, Iris Rivera, and Zhacharias Hajian-Forooshani), seven UM Master's students (Janice Newson, Andriana Miljanic, Jenny Flores, Jacobe Longmeyer, Simone Oliphant, Ember Bradbery, and Bella Mayorga) and three University of Puerto Rico-Utuado undergraduate students (Priscilla Cintron Bartolomei, Ayiana Rivera, Nestor Rodriguez). We also trained 3 field assistants from Puerto Rico (Warren Irizarry, Amarilys Irizarry and Isa Marie Acosta). Of all the UM students trained during this project, 4 were Latinx, 3 were black, and 3 were Asian Americans. We also taught a two-week intensive course at the University of Puerto Rico-Utuado (for 15 students) that was partially based on the results of this project. Finally, we also collaborated and provided research training to Professor Javier Lugo, a faculty member at the University of Puerto Rico-Utuado. How have the results been disseminated to communities of interest?The main results regarding the effect of hurricane Maria on coffee farms were published in one high profile article that received some press coverage (Perfecto et al., 2019). https://seas.umich.edu/news/unexpected-outcomes-damages-puerto-rican-coffee-farms-hurricane-maria-varied-widely We also publish other articles (see list of publications below) in scientific journals and presented part of the results at several departmental seminars and conferences. We also presented the results of this study at a workshop for coffee farmers where 50 farmers participated. What do you plan to do during the next reporting period to accomplish the goals?This is a final report and therefore no additional activities are expected.

Impacts
What was accomplished under these goals? Extent of the damage and resilience of coffee farms:We conducted an extensive survey of the impact of hurricane Maria on 85 coffee farms. We reported that the damage to coffee farms from hurricane Maria was strong with a 35% canopy reduction. Additionally, coffee farms in Puerto Rico had a low level of resistance to a hurricane of the intensity and magnitude of hurricane Maria. We then conducted a more detailed study of 10 farms and found that resilience of coffee farms in Puerto Rico is a socioecological phenomenon, with farmers that have the economic capacity to intervene quickly after the hurricane, being able to avoid complete takeover of the coffee by vines and other weeds, while those that were able to weed their coffee right after the hurricane avoided this weed take over. These results were published in Perfecto et al., 2019. Coffee resistance and resilience and tradeoffs with other ecosystem services in coffee agroforestry systems:We leverage island-wide (USDA Climate Hub) and individual farm-level (data collected by us for this project) data sets from Puerto Rico to evaluate the synergies and trade-offs associated with coffee agroforestry. Using a data set of 351 coffee farms spanning the entire island of Puerto Rico, we first evaluate the claim that intensified agricultural systems are more productive than those managed under principles of agroforestry. At the farm-level, we analyze the effect of management decisions, including shade, ground cover, and crop richness, on the biodiversity of ants, lizards, and birds and on non-provisioning ecosystem services such as carbon storage and hurricane resistance (measured as reduction in NDBI) and resilience (measures as how fast NDVI recovered to levels before the hurricane). We identify where tradeoffs and synergies exist and propose that agroforestry management of coffee results in many more synergies than tradeoffs with no apparent loss in yield. Our results support other findings in the literature and offer new insights on the importance of scale in evaluating agroforestry. With respect to resistance and resilience we found that resistance was positively associate with canopy cover and negatively associate with ground cover, while resilience was positively associated with crop diversity. A manuscript with this results (Mayorga et al.) is currently in revision in Frontiers in Sustainable Food Systems. Impact of the hurricane of the fauna of coffee farms: We assessed the impact of hurricane Maria and prospects of rebounding (resiliency) of 12 resident avian species in the context of two stages of shade layer restoration. We estimated occupancy, colonization and extinction probability using survey data collected March-June 2015-2017 (pre-hurricane) and 2018 (post-hurricane) in 58 coffee farms. The local composition of the resident avian community in farms, pre- and post-hurricane, retained a high degree of similarity (Jaccard Similarity Index = 81%). Local turnover involved eight resident species (2 gains, 6 losses), but with no discernible pattern in terms of particular taxonomic groups or foraging guilds. Local losses measured 6-9 months post-hurricane represented 14% of the resident avian community, slightly below the 21% local loss reported one-year post-hurricane Iris. Higher estimates of local extinction probability were associated with major decreases in occupancy. Prospects of rebounding were more likely for species with invariant or non-significant reductions in colonization probability. Rebounding for species like the Puerto Rican Bullfinch (i.e., colonization rates = 0.04 ± 0.2) would be protracted. These results were published in Irizarry et al., 2021. We also evaluated the response of three species of coquí frogs (Eleutherodactylus antillensis, Eleutherodactylus brittoni, and Eleutherodactylus coqui, to Hurricane María in Puerto Rican coffee agroecosystems within the Cordillera Central region. We compared vegetative and coquí population data taken before and after hurricane Maria hit Puerto Rico. We found that vegetative factors were highly impacted by the natural disaster, whereas coquís populations remained stable. Our results suggest that E. antillensis, E. brittoni, and E. coquí may be resilient to natural disasters even while their agroecosystem habitats are dramatically altered. These results are being written for a publication (and a Master thesis) to be submitted in May 2022. In addition, we collected faces of both coqui frogs and anole lizards after the hurricane to assess their potential impact as biological control of insect pests in coffee farms. DNA extraction and gene amplification has been conducted. Illumina MiSeq will then be run to output the final metabarcode result, which upon upload to BLAST will give each species name and type found in the fecal sample, allowing us to determine if these frogs and lizard species are eating some of the major coffee pests. This study will be finalized by the end of May and will be submitted for publication over the summer 2022.

