Welcome to the MAPS Data Exploration Tool!

This website provides access to data and analytical results from the Monitoring Avian Productivity and Survivorship (MAPS) program, a continent-wide, collaborative effort among public agencies, non-governmental groups, and individuals to assist the conservation of birds and their habitats through standardized mark-recapture monitoring. Created and administered by The Institute for Bird Populations, MAPS is North America’s largest and longest-running program dedicated to the study of avian vital rates (productivity, survival, and recruitment). Since 1989, more than 1,300 MAPS stations in nearly every state and Canadian province have contributed important data that can improve understanding of the population dynamics and conservation needs of terrestrial bird species; approximately 300 stations are currently active.

The MAPS program uses a standardized protocol of constant-effort, passive mist netting during the North American breeding season. Though IBP administers the network, the vast majority of MAPS stations are operated by independent researchers, land managers, or dedicated bird banders who collaborate via the MAPS network.

Analyses of MAPS data can be used to identify spatial and temporal patterns in vital rates and population dynamics and to provide inferences about which vital rates are most strongly linked to population changes. MAPS data can also be used to help understand how factors like weather, climate, land use, and disease affect bird populations. Finally, data collected at MAPS stations can also be used to provide insight into other aspects of avian ecology such as molt strategies, migration, and other seasonal movements.

This website allows users to explore and download MAPS capture/recapture records, as well as view and download results of demographic analyses conducted on the program-wide 1992-2018 MAPS data set. Results from the program-wide analyses include estimates of adult and juvenile population size indices, a productivity index, and adult apparent survival and residency probabilities at the scale of Bird Conservation Regions (BCRs). For more information about BCRs, including an interactive map, visit the US Geological Survey’s BCR page.

Click play on the videos below to see tutorials on how to navigate this application.

MAPS Data Download

The MAPS database has more than 1,850,000 capture records for 1992-2018, including records from 432 bird species in nearly every U.S. state and Canadian province. By providing your name and a short description of your intended use of the data, you will be able to download MAPS capture records and the meta-data associated with them including location, effort, and species breeding status.

The following data types can be downloaded:

• Station location data - names, spatial coordinates, and operational history for the chosen stations
• Capture records - capture records of birds caught at MAPS stations during the MAPS season
• Effort data - net-specific opening and closing times for calculating capture effort
• Net-hours - the calculated net hours by station, year, intended period, and subperiod
• Breeding status data - species-specific observation data for each MAPS period and a yearly determination of whether the species was breeding within the MAPS station boundaries
• Breeding status data (compiled) - yearly breeding status determinations over the time period chosen by the operator and calculated final breeding status determination for that time period
• Analysis suitability - suitability of each station’s year-specific data for use in productivity and survivorship analyses based on the year’s effort level as defined in the MAPS Manual

Ready to get started? Use the tabs at left to agree to our data-use policy and choose the files you wish to download.

Exit and reenter the application if you wish to look at a different combination of stations or years.

MAPS Data Use Policy Agreement

The information entered into this form will allow IBP and MAPS operators to know how their data are useful to other researchers.

Please read the IBP Data Use policy. You will need to agree to the policy before the download will proceed.

Data Download Instructions

Follow the prompts below to choose the data parameters you are interested in. The prompts will only appear after you agree to the data policy on the previous tab and submit your download information.

The final zip file will contain .csv files of the data you chose. Keys to the database structure and data codes for each of the data file types will be included in each data download and are also available here.

To download another data set, exit the application and restart.