Publications

• Type: Journal Articles Status: Published Year Published: 2019 Citation: Perfecto, I., Z. Hajian-Forooshani*, A. Iverson, A. Irizarry, N. Medina, C. Vaidya, A. White, J. Vandermeer. 2019. Response of coffee farms to Hurricane Maria: resistance and resilience from an extreme climatic event. Scientific Reports (2019) 9:15668
• Type: Theses/Dissertations Status: Published Year Published: 2019 Citation: Harmon, K. 2019. Eleutherodactylus in coffee agroecosystems: Effects of farm management on vegetation selection, species richness, and abundances of Puerto Rican frogs. Master's Thesis, University of Michigan.
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Newson, J., J. Vandermeer and I. Perfecto. 2021. Differential effects of ants as biological control of the coffee berry borer in Puerto Rico. Biological Control 160: 104666. https://doi.org/10.1016/j.biocontrol.2021.104666
• Type: Journal Articles Status: Published Year Published: 2021 Citation: Irizarry, A.D., Collazo, J.A., Vandermeer, J. and Perfecto, I., 2021. Coffee plantations, hurricanes and avian resiliency: insights from occupancy, and local colonization and extinction rates in Puerto Rico. Global Ecology and Conservation, 27, p.e01579. https://doi.org/10.1016/j.gecco.2021.e01579
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Vandermeer, J. and I. Perfecto. 2020. Endogenous spatial pattern formation from two intersecting ecological mechanisms: the dynamic coexistence of two invasive ant species in Puerto Rico. Proceedings of the Royal Society B. 287, no. 1936 (2020): 20202214. https://doi.org/10.1098/rspb.2020.2214
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Perfecto, I. and J. Vandermeer. 2020. The assembly and importance of a novel ecosystem: The ant community of coffee farms in Puerto Rico. Ecology and Evolution 10(23): 12650-12621. (https://doi.org/10.1002/ece3.6785)
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Perfecto, I., Z. Hajian-Forooshani, A. White and J. Vandermeer. 2020. Ecological complexity and contingency: ants and lizards affect biological control of the coffee leaf miner in Puerto Rico. Agriculture, Ecosystems and Environment 305: 107104. https://doi.org/10.1016/j.agee.2020.107104
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Hajian-Forooshani, Z., J. Vandermeer, I. Perfecto. 2020. Insight from excrement: invasive gastropod shift diet to consume the coffee leaf rust and its mycoparasite. Ecology 102(5) e02966 https://doi.org/10.1002/ecy.2966
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Perfecto, I. and J. Vandermeer. 2020. Antagonism between Anolis spp. and Wasmannia auropunctata in coffee farms on Puerto Rico: Potential complications of biological control of the coffee berry borer. Caribbean Journal of Science 50 (1): 43-47. Highlighted in Project Biodiversify www.projectbiodiversify.org/ivette-perfecto
• Type: Journal Articles Status: Published Year Published: 2019 Citation: Iverson, A., D. Gonthier, D. Pak, K. Ennis, R. Burnham, I. Perfecto, M. Ramos Rodriguez, J. Vandermeer. 2019. A multifunctional approach for achieving simultaneous biodiversity conservation and farmer livelihood in coffee agroecosystems. Biological Conservation 238: 108179. https://doi.org/10.1016/j.biocon.2019.07.024
• Type: Journal Articles Status: Published Year Published: 2019 Citation: McCune, N., I. Perfecto, K. Aviles-Vazquez, J. Vazquez-Negron, J. Vandermeer. 2019. Peasant balances and agroecological scaling in Puerto Rican coffee farmers. Agroecology and Sustainable Food Systems 43 (7-8): 810-826 (DOI: 10.1080/21683565.2019.1608348)
• Type: Journal Articles Status: Under Review Year Published: 2022 Citation: McCune, N., J. Vandermeer, K. Aviles-Vazquez, J. Vazquez-Negron, I. Perfecto. Disaster colonialism, populism, and resistance in Puerto Rico. Journal of Rural Studies
• Type: Journal Articles Status: Under Review Year Published: 2022 Citation: 7. Vandermeer, J., J. Flores, J. Longmeyer, I. Perfecto. Spatiotemporal foraging dynamics of Solenopsis invicta and its potential effects on the spatial structure of interspecific competition. Submitted to Environmental Entomology
• Type: Journal Articles Status: Under Review Year Published: 2022 Citation: Mayorga, I., J. L. Vargas, Z. Hajian-Forooshani, J. Lugo Perez, J. Vandermeer, I. Perfecto. Tradeoffs and synergies among ecosystem services, biodiversity conservation, and food security in coffee agroforestry. Submitted to Frontiers in Forests and Global Change
• Type: Journal Articles Status: Under Review Year Published: 2022 Citation: 8. Hajian-Forooshani*, Z., I. Perfecto, J. Vandermeer. Novel community assembly and the control of fungal pathogen in coffee agroecosystems. Submitted to Biological Control