Explore MAPS Results

A primary goal of the MAPS program is to provide indices or estimates of landbird demographic rates across space and time. Here, we update results provided by DeSante et al. (2009) by incorporating additional years of data (here 27 years of data [1992-2018] compared to 15 years in DeSante et al. 2009 and applying more recently developed Bayesian hierarchical models that provide region (Bird Conservation Regions) × year specific estimates of demographic parameters, as well as region-specific estimates of trends and annual variation in these parameters. We present results for 5 demographic parameters: 1) adult abundance, 2) juvenile abundance, and 3) productivity indices from generalized linear mixed models of capture data; and 4) adult apparent survival and 5) residency probabilities from Cormack-Jolly-Seber (CJS) models of capture-recapture data. We include parameter estimates for 50 of the most commonly captured species for all regions and years for which data were available and for which parameters were estimable. In future iterations of the website, we intend to incorporate more species and years of data, and additional demographic metrics (e.g., recruitment).

A tremendous level of commitment is needed to establish and operate a MAPS station year-in and year-out. As is to be expected from what is largely volunteer effort with this level of commitment, coverage across species, regions, and years is uneven, presenting challenges for large-scale analyses and limiting the number and resolution of inferences that are possible. Furthermore, guidelines for MAPS station site selection are primarily aimed at generating representative samples for a particular local habitat type or management unit. Thus, it is uncertain whether aggregating these samples at larger regional and range-wide scales can truly provide estimates that are representative of the underlying populations of interest. Finally, the estimates provided herein are also dependent on model structure and how well models fit the data. While we acknowledge that alternative models may be better suited to individual species, and that not all parameters were estimable in cases where there were relatively few captures or recaptures for a given region and year, we nonetheless assume identical model structures across species to allow for as much spatiotemporal resolution as possible and to maintain consistency and facilitate comparisons. Despite these various caveats, the overall scale of the MAPS effort is unprecedented in North America, and we hope that comparisons of demographic estimates across space and time will lend new insights into recent demographic dynamics and trends for many of these species and will be used to inform new research and conservation efforts.

Please See the “About this Site” tab for information on how to cite these results.

Explore MAPS Results

The MAPS Data Exploration Tool allows you to explore summarized data on the vital rates (productivity, survivorship, and recruitment) for scores of species across North America. Below you can choose the species and results you wish to display.

The result popups for adult survival probability, residency probability, adult captures per station per year, young captures per station per year, and productivity (young/adult) include a plot of time specific estimates. Click on a Bird Conservation Region (BCR) to see a popup of the results. Note: If popupgraphs don’t appear for one of the listed parameters, please refresh the program. The program only allows for the parameters to be graphed once per session.

Click here to see a map of all the Bird Conservation Regions (BCRs). For more information about BCRs, including an interactive map, visit the US Geological Survey’s BCR page.

Explore MAPS Results

You can download the calculated results by species and Bird Conservation Regions (BCRs). Choose the species and BCRs for which you would like to download the data. Both the time constant values and the yearly values (when available) will download, but in two separate CSV files. Click the download button to create and download the CSV files.

For more information about BCRs, including an interactive map, visit the US Geological Survey’s BCR page

MAPS Website Regional Demographic Summaries

For each species, we provide Bird Conservation Region (BCR)-scale estimates for five demographic parameters: 1) adult abundance index, 2) juvenile abundance index, 3) productivity (no. juveniles/adult), 4) adult annual apparent survival probability, and 5) residency probability for newly marked birds. For each of these parameters, we proved mean regional estimates, regional trend estimates, region x year estimates, and estimates of annual variance for each region.

Methods

MAPS data

The design of the MAPS program and field methods were standardized in 1992 and are described in Desante et al. (1995, 2004, and 2015). Sampling sites, or MAPS stations, were established by MAPS program cooperators at sites where long-term mist-netting during the breeding season was practical and permissible. Each individual MAPS station typically consisted of about ten 12m x 2.5m, 30mm-mesh mist nets erected at fixed sites that are distributed across the central 8 ha of an approximately 20-ha sampling area. Nets at each station were operated consistently among years (usually 6 hrs/d on 6-10 occasions at approximately 10-d intervals), except for occasionally closing nets due to inclement weather, high capture rates, or other logistical reasons. With few exceptions, all birds captured were marked with uniquely-numbered aluminum leg bands provided by the U.S. Geological Survey or the Canadian Wildlife Service, and their ages and sexes were determined based on Pyle (1997). Band numbers of all recaptured birds were recorded.

For each species, we stratified data by Bird Conservation Region (BCRs). We included all BCRs with at least 100 banded adult individuals. For models of adult apparent survival probability and residency probability (see below), we included data only for stations where the species was confirmed as a breeding species and that had a minimum of 4 years of data. For adult capture and productivity models, we included only data from the first day of station operation per 10-d period in cases where stations operated more frequently than the standard once per 10-day protocol (DeSante et al. 2004).

Statistical analyses

We used Bayesian hierarchical models to model avian response variables. To model the adult (‘after hatching year’) abundance index, we assumed adult captures AHY_j,k,t to be distributed as AHY_j,k,t ~ Pois(μ_AHY[j,k,t]), where the μ_AHY[j,k,t] represents the mean count at station j, BCR k, and year t. To model productivity, we assumed juvenile (‘hatching year’) captures, HY_j,k,t, to be distributed as HY_j,k,t ~ Bin(pHY_j,k,t ,N_j,k,t ), where the pHY_j,k,t represents the probability of a captured bird being a juvenile bird and the N_j,k,t is the total number of captures of adult and juvenile age classes. We then defined generalized linear mixed models for each response of the form:

where the g(θ_j,k,t ) are the link-scale response parameters with log-link for the adult abundance mean parameter, μ_AHY[j,k,t] and logit-link for the productivity parameter, pHY_j,k,t . The β_0[k] represent fixed BCR-specific intercepts; β₁ represents the regression coefficient for an effort ef_j,k,t effect; the β_2[k] represents a continuous random BCR-specific year (trend) effect distributed as Norm(μ_β1, σ²_β1);, where t* = 13 to center the covariate; sta_j is a zero-mean random station effect distributed as Norm(0,σ²_sta); and yr_k,t is a random BCR x year effect with zero mean and region-specific variances, Norm(0,σ²_BCR:year[k]). For the adult and juvenile capture models, we calculated the ξ_j,k,t as a weighted mean number of net-hours across MAPS sampling periods with weights being the average age-specific capture rates specific to each sampling period and BCR. For the productivity model, the ξ_j,k,t covariate was the ratio of juvenile effort, calculated identically to the adult effort covariate in eq. 1 but for hatching-year birds, to adult effort. Thus, this covariate controlled for the proportion of effort in capture periods when juveniles were captured relative to effort in periods when adults were predominantly captured. We derived region- and year-specific adult abundance, N _k,t, and productivity, RI _k,t, indices from these models as exp(β_0[k] + β_2[k](t-t*) + yr_k,t).

We applied a state-space version of the Cormack-Jolly-Seber (CJS) model that accounts for transients (Saracco et al. 2012) to i = 1, …, M individual capture histories to estimate adult apparent survival, Φ_i,j,k,t, and residency, π_i,j,k,t, probabilities. For both response variables we defined logit-linear mixed models including effects from eq. 1 that relate to spatial and temporal variation in the state processes:

Year-specific parameter estimates were then calculated by inverse-logit transformation of these models. In addition to models describing the state processes, the CJS model included sub-models for parameters describing the observation process, capture probability, p_i,j,t, and probability of predetermining a newly banded bird as a resident bird, ρ_i,j,t (i.e., recapturing a bird ≥ 9 d apart in the season it was banded; Saracco et al. 2012). For these parameters, we defined logit-linear models including intercepts and zero-mean random station effects.

We implemented models with JAGS (Plummer 2003) in R (R Core Team 2021) via the jagsUI package (Kellner 2021). We assigned Norm(0,10000) priors to regression coefficients of fixed effects and mean hyperparameters for the random BCR-trend effects and U(0,1) priors to inverse-logit transformed intercepts.