Progress 06/01/20 to 05/31/21

Outputs
Target Audience:1. The academic community - through academic presentations and journal publications 2. Farmers - no contacts with farmers were made this years due to COVID-19 travel restrictions Changes/Problems:Travel regulations due to COVID-19 prevented us from traveling to Puerto Rico in 2020 and part of 2021. We were planning on doing a final assessment of coffee farms with regard to vine cover and general health of the coffee crop that year, but this was not possible. We were also planning on meeting with at least 25 farmers to discuss results from the project, Again, this was not possible. What opportunities for training and professional development has the project provided?This year, we provided data collected from the research project for a student to conduct data analysis for her thesis project since it was not possible for her to go to the field. This student (Ember Bradbury) has been analyzing the acoustic monitoring data for the coqui frogs in coffee farms after the hurricane and comparing it to data we had prior to the hurricane. She will prepare a manuscript that will examine the resistance and resilience of coqui frogs to hurricane disturbance in coffee farms in Puerto Rico central mountain range. How have the results been disseminated to communities of interest?In addition to the academic publications, the PI (I.Perfecto) gave two presentations, one at the University of Puerto Rico-Utuado, and one at the Organization for Tropical Studies (Biodiversity course), where she presented results from this study. We are also working on preparing a flyer/poster from main results to distribute to the farmers. What do you plan to do during the next reporting period to accomplish the goals?We are wrapping up this project. The final phase consists of analyzing the acoustic monitoring data for the coqui frogs and comparing coqui frogs in coffee farms before and after the hurricane to see how the various species were affected by the hurricane.

Impacts
What was accomplished under these goals? Evaluate the extent of damage done by the hurricane to coffee farms- This was achieved the first year of the grant and one publication resulted from the research. Determine the resistance of coffee farms and how it relates to the management style; This was evaluated with respect to the farm vegetation structure and through the response of birds and coqui frogs. The vegetation resistance was included in a manuscript that is now in revision in Frontiers of Sustainable Food Systems (Mayorga et al, in review). The resistance of birds was published in a paper in Global Ecology and Conservation. The data for the resistance of coqui frogs is being analyzed now and will be submitted for publication in the fall 2021. Determine the resilience of coffee farms and how it relates to the management style. This was evaluated with respect to the farm vegetation structure and through the response of birds and coqui frogs. The vegetation resilience was included in a manuscript that is now in revision in Frontiers of Sustainable Food Systems (Mayorga et al., in review). The resilience of birds was published in a paper in Global Ecology and Conservation (Irizarry et al., 2021). The data for the resilience of coqui frogs is being analyzed now and will be submitted for publication in the fall 2021.