Posterior inferences were based on running 4 Markov chain Monte Carlo (MCMC) simulations (Gilks et al. 1996) for a minimum of 30000 iterations and thinning by 4-10 samples, following an adaptive phase of at least 20000 iterations, and a burn-in phase of at least 10000 iterations. We assessed convergence based on Gelman-Rubin statistic values < 1.2. For some species x region x year combinations, convergence was not achieved, even with longer MCMC runs. In these cases we do not report values. For all other parameter combinations, we report posterior distribution medians with 90% credible intervals (CIs).

Literature Cited

Desante, D. F., K. M. Burton, J. F. Saracco, and B. L. Walker (1995). Productivity indices and survival rate estimates from MAPS, a continent-wide programme of constant-effort mist-netting in North America. Journal of Applied Statistics 22:935–948.

DeSante, D. F., J. F. Saracco, D. R. O’Grady, K. M. Burton, and B. L. Walker (2004). Methodological considerations of the Monitoring Avian Productivity and Survivorship (MAPS) program. Studies in Avian Biology 29:28–45.

Gilks, W. R., S. Richardson, and D. J. Spiegelhalter (1996). Introducing Markov chain Monte Carlo. In Markov chain Monte Carlo methods in practice (W. R. Gilks, S. Richardson and D. J. Spiegelhalter, Editors). Chapman and Hall, New York, New York, USA, pp. 1–20.

Kellner, K. (2021). jagsUI: A Wrapper Around “rjags” to Streamline “JAGS” Analyses. R package version 1.5.2. https://CRAN.R-project.org/package=jagsUI. [Online.] Available at http://cran.r-project.org/package=jagsUI.

Plummer, M. (2003). JAGS: A Program for Analysis of Bayesian Graphical Models Using Gibbs Sampling. Proceedings of the 3rd International Workshop on Distributed Statistical Computing (DSC 2003), March 20-22, Vienna, Austria. ISSN 1609-395X.

Pyle, P. (1997). Identification Guide to North American Birds, Part 1. Slate Creek Press, Bolinas, California.

R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/. R Foundation for Statistical Computing. [Online.] Available at https://www.r-project.org/.

Saracco, J. F., J. A. Royle, D. F. DeSante, and B. Gardner (2012). Spatial modeling of survival and residency and application to the Monitoring Avian Productivity and Survivorship program. Journal of Ornithology 152:S469–S476.

The Institute for Bird Populations developed this website to provide easy access to Monitoring Avian Productivity and Survivorship (MAPS) data records and processed results. We hope that this enhanced capacity to disseminate information from MAPS will help advance bird conservation and research across North America.

Recommended Citations

MAPS capture records and ancillary data

The Institute for Bird Populations. 2023. MAPS Data Exploration Tool: Downloaded MAPS data. Version 1.0. https://ibp-maps-data-exploration-tool.org (Accessed [DATE of USE])

Processed MAPS results

The Institute for Bird Populations. 2023. MAPS Data Exploration Tool: Processed MAPS results. Version 1.0. https://ibp-maps-data-exploration-tool.org (Accessed [DATE of USE])

Thank you!

We thank everyone who has contributed to the MAPS database - both those who have put in countless hours in the field collecting data, and also those who have facilitated and funded the data collection efforts. We also thank the Knobloch Family Foundation for largely funding the creation of this website, and Tracy Aviary for additional financial assistance at a crucial time.

Tutorial videos

Please see the Overview page for tutorial videos on navigating this application.

Contact

For functionality issues contact: Bob Wilkerson (bwilkerson@birdpop.org)

For suggestions for additions/changes to the application: Danielle Kaschube (dkaschube@birdpop.org)

For analytic questions contact: Jim Saracco (jsaracco@birdpop.org)

Updates

Version 1.0 - 2023_03_30

• initial release
• Data download includes MAPS data 1992-2018. Includes MAPS data available as of 2022_08_22.