Publications

• Type: Journal Articles Status: Published Year Published: 2021 Citation: 1. Irizarry*, A., J. A. Collazo, J. Vandermeer, and I. Perfecto. In press. Coffee plantations, hurricanes and avian resiliency: insights from local occupancy, colonization and extinction rates in Puerto Rico. Global Ecology and Conservation. https://doi.org/10.1016/j.gecco.2021.e01579

Progress 06/01/19 to 05/31/20

Outputs
Target Audience:1. The academic community - through a couple of academic presentations, and 2-week intensive courses o ecological complexity offered at the Universidad de Puerto Rico-Utuado 2. Farmers - through the organization of a workshop for farmers to play ecological games we were able to reach approximately 40 farmers and aspiring farmers in Puerto Rico. We also participated in several radio programs that describe some of the activities of the project focusing on the ecological games. We used this opportunity to talk about the preliminary results of this project. Changes/Problems:The travel restrictions during COVID 19 did not allowed us to travel to Puerto Rico to finish the study in the time espected. We lost an entire year of data collection. We are hopping to recover that time this summer when we will travel to Puerto Rico to collect fecal samples of coquis an conduct a final survey of theiur populations in coffee farms. What opportunities for training and professional development has the project provided?In spite iof the pandemic, during the last year we have trained two graduate students from the University of Michigan (Isabell Mayorga, Ember Bradberry). These students worked with data from the project that was collected the previous year. How have the results been disseminated to communities of interest?Koralis Reyes, the project Coordinator gave an open public talk at the University of Puerto Rico, Utuado. We are also in the process of preparing a website with information about the project results. What do you plan to do during the next reporting period to accomplish the goals?We still need to finish data collection related to the study of the iimpact of hurricane Maria on Eleutherodactylus spp (coqui frogs) and the impact of the frogs on insect pests of coffee. This is the final study of the project and was delayed due to COVID 19. We also plan on finishing a website that will include information obtained from this study and will inform the general public, including farmers, about the resistance and resilience properties of shaded diverse coffee agroforestry systems.

Impacts
What was accomplished under these goals? We already published one peer-reviewed publication documenting theeffect of the hurricane and the resistance and resilience of coffee farms (Perfecto et al., 2019). One article that include resistance and resilience of coffee farms (to hurricane Maria) and how it relates to other ecosystem services is currently in review (Mayorga et al, in review in Frontiers in Forests and Global Change). In addition, we just got a paper accepted in Global Ecology and Conservation (Irizarry et al., In press) that documents the impact of hurricane Maria on birds inhabitting coffee farms in the Central Mountain region of Puerto Rico. We are in the last phase of the project, which was interrupted by the COVID-19 pandemic. Although we already collected data on the effect of the hurricane on Eleutherodactylus spp (coqui frogs) on coffee farms in Puerto Rico. We are still collecting data on the impacts of these frogs on pest of coffee. We espect the data and laboratory analyses for this project to be finished in 6-8 months and we hope to have a final paper by May 2022. Koralis Reyes, the project coordinator also gave a presentation of some of the results at the University of Puerto Rico, Utuado campus.

Publications

• Type: Journal Articles Status: Accepted Year Published: 2021 Citation: Irizarry, A., J. A. Collazo, J. Vandermeer, and I. Perfecto. In press. Coffee plantations, hurricanes and avian resiliency: insights from local occupancy, colonization and extinction rates in Puerto Rico. Global Ecology and Conservation
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Vandermeer, J. and I. Perfecto. 2020. Endogenous spatial pattern formation from two intersecting ecological mechanisms: the dynamic coexistence of two invasive ant species in Puerto Rico. Proceedings of the Royal Society B. 287, no. 1936 (2020): 20202214
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Perfecto, I. and J. Vandermeer. 2020. The assembly and importance of a novel ecosystem: The ant community of coffee farms in Puerto Rico. Ecology and Evolution. (https://doi.org/10.1002/ece3.6785)
• Type: Journal Articles Status: Published Year Published: 2020 Citation: Hajian-Forooshani*, Z., J. Vandermeer, I. Perfecto. 2020. Insight from excrement: invasive gastropod shift diet to consume the coffee leaf rust and its mycoparasite. Ecology, Early View (https://doi.org/10.1002/ecy.2966)
• Type: Journal Articles Status: Under Review Year Published: 2021 Citation: Mayorga, I., J. L. Vargas, Z. Hajian-Forooshani, J. Lugo Perez, I. Perfecto. Tradeoffs and synergies among ecosystem services, biodiversity conservation, and food security in coffee agroforestry. Submitted to Frontiers in Forests and Global Change

Progress 06/01/18 to 05/31/19

Outputs
Target Audience:1. The academic community - through a couple of academic presentations, and 2-week intensive courses o ecological complexity offered at the Universidad de Puerto Rico-Utuado 2. Farmers - through the organization of a workshop for farmers to play ecological games we were able to reach 55 farmers and aspiring farmers in Puerto Rico. We also participated in several radio programs that described some of the activities of the project focusing on the ecological games. We used this opportunity to talk about the preliminary results of this project. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The project has provided significant opportunities for professional development and training of students. Students- 4 PhD students (Alexa White [UM], Chatura Vaidya [UM], Nicholas Medina [UM] and Zack Hajian Forroshani [UM]), one master student (Princejonathan Pruitt [UM]), and two undergraduate students (Priscilla Cintron-Bartolomei [UPR] and Ema Johnson [UPenn] have been trained in ecological research (these students have worked part time in the this project and half time in my other USDA-NIFA founded grant on multifunctionalily in coffee farms). Professional development - Four technical personnel have received training on acustic monitoring (Kris Harmond) and bird surveys (Amarilys Irizarry, Warren Irizarry and Isamarie Acosta Morales) . How have the results been disseminated to communities of interest?We have presented the work in various academic seminars and also part of the results were presented at the USDA-NIFA annual PI meeting in Wahsington DC in December 2018. During a workshop organized for farmers (relared to our other NIFA grant) we also discussed the impacts of Hurricane Maria on coffee farms in Puerto Rico based on the results of this project. What do you plan to do during the next reporting period to accomplish the goals?We still need to do the landscape level study with images bafore and after the hurricane. The analyses that we have performed up to now was based on only 200 meter radius from the centers of the plots in all 85 farms (before and after the hurricane). The land use analyses was based on visual recognition of land use types from google earth images. Now we have to aquire other images and do a 3 km radius to assess landscape level factors. We are expecting to do this in the fall 2019. Once we have that landscape level data we can then proceeed to investigate the relationship between pests (coffee leaf miner, coffee berry borer, coffee rust, leaf hoppers, scale insects) and natural enemies data (ants, birds, Eleutherodactylus frogs, anoles) with landscape level factors before and after the hurricane.

Impacts
What was accomplished under these goals? 1) Damage to coffee farms: We evaluated the extent of the damage using a subsample of 28 farms for which we had data for before and after the hurricane and related it to resistance and recilience of the farms. We found thatthe reduction in shade/canopy cover was 37.5%. However, there is no suggestion that the proportion of shade reduction is somehow related to the level of shade observed on the farm prior to the hurricane. However, we found a clear negative relationship between fertilizer application and hurricane damage.While our observations (both quantitative and qualitative) reflect the very large variability in damage and seeming unpredictability of response variables, one pattern seems to emerge clearly, based on the resilience of farms depending on vine management. Judging resilience based on vine coverage suggests that the combination of management style and direct farmer response combined to set the stage for what we encountered. Farmers able to respond to the hurricane quickly, did so, and the current state of their farms seems well positioned for rapid recovery, while those farms whose owners were unable to respond quickly were put into a distinct ecological regime - excessive vine coverage, and possible loss of production over a long period of time or even abandonment of the farm. These processes makes the resilience of coffee farms to hurricane disturbance in Puerto Rico partly a function of shade (or the percent tree cover in the farms), but also, and perhaps more importantly, a function of the socioeconomic position of the farmer. In other words, the resilience of the system is a property of socioecological factors. These results are in a manuscript that is in review in Scientific Reports. 2) Impact of hurricane on birds and coqui frogs: Bird surveys in coffee farms after the hurrican detected 39 species, as compare to 45 species before the hurricane (using same point counts on same farm locations). Sharp declines were recorded for frugivore forest species (mainly the Puerto Rican Bullfinch, Antillean Euphonia, and the Scaly-naped Pigeon). Through acustic monitoring of the Eleutherodactylus frogs (coqui), we found little change in the number of species detected per farm. Three out of 24 farms had higher species richness and one had lower richness after the hurricane.

Publications

• Type: Journal Articles Status: Under Review Year Published: 2019 Citation: Perfecto, I., Hajian-Forooshani,Z., Iverson, A., Irizarry, A.D., Lugo-Perez, J.,Medina, N., Vaidya, C., White, A. & Vandermeer, J.Response of coffee farms to hurricane Maria:Resistance and resilience to an extreme climatic event. In review in Scientific Reports